[From the U.S. Government Printing Office, www.gpo.gov]









                                                                          EPA/ERLN Report # 1233
                                                                          June, 1991



                  Assessment of Responses to Climate Variation
                  in the Marine Environment of Coastal Regions
                                         of the United States










                                              PROJECT OFFICER



                                                Henry A. Walker

                              United States Environmental Protection Agency
                            Environmental Research Laboratory, Narragansett
                                                27 Tarzwell Drive
                                          Narragansett, Rhode Island
                                                       02881




                                 GLOBAL CHANGE RESEARCH PROGRAM









                              United States Environmental Protection Agency
                                                  OEPER,GCRP
                                            Washington, D.C. 20460


                                       US Department of Commerce
                                       Environmental Services Center Library
               QC                     
               981.5                2236 South Hobson Avenue
               .C5                     Charleston, SC 29405-2413
               G637
               1991
 















                                             Disclaimer

               This report has been reviewed by the Environmental Research Laboratory, U.S.
           Environmental Protection Agency, Narragansett, Rhode Island, and approved for
           publication. Approval does not signify that the contents necessarily reflect the views
           and policies of the U.S. Envirorunental Protection Agency, nor does mention of trade-
           names or commercial products constitute endorsement or recoxnmendation for use.







             Assessment of Responses to Climate Variation
             in the Marine Environment of Coastal Regions
                                 of the United States









                                           AUTHOP.S




                                      Fredric A. Godshall'
                                  Computer Sciences Corporation

                                        Henry A. Walker'
                               U.S. Environmental Protection Agency

                                        George IL Mapp'
                                  Computer Sciences Corporation







                          'United States Environmental Protection Agency
                         Environmental Research Laboratory, Narragansett
                                        27 Tarzwell Drive
                                    Narragansett, Rhode Island
                                              02881

                      'United States Environmental Protection Agency
                   Atmospheric Research and Exposure Assessment Laboratory
                                      Research Triangle Park
                                       North Carolina 27711





                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                                                                               I
                                               I
                     US Department of Commerce                                                I
                     NOAA Coastal Services Center Library
                     2234 South Hobson Avenue
                     Charleston, SC 29405-2413                                                 
 







                              Assessment of Responses to Climate Variation
                              in the Marine Environment of Coastal Regions
                                               of the United States



             EXECUTIVE SUMMARY:

                 The scientific objectives of the U.S. Global Change Research Program (GCRP),
             coordinated by the Committee on Earth Sciences are to monitor, understand, and
             ultimately predict global change. A critical challenge in meeting these objectives will
             be to develop understanding of interactions among terrestrial, riverine, oceanic, and
             atmospheric systems that occur because of climate change. Research reported in this
             assessment of marine environmental change describes some effects of past climate
             variations on physical characteristics in United States coastal regions, and infers
             some of the changes that may occur due to global warming.

                 The motivation for an estuarine and coastal component of the EPA GCRP is that
             both sensitive ecological systems and highly populated areas are concentrated along
             coasts throughout the world, and many could be adversely effected if anticipated
             global warming occurs. Changes in physical and chemical parameters such as
             temperature, salinity, light, nutrient fluxes, sea level, and circulation have
             substantially effected coastal zone ecology. Physical conditions of coastal waters may
             provide some of the best opportunities for early detection of response to climate
             change. River runoff is an integrator of continental hydrologic processes and changes
             of runoff into coastal areas link coastal oceanic environment to changes in
             precipitation and land use. Coastal sea level change is caused by change in runoff,
             water temperature, and wind-forced circulation as well as land subsidence.
             Continued research will improve our understanding of coastal marine ecosystem
             responses. This research will permit interpretation of future environmental changes
             with respect to climate change.
             are: Key research questions for the estuarine and coastal component of the EPA GCRP

               (1)  How has past and present climate variability influenced estuarine and coastal
                    ecosystems?

               (2)  What physical, biological, and geochemical coastal zone processes and human
                    activities interact with and may be effected by climate change?

               (3)  How accurately can predictions of future coastal impacts of global climate
                    change be made?









                 These questions were discussed at workshop on "Coastal Ocean Physics and
             Climate Change: Approaches for the Assessment of Ecosystem Response" jointly
             sponsored by EPA, NOAA and the Texas Institute of Oceanography in January, 1990.
             A workshop proceedings document is in press. This workshop proceedings document
             briefly reviews:

               0    Important physical features of U.S. continental shelves and estuaries,

               0    Coastal physical processes effected by climate change,

               0    Coastal ocean modeling methodologies,

               o    Modeling scenarios of potential circulation and water mass changes,

               0    Coastal ecosystems as analogues for climate change, and

               0    Research task areas for modeling, observations and monitoring.


                 In this document, "Assessment of Responses to Climate Variation in the Marine
             Environment of Coastal Regions of the United States," assessments focus on regional
             spatial scale responses to potential changes in large scale atmosRheric and oceanic
             circulation. Three coastal regions are selected from United States East, South and
             West coasts, to serve as examples of coastal responses to climate variation in differing
             coastal regimes. The Mid-Atlantic Bight, the Northwest Gulf of Mexico, and
             Southern California were selected for analysis because of exposure to oceanic and
             atmospheric circulation. This research seeks explanation of environmental effect of
             past climate variability in the selected regions and then, by analogy, relates predicted
             climate changes to changes in sea level, wind, and river runoff.

                The order analysis follows a consideration of factors operating at a range of spacial
             scales, from global to local (Figure E-1, Page v). Section 1 of this report provides a
             brief review of important processes operating in different U.S. coastal regions that
             may be effected by climate change. Section 2 addresses a model based prediction of
             future climatic conditions based on doubled global atmosphericC02 concentrations.
             Section 3 describes regional changes in key parameters affecting specific coastal
             regions for periods of Northern Hemispheric warming and cooling. Section 4
             compares model based predictions with regional scale changes based on historical
             data. Thus, this report considers the wide range of process scales depicted in Figure
             E-1.






                                                        iv











                                              ISCALE                                        I PROCESSES




                                                                                     RADIA TIVEL Y ACTIVE GASES
                                                        G                                  (CO2 EMISSION RATE)
                                                        L
                                                        0
                                                        B
                                                        A
                                                        L                              GENERAL At MOSPHERIC
                                                                                          CIRCULA TION MODEL


                                                        H
                                                        E                               HADLEY ATMOSPHERIC
                                                        M                                       CIRCULATION
                                                        I

                                                        P
                                                        H
                                                        E                         OCEANIC AND ATMOSPHERIC
                                                        R                                      CIRCULATION



                                                                                                                          ...... ..... .. ...
                                                                                                                          . . .........
                                                                                       .0, "..rA     OCEANOGRAPN.


                                                                                               REGIO


                                                        N



                                                                                                                               ...........


                                                        @ m:












                       Figure E-1. Atmospheric and Oceanographic Process Scales




                                                                                                    v








                                  Large Scale Atmospheric Circulation

                Wind generated stress on oceans from large scale wind regimes drive large-scale
             oceanic currents. Global climate change affects atmospheric circulation and wind-
             driven currents. Most coastal areas have different exposure to oceanic influences and
             wind than other areas. Therefore, environmental responses to effects of global-scale
             climate change will differ over United States coasts. Environmental changes may
             include change of any variable and changes could be opposite in different coastal
             areas depending on local effects.

                A primary component of global atmospheric circulation is the Hadley Circulation.
             This circulation, driven by atmospheric heating at the equatorial surface, includes
             vertically rising air with surface easterly wind in equatorial regions and subsidence
             over the latitudes of the subtropical anticyclones (Figure E-2, Page vii). These Hadley
             Circulation subtropical anticyclones of the Pacific and Atlantic Oceans effect the
             meteorology of southern United States through wind-driven broad-scale ocean
             circulation and through coastal north/south winds which cause marine upwelling
             circulation. Two to three year interannual periods of change in Hadley Circulation
             have been observed and longer periods of change have been hypothesized from past
             research. Shorter periods of change in Hadley Circulation have been identified
             through cloud cover over oceanic areas monitored by satellite. Present research
             assesses past and future changes of the Hadley Circulation surface wind fields during
             periods of transition from "cold climatic regimes" and from "warm climatic regimes."
             The research relates circulation changes to changes in marine environments along
             U.S. coasts. Dramatic climate change effects have occurred at northern latititudes
             (Figure E-3), reflecting changes of the westerlie winds in mid-latitudes.

                       Modeling Large Scale Atmospheric Circulation Response
                                       to Doubled Atmospheric C02

                The NOAA General Fluid Dynamics Laboratory (GFDL) Q-Flux Numerical
             General Climate Model (GCM) predictions in 1988 provided estimates of winds over
             continental United States and coastal areas for an atmosphere that contains the
             present concentration of carbon dioxide and for an atmosphere that contains double
             present carbon dioxide concentration. Winter (January mean) and Summer (August
             mean) wind fields, predicted for an atmosphere with the present concentration of
             carbon dioxide, are similar to wind fields of these months that were computed from
             the Comprehensive Ocean-Atmosphere Data Set (COADS). August wind fields
             predicted for a doubled carbon dioxide atmosphere, were also similar to the present
             wind climatology. January wind fields were different in an atmosphere with double
             carbon dioxide concentration, particularly in the area of the south and eastern coast;
             wind direction is predicted to veer (rotate in a clockwise direction) and turn
             southward in apparent association with a larger and westward extension of the
             Bermuda High (part of the Atlantic subtropical anticyclone). In effect, this change
             produced a winter wind that was more like the normal summer wind regime.


                                                       vi





























                                                                               Polar high
                                                           PoOar eastedies




                                         Polar



                                                                           wested es
                                Horse latitudes


                                Hadley Call



                                Dolclrurns-@,



                                                                                                     Tt
















                   Figure E-2. A Schematic Representation of the Hadley Circulation( Lutgens, 1989)


                                                                              vii








                     Historical Analysis of Large Scale Atmospheric Circulation
                                     During U.S. Climatic Transitions

                Area-weighted mean annual air temperature data from the United States and
             temperature analyses from northern hemisphere summarized temperature data, were
             used to select three periods from the historical COADS record. Two periods, 1889-
             1899 and 1970-1979 represent times when U.S. and northern hemisphere climate was
             warming and the period 1935-1944 represents time when change was from a warmer
             to a cooler climate. Based on COADS 2' area averaged data, wind anomaly was
             computed for Pacific and Atlantic oceans in the geographic area 130'E eastward to
             10'W, 0' to 40'N. The anomalies show wind direction veers, turns southward, and
             speed increases along U.S. coasts when climate was becoming warmer. Wind
             direction backs (rotates in a counter-clockwise direction) with speed decreasing along
             U.S. coasts when climate was becoming cooler. Changes of regional wind speeds
             during the warming and cooling periods, westerly wind speeds, and hemispheric
             change of air temperature are illustrated in Figure E-4. Changes of winds during
             climate warming are similar to changes predicted for the earth atmosphere that has
             doubled carbon dioxide concentration and with a climate that becomes warmer.
             COADS surface pressure data were summarized for the same periods as for wind
             data. Subtropical anticyclones of the coastal Pacific region migrated toward the
             northwest during periods of transition toward warmer climate but are displaced
             southeastward during the period with climate cooling.


                       Potential Consequences in Selected U.S. Coastal Regions

             Precipitation:

                Increased southerly wind components in the southeast may increase moisture
             advection into continental U.S. and cause increased precipitation and land runoff.
             Runoff from the Potomac, Delaware, and Hudson River watersheds into the Mid-
             Atlantic Bight during the 1970's has been large and variable over 11 year periods
             but without large increase relative to earlier decades. Predicted East Coast
             precipitation by the GCM for the warmer climate is also increased. Records of river
             flow of the Arroyo Seco, representative of land runoff in Southern California, show
             the runoff tends to be decreased during periods of transition to warmer conditions.
             Increased river runoff in Southern California was observed during 1935-1944, a
             transition to colder climate. Flow records from the Mississippi show increases in the
             1970's but multidecade trends are not well defined by records available for this study.








                                                       viii





















                              Temperature Differences for Dec. - Feb.
                                   (GFDL    2xCO, - IxCO,)



























              Figure E-3. Mode I-Est.,.-ated !;::rthern Henlsphere winter Air
                         Temperature Change for Doubled Atmospheric Carbon
                                                                            1
                         Dioxide Concentration (Weatherald and Manabe,       986).

                                               ix


















                                  ANNUAL MEAN `4URFACP 'ITNIPFRATURE ANOMALY (C.)
                                         ESTIMATED FROM TEMPERATURE MEANS (1945-60)
                                                  IN THE NORTHERN HEMISPHERE
                 EMP                                                                                   A
                    0-i                                                                                                           vv



                      1890         1900                     1920                     1940                     1960                     1980


                           -41



                                                                                              IT'



                                                                                                                                   197G 1979
                                Ism 1899                                                     111    1 RROW,   1,               WindSpewfAnowWy k''11,
                           WirA Sv-d A-wy                                                            4mowl
                                                                                                   r
                                                                                    35 IN4
                                                                                       An

                 w                                                                WfND     SPEED
                 I
                 N   6                                                    0-40N,  130E-IOW
                 0



                 E
                 E
                 D   5-


                 M

                 S    4

                      1890 1900                1910        1920 1930 1940 1950 1960 1970 1980




                  Figure E-4. Changes in Northern Hemispheric Air Temperature, Regional Wind
                                          Regimes, and Westerly Wind Speeds.
                                                                                                                                       0@, 4@1






                                                                                    x












            Wind-Driven Coastal Circulation:

                During periods with warming climate, increased northerly wind increases wind-
            driven ocean circulation on the Southern California coast. Increased upwelhng on
            the coast resulting from increased northerly wind causes lower coastal water
            temperature higher coastal salinity. Also, this will cause lowering of sea level.

                Water-mass dynamics in Mid-Atlantic Bight promote southward drift that
            carries chemically enriched water southward along the New Jersey coast. Increased
            southerly wind components, which were found during periods of climate warming,
            produce wind stress to oppose near-shore drift. Increased upwelling resulting from
            wind stress along some coasts will bring oxygen rich "cold pool" water shoreward.
            Periods of upwelling that are extended into seasons of colder weather cause coastal
            water to be unseasonably warm and more saline.

               Westward extension of the Bermuda High (part of the Atlantic area subtropical
            anticyclone) may promote the occurrence of late-spring northeasterly wind over the
            Mississippi delta area. Events of severe anoxic bottom water on the Northwestern
            Gulf of Mexico shelf have occurred because late spring northeasterlies caused an
            increase in the areas with highly stratified shelf water masses.


            Sea Level Rise:

                Steric effects are predicted to cause continued rise of sea level in the Mid-
            Atlantic Bight. With a warming climate the land runoff is expected to increase but
            most effects of sea level increase and runoff into this coastal area are expected in
            lacustrine estuaries. Future assessments of coastal environmental response to global
            climate change should include modeling studies of Chesapeake and Delaware Bay
            using predicted sea levels and runoff amounts.

                Sea level increases in the Northwestern Gulf area during recent times are
            primarily associated with land subsidence and these changes are expected to
            continue.









             GLOSSARY:



             anticyclonic: Direction of motion which is clockwise in the northern
                 hemisphere of earth.
             anticyclone: A region on the earth surface in the northern hemisphere where wind
                 motion is clockwise and atmospheric pressure on the earth surface is relatively
                 high.
             area weighted mean: A statistical parameter computed by summing the products
                 of variables by the measure of the region of variable influence and then dividing
                 the sum of products by the total sum of the region measures.
             backing: Counter clockwise change in direction of wind relative to an observer where
                 the wind direction of air movement is considered to be toward the observer.
             California current: A southward flowing current near the west coast of the United
                 States.
             cyclonic: Direction of motion which is counter clockwise in the northern hemisphere
                 of earth.
             cyclone: A region on the earth surface in the northern hemisphere where wind
                 motion is counter clockwise and atmospheric pressure on the earth surface is
                 relatively low.
             Davidson current: A northward flowing current near the west coast of the United
                 States.
             dynamic depth/height: A relative magnitude of a field or surface parameter that
                 is associated with fluid motion.
             Ekman transport: Net displacement of air or water from one location to another,
                 relative to a fixed location on the earth, that is caused by acceleration from earth
                 spin.
             evapotranspiration: Change of water phase with plant extruded water going
                 to a gas.
             general circulation: Circulation in the atmosphere or oceans which has dimensions
                 commonly associated with the size of earth hemispheres or ocean basins.
             geopotential: A potential energy that is produced by the relative position of earth
                 fluid mass. Surfaces with equal geopotential are designated geopotential surfaces.
             geostrophic: A descriptive term that refers to air or water motion when, in the
                 absence of significant friction forces, the forces of pressure are balanced by the
                 Coriolis forces.
             Hadley Circulation: Circulation of air in the troposphere of earth equatorial regions
                 which is driven by atmospheric heating at the earth surface. In this Circulation,
                 a net ascension of air from low latitudes is followed by latitudinal movement of
                 air to higher latitudes, air subsidence at latitudes of the earth subtropical
                 anticyclones, and surface wind which is caused by the resulting differences of air
                 pressure between the anticyclones and equatorial low pressure where the heating
                 occurs.





                                                        xii








            intertropical convergence zone: The earth equatorial zone where wind that is
                associated with the northern hemisphere meets wind that is associated with the
                southern hem; sphere.
            latitudinal: Motion across latitudes.
            lacustrine: A water body characteristic in which the water area is semienclosed
                geographically from a larger water area which is commonly a sea or ocean.
            long wave: Periodic variation in the atmosphere and oceans that is associated with
                influence regions with distances of thousands of kilometers.
            orographic: Influenced by geographic features, usually mountains.
            meridional: Motion along earth longitudes or a measure of extent that is associated
                with longitudes.
            Afilankovich cycles: Periodic changes in orientation of the earth in space relative
                to the sun with periods in the order of thousands of years.
            palustrine: An adUective that refers to association with a marsh environment.
            phenological: A term that references association with plants.
            steric: Refers to conditions associated with pressure and density effects.
            trough: A term that refers to a minimum of a field or surface.
            upwelling: Oceanic circulation of water commonly caused by wind force which brings
                subsurface water toward the surface.
            veering: A clockwise change in direction of wind relative to an observer where
                the wind is considered to move air toward the observer. This direction change is
                opposite to backing of wind.











                                            Acknowledgement


                Resources supporting research, which developed results presented in this report,
             were provided through the United States Environmental Protection Agency Global
             Change Research Program (GCRP).

                 This research developed from ideas presented at the "Coastal Ocean Physics and
             Climate Change" workshop in January, 1990. The workshop participants presented
             many ideas that were eventually used in the research conducted here at the EPA
             Environmental Research Laboratory, Narragansett (ERLN). Don Block and Cary
             Gentry, of the Atmospheric Research and Exposure Assessment Laboratory (AREAL)
             mapped data summaries for many of the figures included within this document.
             Document reviews were provided by George Mapp, Sharon LaDuc and Peter
             Finklestein at AREAL, and by Jan Prager, John Gentile and Wayne Munns at ERLN.




























                                                     xiv









                                                   CONTENTS




                         Figures and Tables        ................................                 xvii

                         PREFACE        ........................................                     xxi


              1          RqTRODUCTION           ..................................                     1

              1.1        Climate Change in the United States            .................              3

              1.2        Atmospheric Circulation         ............................                10

              1.3        Physical Properties of the Marine Environment              .......          11

              1.3.1         Mid-Atlantic: Bight      ...............................                 12

                              o Mid-Atlantic Circulation         ......................              12


                              o Mid-Atlantic Sea Level         ........................              13


                              o Mid-Atlantic River Runoff          .....................             14


              1.3.2         Northwest Gulf of Mexico         .........................               16


                              o Northwest Gulf of Mexico Circulation             ...........         16


                              o Northwest Gulf of Mexico River Runoff             ..........         17


                              o Northwest Gulf of Mexico Sea Level            .............          17


              1.3.3         Southern California       ..............................                 21


                              o Southern California Ocean Circulation             ..........         21

                              o Southern California Atmospheric Circulation              .....       22

                              o Southern California River Runoff           ...............           25

                              o Southern California Sea Level           .................            25






                                                         XV






            2         MODEL PREDICTION OF CLIMATE             ........   o........        29

            2.1          Interpretation of Modeling Results    ...  o.....                31

            3         ANALYSIS FROM COADS           ...........................           41


            3.1          Differences in Ocean-Area Wind
                         Regimes for Selected Periods     .........  o o......o ....      41

            .3.2         Differences in Ocean-Area Atmospheric
                         Pressure for Selected Periods     ........  o...........         51



            4         CONCLUSIONS .... o o o o    ... o o o o .......... o o.... o o...   60


            5         REFERENCES AND BEBLIOGRAPHY               ..............   o..      63


            APPENDIX A .. o      ...............    o ............   o............        72


            APPENDIXB        .....  o.......................      o ..... o.... o...      76


            APPENDICKC         ....................     o ..........                      79



























                                                  xvi










                                           FIGURES and TABLES

             Figures:                                                                            Page

             E-1 Atmospheric and Oceanic Processes Scales          ........................         v

             E-2   A Schematic Representation of the Hadley Circulation       ..............       % M**

             E-3   Model-Estimated Northern Hemisphere Winter Air Temperature
                   Change for Doubled Atmospheric Carbon Dioxide Concentration
                   (Wetherald and Manabe, 1986)       .................................             ix

             E-4   Changes in Northern Hemispheric Air Temperature, Regional Wind
                   Regimes, and Westerly Wind Speed       ..............................            x

             la.   The Decadal Mean of the Number of Days to Grape harvest
                   From September One in Central Europe: An Analog of
                   Atmospheric Surface Temperature Variation.
                   (illustration from Pfister, 1988)  .................................             4

             1b.   Annual Mean Surface Temperature Anomaly (OC) Estimated
                   from Temperature Means of the Period 1946-60 in the
                   Northern Hemisphere (idlustration from Jones, Wigley,
                   and Kelly, 1982)  .............................................                  4

             2.    Estimated Mean Temperature Differences(*C) for Decades
                   of the Seventeenth Century Based on Tree-Ring Analysis
                   (illustration by Fritts, 1980)  ....................................             5

             3a.   Monthly Mean Isobars, Isotherms, and Wind for October
                   1873 (U.S. Signal Office, 1891). -    ...............................            7

             3b.   Synoptic Meteorological Analysis for the Morning,
                   September 10, 1893 (Koppen, 1899)      ..............................            8

             4.    Annual Mean Temperature of the United States
                   (M. La Brecque, 1989)     ........................................               9

             5.    Northern Hemisphere Surface Air Temperature Variation Averaged for
                   Selected Latitude Bands (Hansen and Lebedeff, 1987)        ..............        15





                                                       Xvii








             6.   Runoff Rates from the Potomac River (Point of Rocks guage), Delaware
                  River (Wallenpaupack and Bush Kill Creeks), and Hudson
                  River (Rondout Creek) Watershed. Annual Means of Guaged
                  Flow are Smoothed by a 5-Year Running Mean          ..................       18

             7.   The Mississippi River Watershed (N. Rabalais and
                  D. Boesch, 1990: EPA Coastal Ocean Physics Workshop
                  Jan. 8-9, 1990, Galveston, Texas, illustration by R.E.
                  Turner) . .................................................                  19

             8.   Annual Discharge of the Mississippi River, Measured
                  at Vicksburg, MS 1931-1977 (Dinnel and Wiseman, 1986)       ............     20

             9.   Sea Surface Layer Temperature Anomaly ('C) of the U.S.
                  West Coast (analysis by Godshall and Williams, 1981)    ...............      24

             10.  Monthly Mean Flow Rate of the Arroyo Seco River
                  Smoothed by a 12-Month Running Mean          ........................        26

             11.  Water temperature of Coastal, Southern California
                  (California State Water Quality Control Board, 1965)   . ..............      28

             12.  Longitudinal Mean Temperature Changes Predicted
                  by GCMs Under Doubled Atmospheric Carbon Dioxide
                  Concentration (Grassl, 1988)  . .................................            30

             13a. August Wind Vectors Estimated by Numerical General
                  Climate Model for an Atmosphere with CO, Concentration
                  of the Present (NOAA GFDL Q-Flux Model Runs, 1988)          .............    32

             13b. August Wind Vectors Estimated by Numerical General
                  Climate Model for an Atmosphere with Double the CO,
                  Concentration of the Present (NOAA GFDL Q-Flux Model
                  Runs, 1988)   ...............................................                33

             14.  Comparison of Model Predicted Surface Wind &om an
                  Atmosphere with 1 X C02 and with 2 X C02 Concentrations
                  in August (NOAA GFDL Q-Flux Model Runs, 1988)          ................      34

             15a. January Wind Vectors Estimated by Numerical General
                  Climate Model for an Atmosphere with C02 Concentration
                  of the Present (NOAA GFDL Q-Flux Model Runs, 1988)          .............    37




                                                     xviii








            15b.  January Wind Vectors Estimated by Numerical General
                  Climate Model for an Atmosphere with Double theC02
                  Concentration of the Present (NOAA GFDL Q-Flux Model
                  Runs, 1988)    ..............................................               38

            16.   Comparison of Model Predicted Surface Wind from an
                  Atmosphere with' X C02and with 2 X C02Concentrations
                  in January (NOAA GFDL Q-Flux Model Runs, 1988)          ...............     39

            17.   Estimated Runoff from the Delaware River Watershed With
                  Increased Air Temperature and Precipitation Changes (mrn)
                  in the Watershed (M. Airs, 1990: Watershed Hydrology,
                  Coastal Ocean Physics Workshop, Jan. 8-9, 1990,
                  Galveston, Texas) ...........................................               40

            18.   Model-Predicted Runoff (mm) in the Delaware River
                  Watershed for +2'C Climate Scenario (M. Airs, 1990:
                  Watershed Hydrology, Coastal Ocean Physics
                  Workshop, Jan. 8-9, 1990, Galveston, Texas)  . .....................        40

            19.   Wind Speed Variations from COADS Analysis over the
                  Period 1890-1979     ..........................................             42

            20a.  Wind Speed Anomaly for the Period 1889-1899 Computed
                  from COADS (Shaded Circles Indicate Larger, Open Circles
                  Indicate Smaller) ...........................................               44

            20b.  Wind Direction Anomaly for the Period 1889-1899 Computed
                  from COADS (Shaded Circles Indicate Backing, Open
                  Circles Indicate Veering) . ....................................            45

            21a.  Wind Speed Anomaly for the Period 1970-1979 Computed
                  from COADS (Shaded Circles Indicate Larger, Open Circles
                  Indicate Smaller) ...........................................               46

            21b.  Wind Direction Anomaly for the Period 1970-1979 Computed
                  from COADS (Shaded Circles Indicate Backing, Open
                  Circles Indicate Veering)  ......................................           47

            22a.  Wind Speed Anomaly for the Period 1935-1944 Computed
                  from COADS (Shaded Circles Indicate Larger, Open Circles
                  Indicate Smaller)    ..........................................             48




                                                     xix








             22b.  Wind Direction Anomaly for the Period 1935-1944 Computed
                   from COADS (Shaded Circles Indicate Bacldng, Open
                   Circles Indicate Veering) . ....................................               49

             23.   Monthly Average Sea Level Observed on the Coast South
                   of San Francisco During the Period 1900-1979
                   (Chelton et al., 1982)   .......................................               50

             24a.  January Average Atmospheric Surface Pressure for
                   the Period 1889-1899 Computed from COADS           ....................        54

             24b.  August Average Atmospheric Surface Pressure for
                   the Period 1889-1899 Computed from COADS        .....................          55

             25a.  January Average Atmospheric Surface Pressure for
                   the Period 1935-1944 Computed from COADS        .....................          56

             25b.  August Average Atmospheric Surface Pressure for
                   the Period 1935-1944 Computed from COADS        .....................          57

             26a.  January Average Atmospheric Surface Pressure for
                   the Period 1970-1979 Computed from COADS        .....................          58

             26b.  August Average Atmospheric Surface Pressure for
                   the Period 1970-1979 Computed from COADS        . ...................          59


             A-1.  January Mean Atmospheric Surface Air Pressure        ..................        74

             A-2.  August Mean Atmospheric Surface Air Pressure      . ..................         75


             C-1.  January Vector Mean Surface Winds         ...........................          81

             C-2.  August Vector Mean Surface Winds        ............................           82


             Table:



             1.    Pressure Anomaly Averaged Over Two-Degree Latitude
                   Bands From the Periods, 1889-1899, 1935-1944, and
                   1970-1979 in Pacific (160'-130*W) and
                   Atlantic (50'-20*W) Areas    ....................................              53


                                                       xx













                                                 PREFACE

                Scientific objectives of the U.S. Global Change Research Program (GCRP),
            coordinated by the Committee on Earth Sciences are to monitor, understand, and
            ultimately predict global change. A great challenge in meeting these objectives will
            be to develop understanding of interactions among terrestrial, riverine, oceanic, and
            atmospheric systems that occur because of climate change.

                Motivation for the estuarine and coastal component of the EPA GCRP is that both
            sensitive ecological systems and highly populated areas are concentrated along the
            coast throughout the world, and many could be adversely effected if anticipated sea
            level and climate changes occur. River runoff is an integrator of continental
            hydrologic processes that link coastal oceanic environment to changes in precipitation
            and land use. Coastal sea level change is caused by runoff, water temperature
            change, and wind forced circulation as well as land subsidence. Physical conditions
            of coastal waters are expected to respond quickly to climate change, and may be
            interpreted for early detection of climate change. Changes in key physical and
            chemical parameters such as temperature, salinity, light, nutrient fluxes, sea level,
            and circulation have effected coastal zone ecology in the past. Continued research
            will improve our understanding of coastal marine ecosystem responses and permit
            interpretation of ecological changes with respect to climatic changes.

                Key research questions for the estuarine and coastal component of the EPA GCRP
            are:


              (1) How has past and present climate variability effected estuarine and coastal
                  ecosystems?

              (2) What physical, biological, geochemical coastal zone processes, and human
                  activities interact with and may be effected by climate change.

              (3) How accurately can predictions of future coastal impacts of global climate
                  change be made?








                These questions were discussed at a workshop on "Coastal Ocean Physics and
             Climate Change: Approaches for the Assessment of Ecosystem Response" jointly
             sponsored by EPA, NOAA and the Texas Institute of Oceanography in January, 1990.
             A workshop proceedings document is in press. This workshop proceedings document
             briefly reviews:

              o Important physical features of U.S. continental shelves and estuaries,

              o    Coastal physical processes affected by climate change,

              o    Coastal ocean modeling methodologies,

              o    Modeling scenarios of potential circulation and water mass changes,

              o    Coastal ecosystems as analogues for climate change, and

              o    Research task areas for modeling, observations and monitoring.


                In this document "Assessment of Responses to Climate Variation in the Marine
             Environment of Coastal Regions of the United States," assessments focus on regional
             spatial scale response to potential changes in large scale oceanic and atmospheric
             circulation as a result of global warming. Three coastal regions have been selected
             from United States East, South and West coasts, to serve as examples of differing
             coastal regimes. The Mid-Atlantic Bight, the Northwest Gulf of Mexico, and
             Southern California were selected because of regional exposure to large scale
             oceanic and atmospheric circulation. This research seeks explanation of effect of past
             climate variability in the selected regions, and then, by analogy, relates predicted
             climate changes to changes in sea level, wind, and inflow of river runoff in the
             selected regions.


















                                                      xxii









                            Assesssment of Responses to Climate Variation
                            in the Marine Environment of Coastal Regions
                                            of the United States



            1. EWRODUCTION

                Recent increases in the atmospheric concentration of a number of radiatively
            important trace gases (RITGs) are attributed to activities such as fossil fuel
            combustion, deforestation, cement rnanilfqcture, agricultural practices and production
            of chlorofluorocarbons (CFCs). These increases raise concerns about global warming
            because these gases absorb long wave length radiation.              Atmospheric CO,
            concentrations may double by the middle of the next century. Regional climate
            changes, predicted through general climate models (GCMs) operating with double CO,
            atmosphere, are expected to cause significant changes in agricultural productivity.
            (Izrael and Hashimoto, 1990). These expectations are developed through use of
            productivity analogues based on historical productivity records. Analogues from
            historical records are used in this assessment of the marine environment to develop
            prediction of change in coastal regions.

                During the Pleistocene (approximately the last 1.6 million years), atmospheric
            C02 fluctuations of 30-50% occurred with lows during glacial and highs during
            interglacial periods. Atmospheric CH4, another RITG, was also low during glacials
            and high during interglacial periods. Thus, changes in atmospheric RITG
            concentration occur with climate change during glacial cycles. Ice core analysis from
            Antarctica (Barnola et al., 1987) provides a historical record of interpreted climate
            over a period of 160,000 years from the past withC02 concentration. SinceC02
            changes parallel climate change, the importanceof C02as a cause of climate change
            is not clearly defined (Siegenthaler, 1988).

                Recently, atmosphericC02concentrations have increased by about 0.4% per year.
            Since mid 1987 this rate has increased to about 0.7% per year. The rate of increase
            in atmospheric carbon sources is larger than the rate of change in the terrestrial and
            marine "sinks" for these gases. AtmosphericC02 increase is much greater than
            expected from fossil fuel combustion alone. The increased rate of change may be
            caused by deforestation and earth warming that led to decrease in the rate of oceanic
            uptake of C02. The measured rate of C02 increase and projected doubling of
            atmosphericC02concentrations by the middle of the next century is unprecedented
            in the geological record during the quaternary.








                Atmospheric temperature and radiation are parameters of climate and these
             variables cause change in all of the earth's environments. However, climate cannot
             be measured directly because it is the collective effect of all its parameters.
             This ensemble of effects has been in continuous change over time in a generally
             defined periodicity of tens of thousands of years (Flint, 1957). There are measurable
             changes, apart from seasonal change, over much shorter periods but the significance
             of change compared to a normal or expected climate condition is poorly defined. In
             the United States, meteorological variability may hinder development of a precise
             normal since record periods are commonly short relative to temporal scales of
             variability. However, climate change can be assessed with respect to reaction of
             selected variables in specific environments. This study describes coastal marine
             environmental response to the variables river-runoff, sea-level, and wind effects.
             Each near-shore coastal area and estuary of the United States is uniquely different
             from all others. Each area is expected to respond to climate change in unique ways.
             Initially, estimates of response for a few areas are assumed adequate to develop a
             sense for general trends of response to climate change and analogues of change. For
             initial analyses the regions of Southern California, North Western Gulf of Mexico,
             and Afid-Atlantic Bight are chosen. These are coastal areas influenced by large
             continental watersheds or exposure to geographically unique oceanic influences, and
             also areas where gufficient environmental data are available to identify present
             characteristics of environment.

                Climate variation forcing during the Pleistocene period is identified with
             Nfilankovich orbital cycles of the earth (Kutzbach and Otto-Bliesner, 1982). Glacial-
             interglacial cycles are related to variations in earth orbital parameters (19K and 23K
             year variations in precession, 41K yr cycle in axial tilt, and 100K yr cycle in
             eccentricity). These orbital variations affect the spatial and temporal distribution of
             solar radiation reaching earth. Additional forces of future climate change are
             anticipated from changes of radiation balances within the earth's atmosphere caused
             by RITGs. Past atmospheric radiation balance changes may have amplified
             astronomical causes of climate change but in modem times radiation balance changes
             may be primary causes of change. Therefore, use of geographically-associated climate
             changes from the past demand judicious use as analogues or examples for the future.
             The analogues of world climate change from coastal areas of the United States in this
             paper are from -model results and extrapolation of climate conditions over the past
             few hundred years. Changes of the past few hundred years, caused by Nfilankovich
             cycles, are assumed small.      Temperature changes estimated in transition from
             interglacial to glacial periods were on the order of a thousandth of a degree per year.
             The yearly amount of change was much greater in transition back to interglacial
             periods. During the past few centuries, long period temperature changes are
             observed on the order of tenths of degree per year. In this report, Nfilankovick cycles
             are ignored, since the research focus is on comparison of climate conditions over the
             past few centuries and in prediction of climate change for the next century, rather
             than on time scales of thousands of years.


                                                        2








             1.1 Climate Change in the United States

                Climate change can occur as change in a single variable of the atmosphere or in
             many variables.     These changes are decreases or increases depending upon
             association among variables. In addition, interdependence of variables may change
             with geographic area.      Therefore, changes in radiation balances within the
             atmosphere could cause different change in meteorological and oceanic variables in
             different geographical areas. This could produce widely varying climate conditions
             and climate change in the United States could be different from other geographic
             areas.


                Paleontologic analyses indicate that global air temperature differences between
             glacial and interglacial periods were on the order of 8* C (Flint, 1957). Changes in
             ocean circulation resulting in the closure of the Nfiddle American Seaway (Hamilton,
             1965; Keigwin, 1976) and an intensified late-Cenozoic orogenic phase (Hamilton,
             1965) explain the relatively sudden onset of Northern Hemisphere ice-sheet
             formation. Atmospheric circulation during the pleistocene is similar to the circulation
             of modem times. Global-mean temperature's are similar to the present but fluctuate
             with periodic glacial changes.       This caused repeated large-scale latitudinal
             displacements of climatic zones by as much as 20 to 30 degrees (Kennet, 1982). An
             enhanced low- latitude atmospheric circulation, the Hadley Circulation, during the
             Pleistocene (Nyberg, 1951) is attributed to large meridional temperature gradients
             caused by cool mid-latitudes. The global temperature fluctuations during glacial-
             interglacial cycles are considerably larger than variations of 1' C, deduced from
             measurements and phenological data over the past 600 years (Figures la.,lb., Page


                Evidence for change toward colder temperatures in central Europe (Pfister, 1988)
             during the 18th century (Figure la., Page 4) is augmented by evidence of glacial
               aximums in Alaska and in Central Europe (Flint, 1957) during the same period.
             This suggests some continuity of climate trends in Europe and North America but the
             1/20 C difference between the 15th century and the 20th in central Europe may be
             different from the magnitude of change in North America. A record of temperature
             anomaly (Figure 1b., Page 4) from the northern hemisphere (Jones et al., 1982)
             appears to overlap the grape harvest data. The combined records provide a record
             of varying temperature from the 14th century in the northern hemisphere.

                Phenological or sedimentary analyses of climate are not totally applicable for
             extension of single variable records backward in time because phenological and
             sedimentary conditions often are dependent on multiple variables. For example, a
             temperature variation record inferred from tree ring analyses (Fritts, 1980) provides
             estimates of relative spatial variation of temperature over the United States in the
             17th century (Figure 2.). Tree ring variations are also caused by moisture



                                                       3





















                                        20


                                        22


                                        24     1
                                        28     1
                                        301    1                           V
                                        32


                                        34


                                        36
                                        2    an =0 VAO "M IM IM sw "a IM IM IM "o WN
                                                                                       decade


                    Figure la. The Decadal Mean of the Number of Days to Grape Harvest From
                                        September One in Central Europe: An Analog of Atmospheric Surface
                                        Temperature (illustration from Pfister, 1988).




                                    1.0





                                    0.5.




                                    0.01                                                                    V.




                                   -0.5-



                                        1880       1 9bo         1 t2-0        1440           1460         19'80

                    Figure 1b. Annual Mean Surface Temperature Anomaly (C) Estimated from
                                        Temperature Means (1945-60) in the Northern Hemisphere (illustration
                                        from Jones, Wigley, and Kelly, 1982).
                                                               1@                                 @ N                       Vil

                                        Tr@e

                                                                                    4






                                              25 -W.        1802-7                   o-25
                                         0
                                         0   so
                                         25



                                                                               25

                                                0 0                    -25

                                                                                              0
                                                                                          -zs
                                                    25           rm    @;                            so
                                         ..50                    .2nS     -Z


                                                                                     75
                                         25  0




                                             -2                                    25
                                                0 2S            25 0                        0
                                                                                               25
                                             -  -50 0                               0,25 @50.75-100  r671-1M
                                                       1@ 0                                                 so

                                         .25         1

                                             25 1                         25    0

                                                                                                                -25
                                            26  0 25    +      w

                                         -25 -50-?S - .100                            26   25-80
                                         0                                 25                            .75
                                         0
                                                                                25                               -so




                                                                                                                .25
                                                                        0
                                                35 25                                          -25

                                                25                               2S    so            IBM-1700
                                             250-            r641 - r660
                                                                             So


                                                                                                                   .25






                                                                                            25

                                                               0





                   Figure 2. Estimated Mean Temperature Differences (OC) for Decades of the
                                    Seventeenth Century Based on Tree Ring Analysis (illustration
                                    by Fritts, 1980).
                                                                                             -'0     1W-
                                                    0            -*30
                                                     50
                                         "2
                                                0              W
                                         '00 "t7"                 157 @O
                                             25' "t
                                                               0














                                                                                   5








             availability. Therefore, these inferred temperature distributions cannot be as precise
             as distributions deduced from modem measurements. However, these 17th century
             temperature distributions are supported by reports of the 1607-1608 and 1683-1684
             winters that were particularly severe in New England and the northeastern United
             States Torry, 1843). These winters occur in the periods shown on the figure by
             Fritts to be the periods of large temperature decreases from normal in the
             northeastern area.

                Long and continuous records of environmental variables from the United States
             are of interest because they give the recent trend of change in the variables that
             produce climate. Unfortunately, the period of recorded environmental measurement
             in the United States is short with respect to periods usually associated with climate
             variation. Although the period of recorded measurements is longer for Europe, the
             possibility of different trends in climatic change in different areas clouds the
             interpretation for the United States. A few records of meteorological measurement
             with breaks are available in the United States from the mid-1700's but records with
             more continuity are available from the mid-1800's. Spatial coverage of observations
             to permit meteorological analysis begins in the late 1800's (Bigelow, 1909 and Figures
             3a. and 3b., Pages 7 and 8). From the 19th century to the present, observation
             stations have moved, observation techniques changed, and observation station
             locations have been encroached by landscape changes and urbanization. These
             changes have caused differences in meteorological measurements that are similar to
             temporal changes associated with climate variation.

                A temperature record for the United States (Figure 4., Page 9) is derived from
             average of air temperature measured by observation stations in the contiguous
             United States. The long-period temporal variation of temperature in the northern
             hemisphere is of larger magnitude in latitudes north of 64'N (Hansen and Lebedeff,
             1987 and Figure 5, page 15). Relatively high Northern Hemispheric temperature
             during 1935-45 and temperature decline to about 1970 is similar to temperature
             anomaly changes in Figure 1b., Page 4. Temperature record variation in the United
             States is similar to variations over the Northern Hemisphere for much of the record
             except since about 1970. Change in any parameter of climate can cause change in
             other parameters. Because of regional differences in parameters of climate, short
             period temperature trend differences are not unexpected. Palutikof et al. (1984)
                                                                                     -1 T
             compared temperatures and precipitation amounts during the periods 9 1-20 and
             1934-53. Temperatures during 1934-53 were about 0.5 - 1.0* C higher in the United
             States mid-west. Comparison of meteorology of the two periods is useful as a
             comparison of possible conditions under warmer climate. Precipitation increased by
             more than 0.5 standard deviations in coastal areas during the warmer period and
             decreased 0 - 0.5 standard deviations in central United States. However, trends of
             temperature and precipitation change are not statistically correlated (Hanson et a-L.
             1989).



                                                       6



















                                          OIL.

                                                                   so
                                                                                 _@A'6

                                                                                           T




                                                    j
                                           4L
           F


                                                                                             Pill


                                                                                   01

                                                                                           t1j.

                                                                                    7S
                                                                                              1w.
                                                                  2w





            Figvm Sa.   Monthly Mean Isobars, Isotherms, and Wind for October 1873 (U.S. Signal Office,1

                                                                  7





                                  Thf.2. SynopfischeWeff@rlrarfe aerDeufsaien Seewarte a a.. Ddniselen MeteorJoffig&OA Ltgfiftts.                                           17.
                                                                                                                                            AO
                                           40             so             to              70                                                                   To
                                                                                --T -io-- -\  TO -10-60-',40      40  30 10 10    10 20 30   40                   70
                                                    7f-            too





                                                                                                                                             Ix








                                    'go




                                  110





                                                                                                                                                                 Jig.


                                   0
                                                                                                                                                   zum winahn-Amd zwijailm
                                                  44,                                                                                             mnimt, T, Ir, Ir. T WKTI &W
                                                                                -4-
                                                                                                                                                        Ad Ts A&* AM, bd T,
                                                                                                                                Udko Xilt&z               ut Y@ -W. N.Tund V
                                                                                                                                                           &0 64 T, aurzw sot
                                                                                                                       71-      M-Ula- W&I.
                                                                                                                                7*.rAv-Sff-2hmvat.T., and T, bast 470 af@izwt
                                                                                                                                DwLzdbvj"w ist int Ikw4&qvJdefkst4c&mt-Z4 wic

                                                                                       30


                        Figum 3b. Synoptic Meteorological Analysis for the Mon-iing, September 10, 1893 (Koppen,


                                                                                                                   8






                UNITED STATES ANNUAL AVERAGE TEMPERATUF


                TEMP

                  13-






                  12-

                                                                 I@A A      11






                  10


                    1890 1900 1910 1920 1930 1940 1950 1960 1970 1980




           Figure 4. Annual Mean Temperature in the United States (M. La Brecque, 1989).




                                                       9








             1.2 Atmospheric Circulation

                 In general, climate in the United States is controlled by global-scale circulation
             of the atmosphere but is influenced by coastal oceans linked to major oceanic current
             systems. Over the United States, this broad-scale atmospheric circulation includes
             a mid-czmtinental trough of the westerly wind and subtropical anticyclones on the
             south east and south west coasts. Other broad-scale circulation features that are
             associated with change in the anticyclones and trough are the Aleutians-area low
             atmospheric pressure (the Aleutians Low), the North Pacific Subtropical Anticyclone,
             and the North Atlantic Subtropical Anticyclone (the Bermuda High). These
             circulation features of the mid-latitude westerly wind are influenced by broad-scale
             atmospheric circulation in equatorial regions. One of these circulation systems is the
             Hadley Circulation.

                 The Hadley Circulation is driven by atmospheric heating in equatorial regions
             that produces vertical air-mass advection in approximate proportion to the amount
             of heating. The vertical motion leads to meridional advection of air-mass to the
             latitudes of the subtropical anticyclones. The size of the anticyclones is correlated to
             the vertical air-mass advection. Changes of the Hadley Circulation in the Pacific and
             Atlantic areas are caused by wind-forced oceanic circulation (and the resulting water
             temperature changes) that is correlated to the size of the subtropical anticyclones
             (Godshall, 1990). The Circulation changes are also correlated to changes in the
             California Current and the Gulf Stream (Wallace and Gutzler, 1981). The eastern
             part of the Pacific subtropical anticyclone influences meteorological conditions in
             Southern California. The western part of the Atlantic subtropical anticyclone, the
             Bermuda High, influences conditions in southeastern United States.

                 Research seeks to identify conveniently monitored features of circulation that may
             be used as analogues of the ensemble of changes related to the circulation. Records
             of change in analogues may be useful indicators of climate change trends. Variation
             in the Hadley Circulation of the equatorial Pacific and Atlantic oceans has been
             related to multi-year variation of the subtropical anticyclones (Godshall, 1971).
             Therefore, this circulation may prove to be a useful analog of meteorological and
             oceanographic changes on southern coasts of the United States.

                 Between 1916-1955, there was intensification of the Bermuda High (Tollner,
             1956), a pressure decrease between Iceland and Labrador, intensification of trade
             winds, and increased water temperatures in the Gulf Stream south of Newfoundland
             (Lamb and Johnson, 1959; Bjerknes, 1960). These changes of the Atlantic ocean and
             atmospheric circulation may have been related to change in the Hadley Circulation
             but there was no measurement of the Hadley Circulation to verify this. Water
             temperature anomaly from the 1900's through 1960's (Jones et al., 1982) showed the
             water temperature differences between equatorial and north Atlantic were larger
             during the period 1915 to 1940 than after 1940.


                                                        10









                In the United States, location of the mid-continental trough of the westerly winds
            influences meteorological conditions over eastern parts of the country including the
            coastal regions. Significant change of meteorological conditions could result with a
            climate-related change in this trough as well as in the long-wave pattern of general
            atmospheric circulation. The trough is influenced by flow of westerlies over the
            Rocky Mountains. The mountains cause wind deceleration in the lower troposphere
            and a pattern of convergence and divergence of air-mass in the lee-waves of the
            mountain range. These wave patterns influence precipitation distribution in central
            United States (Reiter, 1961). Seasonal variations in atmospheric heating in the
            northern Hemisphere do not produce significant shifts in the mean position of the
            long-wave patterns (Bolin, 1950) and the orographic effects seem to be most
            influential for trough position. Therefore, a tentative deduction is that climate
            change from atmospheric warming will not cause shift in the position of the
            mid-continental trough.


            1.3 Physical Properties of the Marine Environment

                Physical properties of selected marine coastal regions are reviewed here to provide
            background for description of changes in some of the variables. Any area of United
            States coast may have different environmental response to global-scale climate
            change than another area because of different exposure to ocean circulation effects
            etc. Three coastal areas are analyzed here to provide a representative assessment
            of possible responses. Within any of the areas many characteristics of the
            environment may change but these analyses will address environmental response to
            wind, river runoff, and sea level changes that are caused by climate change.

                The Mid-Atlantic region, roughly defined as the coastal region seaward to the
            shelf break from Cape Hatteras to Long Island, is selected for analysis because of
            large estuarine environments that are sensitive to variation in precipitation in
            northeastern United States and to oceanic circulation features of the shelf. The
            Northwest Gulf of Mexico is selected for climate impact analysis because of the
            influences of river runoff variations on water quality and the control of shelf
            circulation by wind-stress. A major source process for the nutrient chemicals input
            to the Gulf of Mexico ecosystems is the seaward flux of chemicals that results from
            flooded coastal margins. The processes associated with flooding may be altered by
            sea-level rises commonly associated with global climate change. The Southern
            California coast is selected for analysis because it is an area of coastal-water
            circulation forced by winds that are part of the broad-scale atmospheric processes in
            Hadley Circulation.








             1.3.1 Afid-Atlantic Bight

                The lacustrine estuaries of the area, Chesapeake Bay and Delaware Bay, have
             physical characteristics that are expected to respond to climate change effects
             differently from the responses of the open coast that are analyzed here. The special
             consequences of climate change on these water bodies will not be presented in this
             report.



                  Md-Atlantic Circulation


                The prevailing current on the open shelf is southward at all depths (Bumpus,
             1973) but the near-shore waters, within the 90m isobath, have spatially and
             temporally varying circulation that changes with wind forcing, tidal influences, and
             stratification. In deep waters where the prevailing current is southward in all
             seasons and not influenced by seasonal wind changes, the circulation is a geostrophic
             balanced flow where the denser water is seaward. This southward current is
             assimilated into the western margin of the Gulf Stream over the area between
             Chesapeake Bay and Cape Hatteras.

                A component of the southward drift is a cold water coastal-drift that moves
             southward from the Gulf of Maine (Godshall et al.. 1980). A subsurface part of this
             drift (identified as Gulf of Maine intermediate water) contributes to the cold pool
             water on the southern flank of Georges Bank, and this water is found in a bottom
             layer current of the mid-shelf (Hopkins and Garfield, 1977). Production of the cold
             pool water-mass is dependent on severity of winter weather; the greater the
             wind-forced mixing and surface cooling, the larger the pool.

                Because of out-flow of fresh water from the Hudson River, there is a weak
             southward drift of Bight water along the coast of New Jersey. The drift is commonly
             obscured by local wind forced circulation (Nuzzi, 1973). The chemical-nutrient and
             waste loadings to Bight waters contributes to lowered concentrations of dissolved
             oxygen in the Bight and in coastal waters of New Jersey (Swanson and Sindermann
             (eds), 1979, Segar and Berberian, 1976; Stoddard, 1989). The southward drift may
             continue along the coasts of Delaware and Virginia (Williams and Godshall, 1977)
             under conditions of maximum stratification in late summer. At this time of year,
             winds are relatively weak and the stratification reduces wind forced acceleration of
             the bulk of coastal water mass.










                                                      12








                 Summer season winds are typically southwesterly and, at depths less than 50m,
             the near-shore drift is northerly with upwelling in the area off Chesapeake Bay and
             coast of New Jersey (Norcross and Austin, 1988). This causes the cold pool waters
             to move shoreward and maximum bottom water temperatures are delayed until fall
             (Hicks and Miller, 1980) when wind becomes northerly and upwelhng ceases.
             Autumnal cooling of coastal surface waters causes overturning and isothermal
             conditions (Houghton et al., 1982), general obscuration of the cold pool, and a
             seasonal trend of decreasing bottom temperatures.

                 North of the Bight, variation of the water temperature in the Gulf of Maine (Gulf
             intermediate water is a source of water for the cold pool.) is responsive to oceanic
             circulation in high-latitude regions. Part of this circulation, the Scotian Current (a
             part of the water-mass that compensates northward moving water mass in the Gulf
             Stream), flows past Cape Sable into the Gulf of Maine. This current is a cold low
             saline flow that carries runoff from the St. Lawrence River. It contributes to the
             seasonally varying cyclonic gyre of the north-central part of the Gulf of Maine. Gulf
             circulation has larger southerly currents along the western shore in spring caused by
             seasonal      .mums of river runoff. These inflowing water masses into the Gulf of
             Maine contribute to low Gulf salinity and maintenance of relatively low water
             temperature.



                   Afid-Atlantic Sea Level

                 Sea level changes are not expected to be the most significant consequences of
             global climate change in the Bight because of relatively large slopes of coastal land
             areas. The effect of water level on circulation within estuaries may cause important
             environmental changes that will not be reported here.

                 Change in Bight-area sea level is associated with change in the earth's surface
             elevation, wind stress, and water density change. Monthly average sea level records
             from tide gauges at Charleston, SC and Sandy Hook, NJ (Thompson, 1990) have
             relatively large annual variation. This is attributable to geographic orientation of the
             coastal areas relative to the seasonal wind direction changes.

                 At stations north of Rhode Island, less than half of the annual variability in sea
             level has been attributed to winds and atmospheric pressure (Thompson, 1990). At
             Charleston and Sandy Hook (stations where the coast is oriented in the direction of
             prevailing summer wind) sea level is influenced by wind (Thompson, 1990). Sea level
             change has been associated with changes of the Gulf Stream (Blaha, 1984) in shelf
            .areas south of the middle Atlantic Bight. Since some Gulf Stream changes are
             correlated with large-scale wind fields (Godshall, 1990), association of any specific
             cause of long-period (i.e. interannual) sea level change is obscured.



                                                        13








                 Sea level along the middle Atlantic coast rose 10cm. over the period 1930-1970
             (Meade and Emery, 1972) but components of the change were not resolved in this
             source. Sea level rose about 10cm in the period 1950-75 at Sandy Hook but local
             differences are expected because of the local circulation. Annual variation of sea level
             on the shelf near the mouth of Chesapeake Bay was estimated to be about 17cm;
               * *mum sea-level height corrected for atmospheric pressure effect occurs in spring
             and maximum heights occur in fall (Montgomery, 1938). The rate of sea level rise is
             about twice the rise on the west coast and annual variations are less than those
             generally observed on the west coast. Largest rates of change, by comparison, occur
             along the northwestern Gulf coast (Hicks et al,1983).


                   Mid-Atlantic River Runoff

                 There is an annual variation of salinity on the shelf region of the mid-Atlantic
             Bight that is caused by annual variation in river runoff. The seaward salinity
             gradients from the surface in-flows of fresh water are affected by transient in-drifts
             of saline slope-water which may intrude to regions of the 20m isobath. The in-drift
             water mass is separated from the on-shelf water mass by oceanic salinity fronts that
             bring local change in water column sediment loading and chemical concentrations
             with salinity change. Stratification is variable; it is dependent upon the amount of
             surface in-flow and the occurrence of wind events to force vertical mixing of the water
             column (Bumpus, Lynde, and Shaw, 1973). The stratification of near shore water
             mass, caused by seaward flows of fresh water, is at a maximum in mid to late
             summer in the vicinity of Chesapeake Bay and other fresh water sources.

                 Since variation in evaporation and moisture content of saturated air is directly
             related to the air temperature, a local change of air temperature may be related to
             precipitation changes. The lack of correlation between temperature and precipitation
             change trend (Hanson et al., op. cit) over the United States as a whole may result
             from precipitation change dependence on moisture advection as well as other
             parameters. A change of temperature of the 19 year period average 1950-68
             compared to the temperature average from the period 1931-49 was about -1' C in the
             region of the Delaware River water shed. Concomitantly, the precipitation in the
             relatively cooler' 1950-68 period decreased about 0.5mm per day in the water shed
             (Karl and Riebsame, 1989). Runoff into the mid-Atlantic bight from the Potomac,
             Delaware, and Hudson River watersheds is illustrated in Figure 6. by the graph of
             gauged flow rates from selected stations within these watersheds. Although the runoff
             rates from the periods beginning in the 1970's are large, the rates are not large
             relative to preceding decades. An 11 year period of variability is a significant feature
             of these records. Such changes in runoff effect water quality in embayments of the
             Bight (Officer et a-I., 1984).




                                                        14
























                                                      LATITUDE BAND 64       90ON




                    0.5-

                    0.4-

                    0.3-

                    0.2-
                    0.1-.
                    0.0                                           VV,
                    0.1-

                   -0.2-

                   -0.3

                   -0.4-
                   -0.5-                              LATITUDE BAND 44       640N
                   -0.6'



                    0.4-

                    0.3

                    0.2-

                    0.1-

                    0.0-

                    0.1-
                   -0.2-'
                   -0.3                                 LATITUDE BAND 24       440N
                   -0.4-      ---rTr-rT-rT-T         I..'' ...      ... I I.........7
                       1880     1900      1920     1940      1960     1980      2000

           Figure 5. Northern Hemisphere Surface Air Temperature Variations Averaged for
                      Selected Latitude Bands (Hansen and Lebedeff, 1987).


                                                  15











             1.3.2 Northwest Gulf of Mexico

                The runoff and siltation from the Mississippi/Atchafhlaya rivers are dominant
             control sources for the coastal palustrine system. The bays of the coastal margin are
             important parts of the system because of biota that they support. Therefore, the
             environmental response of these areas to climate change could be described in
             relation to the coastal ecosystem biota. However, the orientation of this report to
             physical parameters of change directs these analyses primarily to the coastal physical
             responses.



                  Northwest Gulf of Mexico Circulation


                The circulation on the northwestern shelf of the Gulf of Mexico is
             quasi-independent of the general circulation in the western Gulf basin. An
             independent, on-shelf circulation is established because the width of the shelf (over
             200krn to the 100m depth) and shallowness tend to isolate the shelf water mass from
             the dynamics of the deep water circulation regimes of the western basin. The coastal
             boundary of the shelf is uniformly arcuate with offshore barrier islands and relatively
             featureless bottom where bathymetric slopes are in the order of 1:1000. These
             characteristics promote the maintenance of seasonally established shelf circulation
             (Cochrane and Kelly, 1986).

                The shelf circulation of fall and winter is an apparent cyclonic gyre with
             peripheral flows parallel to the coastal bathymetry and the shelf break. Weakening
             of this cyclonic circulation in spring begins along the south Texas coast in association
             with seasonal development of a southerly wind regime. By mid-summer the
             south-wind forced anticyclonic circulation dominates the shelf with a small portion
             of the shelf, near the Mississippi delta, under a continuation of the cyclonic gyre.
             Cross-shelf bottom flow results from downwelling along the coast in association with
             the anticyclonic gyre (Snedden et al, 1988) and from tidal currents. With onset of
             northeasterly wind in fall, the cyclonic gyre is reestablished over the shelf The
             near-shore component of the cyclonic gyre is accentuated by the westerly drift of the
             Mississippi and Atchafalaya river outflows. This near-shore drift and other fresh
             water inflows produce an across-shelf gradient of salinity and density stratification
             in near-shore water masses.

                Fall and winter northeasterly wind is associated with weak anticyclones in
             Louisiana and in the southeastern states associated with westward extension of the
             Bermuda High. These northeasterlies become less frequent as the southerly wind
             regime of summer becomes dominant over the northern Gulf. A propensity for ano3dc
             bottom-water development in spring and early summer is established by
             stratification; the seasonal runoff is large, the weak southerly wind regime does not


                                                        16









             force vertical mixing of the water and solar heating of surface water increases
             stratification. An occurrence of northeasterlies late in spring can be significant with
             respect to production of anoxic bottom water. Wind stress from the northeasterlies
             forces the near-shore fresh water to spread south and west over a larger part of the
             shelf area producing a larger area with stratified water and, therefore, increased
             possibility of anoxic bottom water development (Harper et al., 1981). Therefore,
             changes of the Bermuda High may effect wind over the no;ih-em Gulf and cause
             undesirable water quality changes.


                   Northwest Gulf of Mexico River Runoff

                Physical characteristics of the shelf environment are expected to respond to
             climate variations that change the rate of discharge of fresh water from the
             Nfississippi and Atchafalaya rivers. The Nfississippi watershed (Figure 7., Page 19)
             covers most of central United States. This very large area carries runoff that has,
             effectively, integrated over the short time (hours) and space scales of myriad
             precipitation events within the central plains region. During the period 1950 to 1968,
             precipitation of the U.S. Central plains areas increased by about 0.25 mm/day
             (concomitantly with a temperature decrease of about 1' C) compared to the period
             1931-49. This represents an increase of about 13.5% for a representative station,
             Concordia, Kansas, based on U.S. Weather Bureau (1930) precipitation statistics.
             From regression analysis on precipitation data and stream gauged runoff (Karl and
             Riebsame, 1989), changes of runoff were expected to increase in the Nfississippi water
             shed. However, the flow rate at Vicksburg, Afississippi gaging station decreased
             relative to the comparison period, 1931-49 (Figure 8., Page 20). During the 1970's,
             larger flow rates at Vicksburg were associated with larger amount of precipitation
             (Hanson et &I., op. cit). The 1970's increases in runoff also occurred from the
             Appalachians (Figure 6., Page 18).


                   Northwest Gulf of Mexico Sea Level

                 Sea level change in the northern Gulf of Mexico over the past few decades has
             been largely caused by coastal land subsidence. Largest rates, about 0.63 cm per
             year in the vicinity of Galveston, TX (Hicks et a-I., 1983) are attributed to removal of
             gas and petroleum resources from coastal wells. This rate of change is not
             representative of the northern Gulf coast as a whole. Other factors causing change
             are steric effects and river runoff rates.








                                                        17





                                     WATERSHEDS OF THE MID.-ATLANTIC
                                                                 LEGEND: DELAWARE
                                                                              HUDSON
                                                                              POTOMAC


                        FLOW

                          26--



                        S 21'
                       .C
                        A
                        L
                        E 16-
                        D


                        C
                        F
                        S                                                          W6
                                                                                                     (D     (D
                                                                           (D (D       6                        fflb%ï¿½D
                                                                                                         . (1@0    (D
                           6-                                                                     do                  $
                                             ++++++++++++++4-+++ ++++++4     + ++++++4+             +4++++4++
                                                                          ++4++++     + + _++    +++



                            1895                                          1945                                           1995

              Figure 6. Runoff Rates from the Potomac River (Point of Rocks guage), Delaware River (Wallenpaupack and Bush
                          Kill Creeks), and Hudson River (Rondout Creek) Watersheds. Annual Means of Guaged Flow are
                          Smoothed by a 5-year Running Mean.



                                                                          18































                                                             I I --- - - - - - -



























              Figure 7. The Afississippi River Watershed (N. Rabalais and D. Boesch, 1990: EPA Coastal C
                         Workshop, Jan. 8-9, 1990, Galveston, Texas, illustrations by R. E. Turner).


                                                                     19































































                                                           N.
                           A       A            N         /N.
                            @v V            y    V          V


                                             YEAR












           Figure 8. Annual Discharge of the Mississippi River, Measured at
                     Vicksburg, MS 1931-1977 (Dinnel and Wiseman, 1986).


                                                20








             1.3.3 Southern California

                 Southern California is addressed here as the coastal area south of San Francisco.
             The area is chosen as an analog of western-coastal responses because of the
             sensitivity of the marine circulation to broad-scale wind regimes associated with the
             Hadley Circulation.

                   Southern California Ocean Circulation

                 The California shelf width is narrow, about 6km out to 100m depth, but about
             110km to the 2000m isobath off San Francisco and about 260km to the 2000m
             isobath off San Diego. Because of the narrow shelf width north of Point Conception,
             this near-shore area is exposed to deep-water wave and current regimes (Godshall
             and Williams, 1981). In the Southern California Bight, the Channel Islands and
             wider shelf protect the coast.

                 North of Point Conception, an oceanographic trough of dynamic depth is oriented
             along the mid-shelf isobaths in winter and seaward of the Channel Islands of
             Southern California. This trough position is synoptic with surface near-shore
             northward currents. In spring the trough is contiguous with the coast, through the
             Bight and north to Cape Mendocino and northward surface currents in the near-shore
             areas cease.


                 The north flowing Davidson Current in the near-shore area north of Point
             Conception is a winter circulation feature on the surface that is synoptic with a
             deep-water northward flow, the California Undercurrent, in mid-shelf. These
             currents transport a warm, saline water mass northward.

                 The surface layer southward flowing California Current is a feature of mid and
             outer-shelf regions that is poorly defined in winter but well organized in spring and
             summer. This wind-driven current brings cool, low-saline waters southward. The
             near-shore portion of California Current has a seaward component caused by Ekman
             transport that is identified with upwelling along the coast.

                 In the Southern California Bight, surface near-shore drift is northward in late fall
             and early winter; it leads the transition to winter seasonal-circulation of the shelf,
             north of Point Conception. Transition toward southerly flow in early spring is,
             apparently, a combination of weakening in a cyclonic circulation eddy south of Point
             Conception and shoreward movement of the California Current system. Shoreward
             of the Channel Islands, in the area between Santa Barbara and north to Pt.
             Conception, the cyclonic eddy may exist in any month and this produces variable
             northerly drift at the shoreline. In early fall, the flow inshore of the channel islands
             returns northward, that is a surfacing of the California Undercurrent in the Bight.



                                                        21







                Local circulation is strongly influenced by submarine canyons in the southern
             California shelf area. These canyons, e.g. Santa Monica Bay, cause spatially varying
             currents from tide and wind forcing which result in eddies. Eddies cause local
             changes in vertical stirring of the water column and local variations in water
             temperature (Leipper, 1955). Kelp beds affect water temperature on beaches from
             Pt. Conception southward by sheltering beaches from cool water of the California
             Current (Kolpack, 1971).


                  Southern California Atmospheric Circulation

                The general ocean circulation along the coast is forced by broad-scale wind
             regimes with local modifications from sea-breeze circulation. Therefore, appreciation
             for potential impact on the coastal environment must include some explanation of the
             prevailing wind regimes.

                Wind is northerly along the California coast and seasonally variable, primarily
             with respect to speed. Seasonal change occurs with change in the eastern Pacific
             Subtropical Anticyclone, a ridge of high pressure over northwestern United States,
             and a low pressure area in southwestern United States that is produced from thermal
             effects. Change in the subtropical anticyclone is inversely associated with change in
             the Aleutians-area low pressure (the Aleutians Low). Since the California Current
             is wind driven, the seasonal variation in wind is associated with seasonal variation
             in coastal circulation and water temperature.

                In winter the Aleutians Low deepens, the subtropical anticyclonic area decreases
             and moves southward, and the ridge of high pressure in the Northwest may become
             contiguous with a weak ridge that replaces the siimmer-low in the Southwest. The
             decrease and southward migration of the Pacific subtropical anticyclone is synoptic
             with southward migration of the polar front. Storms, developed along the front, may
             move into California coastal areas. Northerly wind along the coast is, relative to
             other seasons, more variable (wind constancy is about 20-40%). In the Southern
             California coastal-bight, wind from interior high pressure areas subsides from
             elevated land areas to produce a dry seaward flow of air (DeMarris et al.. 1965).

                In summer, the Pacific subtropical anticyclone is seasonally largest; it extends
             northward over the latitude of Hawaii and eastward to coastal California. A low
             pressure area commonly develops from surface heating in southwestern United
             States. The coastal northwesterly wind is seasonally strongest and consistent in
             direction. Wind constancy (ratio of vector and scaler means) is about 80-90%. In the
             Southern California coastal bight, shoreward of the Channel Islands, wind is variable
             and influenced by a sea-breeze circulation.




                                                      22







                Seasonal atmospheric circulation along the U.S. west coast is readily associated
            with broad-scale atmospheric circulation and interhemispheric changes (Dickson and
            Livezey, 1984). In the eastern Pacific, low-latitude regions circulation changes are
            correlated with pressure anomaly in the north Pacific (Bjerknes, 1969; Namias, 1985).
            However, the specific processes producing correlation between latitudinal change of
            circulation on these scales are unknown. Long period change in wind along the U.S.
            west coast, such as increased coastal wind stress 1946-1988, was deduced through
            water temperature changes (Bakun, 1990). These temperature changes are typically
            produced on eastern oceanic boundaries from wind effects. A period of cool water
            temperature along the southern California coast in 1955 was produced with buildup
            of the east Pacific High and intensified northerly wind. The subsequent shift to
            warmer water in 1957 (Figure 9., Page 24) was accompanied by a weakened high and
            deepened Aleutian Low. The weakened northerly winds were accompanied by a
            simultaneous warming of air temperature but the warming was not interpreted as
            caused by warmer coastal water (Namias, 1960). At the same time there was a
            substantial rise in sea level (Stewart, 1960).

                Local modifications to the northerly winds produce locally varying water
            temperature. In Santa Barbara Channel, the land-sea breeze regime produces a
            southwesterly wind at the shore that backs to a westerly or northwesterly wind
            further offshore. This local wind affects the amount of upwelling that may result in
            warmer coastal water because of opposing circulation from wind stress. The offshore
            northerly winds cause upwelling and cooler surface waters with the result that there
            is a complicated and variable pattern of water temperatures across the shelf. Santa
            Monica Bay is a trap for locally advected warm water as well is the Ventura
            embayment. Headland sheltering of the embayments also produces locally unique
            coastal water circulations which affect water temperatures. As a consequence of
            these local pattems of water temperature, there is considerable coastal variation in
            climate effects from fog and low stratiform clouds.

















                                                      23

















                            +2.6 7                                         UNITED STATES WEST COASTAL STATIONS
                            +2.0 -                                         SEA SURFACE TEMPERATURE ANOMALIES
                                                                                 12 MONTH RUNNING AVERAGE


                            +1.0

                            +0.5

                                0

                            -0.5

                            -1.0

                            -1.5


                                 1917191819191920192119221923192419251926192719281929193019311932193319341935193619371938193919401941194219431944194519"19471948
                                                      .   .   9   1   .   I   I   .   I       I  I    I  I    .  I   .   .   I  .   - so

                                                    330 - 40*N U.S. WEST COAST STATIONS                                             - 70
                                                  SEA SURFACE TEMPERATURE ANOMALIE5                                                   60
                                          +1.0                                                                                        50
                                                                                                                                      40
                                          +0.5
                                              0                                                                                       30

                                          -0.5
                                          -1.0                                                L               L
                                              I;i-ML9r-OL195ILlidlii@li;tli;tli;@Llii@li;tl959'lii@196lLl9r32LI1196519661967196819691970




                 Figure 9. Sea Surface Layer Temperature Anomaly (1C) of the U.S. West Coast (analysis by Godshall and Williams,
                                1981).


                                                                                               24










                  Southern California River Runoff

                Variation of precipitation and land runoff from the coastal area of southern
             California correlates with broad-scale pressure distribution over the north Pacific.
             The Arroyo Seco river flow rate, for example, is negatively correlated with pressure
             anomalies in the north-central Pacific region (Cayan and Peterson, 1989; Namias,
             1980). However, the correlations are relatively weak and increased runoff, expected
             during the 1957 period of weak coastal pressures, did not occur in all major rivers of
             the study region, e.g. the flow rates of the Sacramento, Merced, and the Arroyo Saco
             rivers graphed by Cayan and Peterson (op. cit). Longer temporal-scaled variations
             of the pressure and associated runoff changes are found in all rivers of the region.
             Changes in precipitation/river-runoff onto the coast are illustrated by the gauged flow
             rate records of the Arroyo Saco river (Figure 10., Page 26). Seasonal changes in
             runoff rates are also expected due to early-season snowmelt with warmer climate
             (King et al. 1990).

                River runoff into the Southern California Bight influences beach configuration by
             supplying sedimentary material distributed along the coast. Large runoff events
             associate with increased concentrations of silicates in the near shore area of the Bight
             (Tibby and Terry, 1959).


                  Southern California Sea Level

                The average of hourly measured water elevations over yearly periods is defined
             as the annual mean sea level. Along most of the California coast, mean sea level has
             been rising relative to sea level for the 1941-1959 National Tidal Datum Epoch with
             elevation rise of about 0.15 cm/year at San Francisco (Harris, 1979). Sea-level
             increase rate of 0.12 cm/year is reported over the period 1855-1922 at San Francisco
             (Smith and Leffler, 1980).

                Coastal sea level varies seasonally with changes of more than 20cm over the year
             (Harris, 1979). Sea level variation caused by transient oceanographic conditions can
             be as much as 30cm (Smith and Leffler, 1980). Off San Francisco, sea level falls with
             the spring seasonal increase of the north-wind stress and surface cooling from
             upwelling. South of Point Conception, sea level decreases occur in late winter and
             decreases are more variable in magnitude. Assuming that the change of California
             coastal water temperature is proportional to the upwelling and decreased coastal sea
             level, change of coastal water temperature from the early 1900's to 1970 (Figure 9.,
             Page 24) suggests that sea level has had large fluctuations in the past. Since 1917,
             sea water temperature change shows no long-period trend of temperature and,
             therefore, there is no evidence to suggest a long period change in upwelhng along the
             coast. However, there was a 3.6cm decrease in sea level over the period (Huang,
             1972).


                                                       25









        50


        40


        30


        20



        10



        0








        50



        40



        30


        20



        10



        0



                     Ln                             co
                                        tQ          IIIJ



      Figure 10. Monthly Mean Mow Rate of the Arroyo Saco River Smoothed by a 12-Month Running Mean.



                                26








                Graphs of water temperature from San Diego and Los Angeles show water
            temperature increased during the period of climate cooling (from about 1945) when
            coastal wind and upwelling were expected to decrease (Figure 11., Page 28).
            However, the temperature record from may be influenced by local process that are
            not typical in other coastal areas of California. The seasonal variation in sea level
            produced by atmospheric pressure variation is about 4cm in the San Diego, CA area
            and seasonal variation is out of phase with seasonal heating and cooling cycles. This
            is interpreted as evidence of influence from the broad-scale geopotential surfaces that
            are established by the major oceanic circulation patterns (Reid and Mantyla, 1976).
            Therefore, rise in water temperature need not be interpreted as a decrease in coastal
            upwelling.


































                                                       27









                     F- ----SOUTH ER_N__ CALI FOR N I A                      ANNUAL MEAN                      TEMPERATURE 1900-1960

                         70






                          65-
                                                                                                                                                          S N
                      LLJ                                                                                                                                DIE
                                                                                                                                                         MEAN
                      LU                                                                                                                _11,,/L.A.INT AIR OR
                      1:60 -
                                                                                                                                    V/              MEAN

                                                                                                                                               SANTA MARIA--'--
                                                                                                                                                   ME A N

                      Uj
                      W
                      ac
                      cq 55-
                      Uj                                                                           SAN DIEGO

                                                                                               - - -L.A. INT. AIRPORT

                                                                                               ....... SANTA MARIA

                         50-

                           1900        05         10         15         20        25          30        35         40         45         50         55        1960
                                                                                                                                                     FIGURE  2.3
                                                                                                                                                             .k
                                                                                                                                                           AN",
                                                                                                                                                           IEGO
                                                                                                                                                         E N
                                                                                                                                                         MI
                                                                                                                                                  _@A R
                                                                                                                                                L.A.'INT
                                                                                                                                                    MEAN














                  Figure 11. Water Temperature of Coastal, Southern California (California State Water Quality Control Board, 1965).


                                                                                             28










            2. MODEL PREDICTION OF CLIMATE

                General Circulation Models (GCMs) used for assessing possible effects of climate
            change operate with the same application of the laws of fluid dynamics as the models
            for meteorological forecasting (Hansen et al., 1983). Although GCMs will not provide
            forecasts of meteorological variables &t are extensions of the usual commercial
            forecast periods, GCMs can produce balanced distributions of energy/energy fluxes
            and fluid mass related to predicted atmospheric forcing. These distributions result
            from simulation of ocean and atmospheric processes by analogues of the actual
            environmental processes and, therefore, they require validation that is commonly
            provided by the climate at' X C02- Modeling technology research seeks improved
            process generalizations of the atmosphere and oceans that reduce requirements for
            process rate specification and data (Grassl, 1988). Spatial scales for modeling are
            sought to optimize model results relative to input data requirements. During the
            present time, modeling operations also are limited by computing capability and
            accurate specification of surface fluxes of moisture, heat, and momentum. The effects
            of these limitations are commonly evaluated by creating a model output of the climate
            of the present (the' X C02 climate) that can be compared to observations.

                Definition of climate change, with a doubled carbon dioxide concentration in
            equilibrium with all parts of the environment, is a basis for modeling result
            intercomparison (Figure 12., Page 30). These latitudinal temperature distributions
            show agreement within the compared modeling result; the temperature of the
            latitudes of the United States are predicted to increase relative to the temperatures
            of the equatorial region with atmospheric carbon dioxide concentration doubled.
            However, the time for earth's atmosphere to reach a doubled carbon dioxide
            concentration is poorly estimated because the rate of approach to equilibrium of
            carbon dioxide in the atmosphere with the concentration in the oceans varies as the
            rate of carbon dioxide increases. Carbon dioxide uptake by oceans is inversely
            proportional to atmospheric carbon dioxide rate of increase, i.e. halving the rate of
            increase will double the time for the ocean uptake to reach equilibrium with the
            atmosphere (Grassl, 1988). Also political and social factors governing the rates of
            carbon dioxide increase are unknown.

                The most important parameters of climate with sensitivity in modeling of climate
            are those used to characterize atmospheric radiation balances. Since carbon dioxide
            influences radiation balances, the concentration of atmospheric carbon dioxide is
            theoretically an important parameter. Cloud amount and atmospheric water vapor
            are as important as carbon dioxide in causing climate change. That places the
            parameters related to surface flux of water vapor to be important also
            (Hartmann, 1984). Doubled carbon dioxide model runs by the Goddard Institute of
            Space Science (GISS) model (Hansen et a-1., 1984) predicted increased air
            temperature of the United States and coastal areas (with increases of about 3* to 5'
            C).


                                                      29
































                                                                             1OC

                                  10-a =Manabe and Wethercld 1980            10
                                  9- b =Schlesinger 1982                     9
                                     d =Washington and Meehl 1983
                                  8-           clouds calculated             8
                                     c =Washington and Meehl 1983


                                  6                                          6


                                  5


                                  4-                                         4

                                  3 - \b                                     3
                                                         PV                  2
                                  2-





                                  9ON 70   50  30  1ON 10S   30 50    70   90S
                                                    Latitude









             Figure 12. Longitudinal Mean Temperature Changes Predicted by GCMs Under
                            Doubled Atmospheric Carbon Dioidde Concentration (Grasal, 1988).
                                     c  Washin







                                           a
                                              gton  a
                                  - @\b

                                                            30








                The prediction of large coastal temperature increases are predicated on a cloud
            model that leads to decreased low and middle-level cloud with increased convection.
            Some argument might be offered to counter this assumption in coastal regions with
            cold surface water. Evaporation increases nonlinearly with temperature which leads
            to larger vapor pressure increases relative to temperature increases. Increased water
            vapor in surface air-mass that may be advected over cold water surface is expected
            to lead to increased low-level clouds.

                Positive temperature feedback from increased water vapor concentration is
            approximately compensated by negative feedback from lapse rate changes. Therefore,
            cloud effects are control factors for local temperature changes. Coastal low-level
            stratiform cloud are produced over water areas where there are cold water surfaces
            and, since these cold surface waters commonly develop from wind-forced upwelling,
            the GCM must be coupled to coastal oceanic circulation simulation. The accurate
            coupling of oceanic circulation and atmosphere is also important for GCM predictions
            of climate in the coastal areas of the United States where low-latitude Hadley
            Circulation influences climate.

                Numerical experiments using the ECMWF (European Center for Medium Range
            Weather Forecasting) general circulation model (Cubasch, 1985) demonstrated
            association of enhanced Hadley Circulation with mid-latitude atmospheric dynamics.
            Enhanced Hadley Circulation, a consequence of positive temperature anomaly in the
             quatorial Pacific, led to Aleutian Low deepening and higher pressure over the area
            of easter*n maritime Canada. Positive surface pressure anomalies in the Pacific are
            e

            correlated with increased size of the Bermuda High and this change of pressure
            distribution is an expected result of an enhanced Hadley Circulation.


            2.1 Interpretation of Modeling Results

                Figures 13a. and 13b., Pages 32 and 33, are vector averaged winds predicted by
            the NOAA, General Fluid Dynamics Laboratory (GFDL) GCM model (Jenne, 1989;
            Manabe and Wetherald, 1987) for August with atmospheric carbon dioxide
            concentration of the present(' X C02) and with atmospheric carbon dioxide
            concentration doubled (2XC02). A comparison of the analyses shows strengthened
            southerly wind components in the Gulf of Mexico are predicted during 2XCO2
            conditions (Appendix C and Figure 14.). Water-mass circulation on the northwestern
            shelf of the Gulf of Mexico is normally under the influence of the summer southerly
            wind regimes (ab in'tra 1.3.2 Northwest Gulf of Mexico and, therefore, little change
            may be expected in the anticyclonic circulation of water on the shelf






                                                     31






















                                                                                      46
                                                                   let



























                                                                                           Ar--






             Figure 13a. August Wind Vectors Estimated by Numerical General Climate Model for an Atr
                           Concentration of the Present.




                                                                  32




























              0
          0   0
          r   0
       0                                                                                                               14




           0


                0 0




             (D (D
                :3 (D




                V











                                                                                                                                                                     10 m/se






                     Figare 13b. August Wind Vectors Estimated by Numerical General Climate Model for an Atm
                                           Double theC02Concentration of the Present.


                                                                                                              33




                                                                                  Not-












                                                                               TO

       PT T      T    ;R
       TIT I
                     . . ....... ....
                           ..........
                                                                         -"T    7'
          ...........  .......
                                                                                   ...........
                                                                             .... .. .. .... ....


                         ..................... .....................
                                                                         ..................... .....................
                ...............                                               ... . ...........


           .. ... .......






                                                                        .... ......... ..........
                                    k*'


                                                                                  . ............ ......... . ..........
                                                                              .... . .....
                                             k-   k-
                                                       711  P
                                                                                .... ...... ..........




   WIND  SPEED CHANGE   2XCO2    IXCO2 (W/SEC)               GFOL  GRID WIND  CHANGE    IXC
       LESS THAN -1.5 RM 0.5  TO 1 0
       - 1:. 5 TO - 10= I - 0 TO 1 5
   1=  -1 0 TO -0 5   = GREATER   T H A NI . 5               AUGUST -  I X C 0 2(BLACK  2 X C
       -0 5 TO 0 . 0
       0 . 0 TO 0. 5


          Figure 14. Comparison of Model Predicted Surface Wind from an Atmosphere with' X C02aj
                    Concentrations in August (NOAA GFDL Q-Flux Model Runs, 1988).


                                                          34








                Moisture advection into continental United States will be increased and this may
             result in increased precipitation amounts with possible increased river runoff.
             However, increased continental temperatures will increase evaporation and decrease
             runoff. In the past, increased flow rates of the Mississippi and Atchafalaya Rivers
             into the Gulf have increased areas of stratified coastal water masses and this has
             lead to anoxic bottom water problems on the shelf. Possible increases in runoff
             volumes under 2XC02could lead to similar anoxic bottom water problems following
             runoff events as occurred in the past. United States east coast August wind regimes,
             predicted by the' X C02 (Figure 13a.), are similar to the winds shown by the mean
             wind field on the COADS derived analysis (Appendix C.), winds are southerly along
             the coast. These winds produce coastal upwelling (ab in'tra 1.3.1 MIID-ATLANTIC
             BIGHT).

                Figures 15a. and 15b., Page 38, show the GFDL GCM predicted wind fields of
             January for' X C02and 2 X CO. respectively. The' X C02wind fields are similar
             to the monthly mean wind fields derived from COADS (Appendix C.). However, the
             2XC02fields are different on the east coast of the United States from the winds of
             the present times. The predicted changes are associated with change in the Bermuda
             High; winter coastal winds are forecast to be more southerly on the coast (Figure 16.)
             with veering in southern coastal areas. This is likely to cause coastal upwelling in
             winter and this could lead to warmer and more saline water along the coast in
             winter. In an analysis of the GFDL model-forecasted 2XC02precipitation over the
             United States (Finkelstein and Truppi, 1990), much of the seasonality of precipitation
             distribution of the present climate is lost over the mid-west and eastern states areas.
             This agrees with expectations deduced from moisture flux associated with westward
             extension of the Bermuda High in winter. Although predicted wind speed win
             increase for a 2XCO2 atmosphere, the increases are about an order of magnitude
             less than wind increases considered by Lantham and Smith (1990) for significant
             increase of aerosol generation at sea.

                Since GCMs commonly fail to produce the correct absolute mean atmospheric
             pressure on the surface (Gates et al., 1990), pressure distribution should be the
             Parameter of analysis. All models reproduce the subtropical anticyclones but there
             is a range in the anticyclone central pressure (which is generally too high in the
             northern hemisphere) and there is variation in the position of the anticyclones
             latitudinally (Gates et al., 1990). However, the similarity of the wind fields on
             coastal United States when the' X C02wind fields are compared to the wind fields
             derived from COADS indicates the GFDL predicted position of the Bermuda High
             must be correct. Therefore, the predicted wind fields under 2XCO2 should be a
             reasonable expectation.






                                                       35








                The high   pressures of the subtropical anticyclones produce large meridional
            pressure gradients in mid-latitudes. These gradients lead to strong westerly wind
            regimes that are sensitive to model representation of drag from mountain ranges
            (Slingo and Pearson, 1987). This could be of consequence in simulation of
            atmospheric circulation and precipitation patterns over central United States because
            of influences from the Rocky Mountains. However, comparison of the trough position
            deduced from mean continental pressure distributions (U.S. Navy, 1981) with the
            GFDL 2XC02pressure fields indicates the position is not predicted to change from
            that of the present.

                GCMs predict temperature increases under conditions of doubled carbon dioxide
            of about 2* to 50 C in the mid-Atlantic Bight area. Figure 17., Page 40, gives
            estimates of runoff that will come from the Delaware River watershed with 20 C
            increase in temperature and with varying amount of precipitation change. These
            variations in runoff at different stations within the Delaware River estuary as
            predicted by the different GCMs is given in Figure 18., Page 40. In comparison with
            these predictions, the Goddard Institute for Space Study (GISS) model result is much
            different from both the General Fluid Dynamics Laboratory (GFDL) model and
            Oregon State University (OSU) model output.



























                                                      36













                                                                                                                                                    54
                                                                                                                                                     0






                                                                                                                                                                                       14




















                                                                                 A-                                          Ar-

                                                                                                           -Af'

                                                         Ae@                                                                                                                                                              .k-








                                                                                                                                                                                                                                 M/Sec





                               Figure 15a. January Wind Vectors Estimated by Numerical General Climate Model for an A
                                                             Concentration of the Present.


                                                                                                                                                       37



















                                                                            (2




























                                       Ar'    hr-                                         Ar--









              Figure 15b. January Wind Vectors Estimated by Numerical General Climate Model for an
                          Double the C02 Concentration of the Present (NOAA GFDL Q-Flux Model Runs

                                                                 38
















                   ef:                                                                                                                                                                                                         -Af

                                          ..... .......
                                                                                                       7,
                           ... ....... ... ... .
                                                                                                                           ..............



                                                                                                                                        . ............

                                                                                                                                                            . . .. .. ............
                                                                                                                                                                  ..... .... .
                                                                                                                             . ...............     ...                .. .... .. .....
                                                                                                                                                                 . . . . ....... ..... ..... .........

                                                                                                                                                                                             .......... ..... ;%
                                                                                                                                                                                                    .. ........ ..

                                                                                                                                                                                                                           g
                                                                                                                                                                                                                           a
                                                                                                                                                                                                                             mg






                            . .... ....................



                                                    .. ...... ..
                                                                                                   .......... . ......
                                                                       .....                       ............ .... ... ....                                                          .................. .
                                                                                                                                                                                       .................


                                                                                                     ............... .... ...... . ... .

                                                                 k-
                                                                                                           . .. .. ......... .
                                                                                                                                                                                                                           ;MP.
                                                              1          F

          WIND SPEED                    CHANGE               2XCO2 - IXCO2                         (M/SEC)                                               GFDL GRID WIND                           CHANGE                   IXC
          1110 LE      SS THAN -1 .5 04 0.5 TO 1 .0
                   -1.5 TO -1.0                                  1 .0       70 1       .5
                   -1 .0 70             -0 . 5           111111111 GREATER             T H A N       I . 5                                               JANUARY - IXCO2 (BLACK), 2X
                   -0 . 5 TO            0 - 0
                   0   0 TO 0           - 5
                         Figure 16.                 Comparison of Model Predicted Surface Wind ftom an Atmosphere with I X C02 aj
                                                    Concentrations in January (NOAA GFDL Q-Flux Model Runs, 1988).


                                                                                                                                                 39








                            Measurement Stations, Delaware River Watershed

                                            Montrose, PA                      Trenton, NJ

            Precipitation at Present,          P = 486                    P   381

                                         +2'C temperature change

            Precipitation Change M          ..... runoff amount (mm) ......
                         P+20                      632                        501
                         P+10                      532                        411
                         P                         411                        329
                         P-10                      351                        258
                         P-20                      268                        188

                                          +4'C temperature change

                         P+20                      571                        431
                         P+10                      476                        359
                         P                         389                        286
                         P-10                      306                        221
                         P-20                      231                        157

            Figure 17. Estimated Runoff from the Delaware River Watershed With
                           Increased Air Temperature and Precipitation Changes (mm) in the
                           Watershed (M. Airs, 1990: Watershed Hydrology, Coastal Ocean
                           Physics Workshop, Jan. 8-9, 1990, Galveston, TX).





                           Measured                          General Climate Models
                       Precipitation (mm)                 GISS         GFDL          OSU

            Montrose, PA         486                      364          453           528
            Trenton, NJ          381                      245          353           397
            Dover, DE            347                      213          324           355

            Figure 18. Model Predicted Runoff (mm) in the Delaware River Watershed for
                           +2'C Climate Scenario (M. Airs, 1990: Watershed Hydrology, Coastal
                           Ocean Physics Workshop, Jan. 8-9, 1990, Galveston, TX).





                                                     40










           3. ANALYSIS FROM COADS:

               The Comprehensive Ocean-Atmosphere Data Set (COADS) included wind and
           atmospheric surface-pressure monthly mean data (for 20 square geographic areas)
           from 1854 through 1979 (National Oceanic and Atmospheric Administration (NOAA),
           1985). Means from the period of record (Appendix A) were mapped over the
           geographic area eastward from 130E to 1OW longitude and in latitudes from the
           equator northward to 40N. These data provide the bases to which monthly means
           from specific periods are compared in production of difference analyses. Three
           periods within the COADS record period are selected for this study. The first period
           represents conditions associated with a trend toward warm temperature in the
           Northern Hemisphere prior to 1900. The second period, 1935-44, is a relatively warm
           climate that is followed by a cooling climate. The third period, 1970-79, is a period
           of cool temperature with a trend to warmer temperature (Figure 1b., Page 4). A
           hypothesis is set forth that cool and warm period variations in climate of the
           Northern Hemisphere are associated with global-scale circulation changes in
           atmosphere and oceans. The net wind of the study area is a westerly with annual
           average speed variations associated with warming and cooling climate. These changes
           of the broad-scale wind regimes is shown by a graph of wind speed over the period
           of COADS analysis, 1890-1979 (Figure 19). A component of the global-scale
           circulation is the Hadley Circulation (ab in%ra 1.2 ATMOSPHERIC CIRCULATION)
           which has features controlling atmospheric advection of moisture (therefore
           precipitation and river runoff control) over the United States and hence freshwater
           input to coastal regions.


           3.1 Differences in Oceanic-Area Wind Regimes for Selected Periods

               Low-latitude wind regimes are part of the Hadley Circulation and these winds
           change over time (Godshall, 1971). Differences in these regimes for selected periods
           are computed to explain effects of global-scale circulation changes on U.S. coastal
           areas. Periods of climate variation are commonly considered over time that includes
           many years; specific months or seasons from a year are not germane in analyses of
           climate variation. Our basic wind-difference components are developed by year and
           then grouped to represent the historical periods of interest in the cool and warm
           climate variations. Wind-difference components were computed with a procedure
           developed by F. Godshall and J. Jalickee (Williams and Godshall, 1977) that is
           described in Appendix B of this report. Wind-difference factors, nu and phi, are
           computed in comparison of pairs of wind vectors (the long-term mean wind for each
           month of each 2* averaging area and the monthly mean winds for the area and year).
           The nu values are the magnitude differences of the pairs for all months and phi
           values are the direction differences of the pairs. In the analyses shown in Figures
           20a,b. through 22a,b., Pages 44 through 49, nu values are mapped by the use of open
           circles (meaning the decade winds were smaller than the long-term mean


                                                     41







                                      WESTERLY WIND SPEED
                                                   0-40N) 130E-IOW

                   6-
                w


                N
                           Ic W
                D         CC


                S              W
                p
                E  5-
                E
                D                   'Coc   Ic                       X                 "cc
                                                                         "cc      10C   X

                                                                        cc W         X



                                                  loccc W W cc Ic

                M
                                                           I
                S  4                              1   1   1
                   1890     1900    1910 1920 1930 1940 1950 1960 1970 1980




                                          e
                          Figure 19. Wind Spe d Variation from COADS Analysis over the Period 1890-1979.



                                                      42








            wind in the 2' square area) or by shaded circles (meaning the decade winds were
            larger than the long-term mean wind). Phi values are mapped by the use of open
            circles (meaning the decade wind direction veered (rotated in a clockwise direction)
            relative to the vector averaged wind direction over the COADS record period) or by
            the use of shaded circles (meaning the decade wind direction of the 2* square area
            backed (rotated in a counter-clockwise direction) from the direction of the long-term
            average direction).

                In reference to the annual mean temperatures of the northern hemisphere
            (Figures Ia. and 1b., Page 4) or the annual mean temperature of the United States
            (Figure 4., Page 9), temperature was relatively low in the late 1800's, high in the
            period 1935-1944, and low again in the period 1970-1979. Wind summary periods are
            selected from these periods to intercompare the characteristics of the wind fields in
            these periods and to relate wind circulation to the known coastal and continental
            climate conditions. Figures 20a. and 20b., Pages 44 and 45, are produced from
            COADS monthly wind averages, over 2' square areas, in the years 1889 through
            1899. These maps show the surface winds along the United States coasts to be
            veered from the long-term average and, generally, of larger speed. Figures 21a. and
            21b., Pages 46 and 47, are similar analyses for the period 1970 through 1979. The
            coastal winds are again generally veered from the long-term average and of larger
            speed and these conditions again preceded a warmer climate. Wind characteristics
            from the period 1935-1944 are mapped in Figures 22a. and 22b., Pages 48 and 49.
            During this period the winds are backed from the long term mean direction in United
            States Coastal areas and, generally, lower speed. These circulation conditions
            preceded climate cooling. Veering and turning southward with strengthening of
            coastal winds is predicted by the GFDL 2XC02wind field (Figures 13a,b. and 15a,b.,
            and Appendix C), an expected circulation associated with climate warming based on
            COADS. However, these COADS analyses results show some differences over the
            coastal regions with some veering and backing occurrences in all analysis periods.


                Wind effects on oceanic circulation on the coast of California (ab in'tra
            SOUTHERN CALIFORNIA OCEAN CIRCULATION) are changes of upwelling and
            coastal water temperature. The COADS analysis period 1935-1944 is within the
            coastal water-temperature anomaly recordsshown in Figure 9., Page 24. From these
            records, the result of wind direction backing and decreased speed during the
            1935-1944 period was increased coastal water temperature. Temperature anomaly
            was equal to about +0.23' C over the decade. The records of measured sea-level
            south of San Francisco (Chelton et al., 1982) also support the expectation of less
            upwelling along the coast during the decade (Figure 23., Page 50); sea-level is of the
            highest recorded that indicates weak southward transport of surface water mass.
            Conversely, the expected result of the wind regimes of the late 1800's and 1970's is
            for colder coastal temperature.



                                                     43












                                                                                       Ilk I





                                                                                             14







                         ---------- 300       -------
                                      009 - - - - - - - -

                                  0



                          0














                                  11Q.






               Figure 20a. Wind Speed Anomaly for the Period 1889-1899 Computed from COADS (Shaded Ci
                              Open Circles Indicate Smaller).


                                                                              44













                                                                                let -











                                    00000





                         0



                                                                                           ........ .. ...

















              Figure 20b. Wind Direction Anomaly for the Period 1889-1899 Computed from COADS (Sha@
                            Backing, Open Circles Indicate Veering).


                                                                     45











                                                                      N
                                                                          x,@


































            @4 X4.

              "if







             Figure 21a. Wind Speed Anomaly for the Period 1935-1944 Computed from COADS (Shaded
                          Larger, Open Circles Indicate Smaller).


                                                                 46













                                                                                      17

                                                                IPA


























             Figure 21b. Wind Direction Anomaly for the Period 1935-1944 Computed from COADS (Sha
                          Bacldng, Open Circles Indicate Veering).


                                                                 47




































            Figure 22a. Wind Speed Anomaly for the Period 1970-1979 Computed from COADS (Shaded
                          Larger, Open Circles Indicate Smaller).


                                                               48























































                           111v
                                                                                       "VV
























            Figure 22b. Wind Direction Anomaly for the Period 1970-1979 Computed from COADS (Sha
                          Backing, Open Circles Indicate Veering).


                                                                49












                         5[-

                       E

                                             1910                 1920
                       Uj
                       > 0

                                                                                      1930                1940
                       w



                         5




                         5


                       E


                       -j
                       Ui                   1950                                       970A
                                                                       x

                                                                1960                                     1960
                                                                         V
                                               10



















            Figure 23. Monthly Average Sea Level Observed on the Coast South of San Francisco During the Period 1900-1979
                        (Chelton et al., 1982


                                                                 50








            3.2 Differences in Ocean-Area Atmospheric Pressure for Selected Periods

                The Hadley Circulation produces changes of air mass convergence in the vicinity
            of the Atlantic and Pacific areas subtropical anticyclones. These changes are
            monitored by the magnitude of surface air pressures and by latitudinal gradients of
            pressure variation. Monthly averaged pressures were compared to long-term mean
            of pressures for each 2' area and these pressure anomalies were averaged over all
            months and along 20 bands of latitude. Latitude bands in the Pacific area were from
            160* W eastward to 130* W and in the Atlantic area 50' W to 20' W. Pressure data
            were analyzed for the same periods as the wind summaries i.e. 1889-1899, 1935-1944,
            and 1970-1979 and these anomaly averages are listed in Table 1, Page 53. Since
            anomalies were computed by subtracting measured pressure from long-term mean,
            negative anomalies indicate pressure of the latitude band increased during the
            summary period. Data in the table shows pressure increased over most of the
            latitude bands in the Pacific area coincidentally with wind veering and increasing in
            speed in the California coastal area. Similar changes occur in pressure and wind
            regime with seasonal changes from winter to summer. Conversely, wind bacldng and
            reduced wind speed occurred during the period 1935-1944, a period with decreased
            pressure over the Pacific latitude bands. Latitudinally averaged anomalies in the
            Atlantic area do not have a distinctive pattern of change from one summary period
            to another.

               In geographic regions of Hadley Circulation, the regions of pressure difference are
            also regions with cloud-cover differences; areas of low pressure and ascending air-
            mass are areas with greater cloud amount than areas of high pressure and
            subsidence. Cloud amounts from the period of climate cooling were compared with
            cloud amounts from the 1970's period with climate warming to see if changes in
            Hadley Circulation intensity were causes of observed wind and pressure variations.
            No distinctive regional differences were discerned but cloud amount in both the
            Atlantic and Pacific areas was larger in the analysis period from the 1970's compared
            to the analysis period 1935-44. The cloud differences in our analysis regions appear
            to be caused by southward extension of cloud associated with meteorological systems
            in the mid-latitude westerlies. These results do not support the hypothesis that
            variation of Hadley Circulation intensity was a cause of the observed wind and
            pressure differences. COADS cloud data were compared to cloud amount derived from
            satellite data by Godshall (1968) from the period 1962-1973 in the eastern Pacific.
            Correlation between the cloud amounts was poor with correlation coefficient of 0.53
            which suggests the COADS cloud data may be of poor quality.








                                                      51








               Averaged surface pressure maps for January months in cold years, with change-to-
             warmer climate (Figures 24a. and 26a., Pages 54-59), show Pacific subtropical
             anticyclones were positioned further north and west of normal positions of the
             anticyclone (Figure A-1.). Cold years anticyclones in the Atlantic were also positioned
             further north but all anticyclone displacements are small, on the order of a few
             degrees of latitude. During August cold years periods, there is less difference in the
             geographic position of the anticyclones (Figures 24b., 26b. and Figure A-2.). Eastern
             Pacific anticyclones during the 1935-44 period of climate cooling (Figure 25a,b.) are
             displaced toward the southeast. Changes in the Atlantic area are less than those in
             the Pacific.    Although there are few of the pressure analyses periods for
             intercomparison, the period differences are evidence for a change in the general
             atmospheric circulation that includes change in the Aleutians Low and local climate
             variation identified by others (Namias, 1980 and 1985, Cayan and Peterson, 1989).

































                                                       52














             Table 1 Pressure Anomaly Averaged Over Two-Degree Latitude Bands
                    From the Periods, 1889-1899, 1935-1944, and 1970-1979
                    in Pacific (1601-130'W) and Atlantic (501-20'W) Areas.

                                         PACIFIC AREA                       ATLANTIC AREA
                                           Anomaly (mb)                        Anomaly (mb)
               NORTH (ON)          1889-99 1935-44 1970-79               1889-99 1935-44 1970-79
               LATITUDES


                     1               -0.61      -0.17     -0.11          0.48      0.11        0.09
                     3               -0.74      -0.73     -0.09          0.24      0.12        0.08
                     5               -0.93      -0.35     -0.18          -0.13     0.12        0.03
                     7               -1.48      -0.09     -0.21          -0.09     0.09        0.04
                     9               -0.86      0.01      -0.15          0.02      0.18        0.09
                    11               0.54       0.09      -0.17          -0.12     -0.04       0.05
                    13               -0.03      0.31      -0.21          0.10      0.08        0.07
                    15               -0.59      0.41      -0.22          0.18      -0.04       0.05
                    17               0.75       0.20      -0.22          0.12      0.10       -0.01
                    19               0.81       0.01      -0.22          0.01      0.07       -0.04
                    21               0.14       0.34      -0.24          0.25      0.01       -0.04
                    23               0.00       0.39      -0.30          0.39      0.00       -0.02
                    25               -0.12      0.44      -0.23          0.55      -0.17       0.01
                    27               -0.66      0.37      -0.13          0.69      0.11       -0.02
                    29               -0.43      0.55      -0.20          0.25      0.21       -0.09
                    31               -1.76      0.71      -0.36          0.91      0.39       -0.03
                    33               -1.41      0.95      -0.35          0.84      0.25       -0.09
                    35               -1.72      0.72      -0.28          1.86      0.51       -0.06
                    37               -1.49      0.92      -0.52          3.01      0.34       -0.12
                    39               0.00       0.99      -0.71          -0.11     0.76       -0.29














                                                        53













                                                                                          N,



















                                                                                         0
                                                               0                          0
                                                                8                        0                             ;.OC8
                                                                                          0
                                                                                           8







                  LEGEND:
                0 < 1016mb                                                                                         t
                Q > 1015mb  and < 1021mb
                  > 1020mb



                 Figure 24a. January Average Atmospheric Surface Pressure for the Period 1889-1899 Compu


                                                                                  54


































                                0                      G&QM QQO 9
                0.          00 00                                 0

                        C,                                    Gio
                                                                 16@
                                                 00              Q
                                                                9  8
                                                                 c8 8
                                                                     0


                                                                 0
                                                                               0
                                                                                C8D
                                                                           oc)@'




               LEGEND:


              0 <1016mb
              j& >1015mb and < 1021mb
              9 >1020mb



              Figure 24b. August Average Atmospheric Surface Pressure for the Period 1889-1899 Comput

                                                                  55























                                                                        All













                                        OCOO               COCO





                                                                                  0
                                                                                    0
                                c8
                                                     0               00
                                           jo            88
                                 0






                   LEGEND:


                 0 < 1016mb
                 Q)> 1015mb and < 1021mb
                 0 > 1020mb




                 Figure 25a. January Average Atmospheric Surface Pressure for the Period 1935-1944 Compu


                                                                                 56


























                                                                                                       ,@41












                                                                                         0
                                    C800C)
                                     0 C80







                    LEGEND:


                 0 < 1016mb
                 Q > 1015mb  and < 1021mb
                 0 > 1020mb
                                                                                                                 O,i,































                 Figure 25b. August Average Atmospheric Surface Pressure for the Period 1935-1944 Compu


                                                                                    57
















                                                                                                 let -


                                                        Ap"                                                             11@6











                                                                                              PO
                                   0






                   LEGEND:
                                                                                                                       t
                 0 < 1016mb
                 Q > 1015mb and < 1021mb
                 0 > 1020mb



                 Figure 26a. January Average Atmospheric Surface Pressure for the Period 1970-1979 Compu


                                                                                    58





















                                                                                                                                                               Q



















                                                                               - - - - - - - - - - - - - -



                                                                  0


                                                     0
                                                                                 0
                                                                                                                              GOOD             0
                                                       0                  0                     0

                   77@
                                      A

                           LEGEND:


                        0  < 1016mb
                        Q  > 1015Mb      and      1021mb
                        0  > 1020mb




                         Figure 26b. August Average Atmospheric Surface Pressure for the Period 1970-1979 Compu


                                                                                                                         59









             4. CONCLUSIONS

                 A basis for assessment of environmental impact in coastal waters is the expected
             change in the parameters sea-level, runoff and wind effects as a consequence of
             Global Climate Change. Changes in these selected parameters serve as analogues
             of change that may occur within the regions for analysis. The analogue approach for
             analysis of possible climate change effects could be usefully applied to coastal regions
             not included in this study.


             Mid-Atlantic Bight:

                 Coastal bays of the Mid-Atlantic Bight are expected to be most affected by river
             runoff changes and water level change. The seaward salinity gradient, the near-shore
             stratification, and dynamic circulation of the Bight are also dependent on the amount
             of fresh water flowing onto the coast. The effect of wind forces on the marine
             circulation is variable over the Bight because of differences in coastal geography
             relative to prevailing wind. Southerly wind components in summer promote
             upwelling along the coast and resultant wind stress opposes the dynamic water mass
             near-shore circulation. These effects retard southward flow of chemical enriched
             waters of the New York area and promote circulation of oxygen-rich "cold pool"
             waters shoreward. A GCM simulated warmer-climate-predicted veering and
             southward turning of surface seasonal winds in winter is supported by the decade
             wind analyses of COADS for the area. The COADS wind veering and strengthening
             during periods of transition to warmer climate indicates lengthening of the summer-
             seasonal circulation period is possible. The effect of this wind forced circulation
             during cold seasons will be warmer coastal waters that are more saline.

                 The veered and more southerly wind regimes over southern coastal areas may
             advect larger amounts of moisture into the U.S. interior that could lead to larger
             amounts of precipitation and land runoff. However changes in evapotranspiration
             could have important influences. Gauged flow rate of the Potomac River (Figure 6.)
             indicates that enhanced river flow rate did occur during the decade of the COADS
             wind analyses in the 1970's and possibly during the 1890's (both periods of transition
             from "cold" to "warmer" climate).

                 Steric effects (water density changes) are predicted to cause continued rise in sea
             level along the east coast. Wind-forced circulation will lead to lowering of water level
             but varied geography of the Bight will cause considerable local differences in this
             effect.







                                                        60









            Gulf of Mexico:

                The summer, seasonal shelf circulation of the northwestern Gulf of Mexico is
            driven by southerly wind. Wind changes that extend the season of these winds will
            promoti, a longer annual period for existence of the anticyclonic gyre on the shelf
            However, it is unclear that an extension of the Bermuda High predicted for a period
            in transition toward a warmer climate will promote prevailing southerly wind
            regimes of the western Gulf. In the past, large areas near-west of the Mississippi
            Delta were impacted by anoxic bottom water events. These conditions occurred with
            wind-forced spreading of low saline surface-layer water mass that increased the
            coastal -area stratification. Northeasterly winds for these events came from an
            apparent extension of the Bermuda High/belt-of- subtropical- anticyclones over the
            southern states.

                Given the extent of the Mississippi river watershed, which occupies a large
            portion of the United States, one might expect to find trends in the river discharge
            attributable to changes in mid-continental precipitation. However, the records of
            runoff rate of the Mississippi/Atchafalaya Rivers for this study do not substantiate
            an association of runoff rate change with transitions from "cool" to "warm" periods.

                Sea level rise is expected to continue in the northwestern Gulf area because of
            land subsidence. Loss of coastal wet lands and barrier islands is expected to continue
            and result in continued change of coastal bays and salt marsh environment (Edward
            Mima, National Maxine Fisheries Service, personal communication).

            Southern California:

                In sou them California coastal area, winter wind veering may have some influence
            on the Davidson Current but, of greater possible significance is northerly wind speed
            increases that will increase upwelling along the coast. From past records, increased
            upwelling has produced cooler coastal water temperatures and increased coastal
            salinity levels. Increased northerly wind will cause increases in the California
            Current and advection of northern California coastal water may have some effect on
            water quality but this is not assessed in this report.

                The COADS wind analysis of the decade of the 1970's was interpreted as a decade
            with charELCtexistics that are associated with trends toward a warmer climate. A
            longer period, 1946-88 (Bakun ob. cite) was also identified as a period with increased
            northerly wind and coastal upwelling even though the longer period included part of
            a period of Northern Hemispheric cooling.






                                                      61







                The negative correlation of river runoff in southern California coastal areas with
            atmospheric pressure reported by Cayan and Peterson (ob. cite) is expected from the
            analyses of COADS pressure data. From COADS analyses lower pressures were
            found to be characteristic of the decade from 1935 and runoff rate of the Arroyo Saco
            River is much greater during this same period (Figure 10., Page 26). Namias (1980)
            found the dry 1975-76 period to be associated with lower pressures in the north
            central Pacific with higher pressures on the southern California coast. The flow rate
            in the Arroyo Seco is very low in this period (Figure 10., Page 26). COADS analysis
            of pressure found the coastal area pressures to be higher than the long-term mean
            in the periods in transition toward warmer climate. Therefore, precipitation and
            runoff from rivers in southern California are expected to decrease in the future as
            climatic warming continues. Although runoff carries plant nutrients into coastal
            water masses, decreased runoff may not be important for coastal productivity because
            the expected increase of northerly wind will promote advection of nutrients into the
            surface layers from upwelling.

                Dynamic causes of sea-level change are important factors for change along the
            coast. Coriolis acceleration of coastal water masses causes lowered coastal sea level;
            seasonal change of southward directed wind stress produces an annual sea level
            change of about 20cm. This amount varies along the coast because of local geography
            and bottom topography of the shelf. Changes in sea level are expected to have little
            impact along the coast because of the general steepness of coastal terrain.


            The Next Steps:

                Changes in the atmospheric coastal circulation of the past substantiate prediction
            of circulation change by the GFDL General Climatic Model. The atmospheric
            circulation is used to infer a coastal-marine circulation along the open coast that is
            expected in an environment with atmosphere doubled carbon dioxide concentration.
            This coastal circulation and other parameters of future climate are expected to cause
            environmental change in coastal bights and bays that are different from change on
            the open coast. Large estuarine areas, such as Chesapeake Bay, are expected to have
            large hydrographic change caused by coastal circulation and sea level and also
            precipitationYriver runoff. These areas are important for assessment of coastal
            responses because the estuaries are, in many aspects, a focus of concern regarding
            coastal resource exploitation. Operation of the numerical models of runoff and
            circulation for the Cheasapeake Bay from scenarios of changes in climate will permit
            evaluation of potential impact in the context of changes related to larger scale
            processes.








                                                      62








            5. REFERENCES and BEBLIOGRAPHY M


            Bakun, A. 1990. Global Climate Change and Intensification of Coastal Upwelling,
                Science, Vol. 247, p198-201.

            Barnola, J.M., Raynaud, D., Korotkevitch and C. Lorius. 1987. Vostok Ice Core:
                A 160000 year Record of Atmospheric C02. Nature, Volume 329, p408-414.

            Bigelow, F.H. 1909. Report on the Temperatures and Vapor Tensions of the
                United States, Bulletin S, U.S. Department of Agriculture, Weather Bureau,
                Washington, DC, 302pp.

            Bjerknes, J. 1960. Ocean Temperatures and Atmospheric Circulation, WMO
                Bulletin Vol. 9, No. 3, p151-157.

            Bjerknes, J. 1969. Atmospheric Teleconnections from the Equatorial Pacific,
                Monthly Weather Review, Vol. 97, p163-172.

            Blaha, J.P. 1984. Fluctuations of Monthly Sea Level as Related to the Intensity of
                the Gulf Stream from Key West to Norfolk, Journal of Geophysical Research Vol.
                89, p9547-9552.

            Bolin, B. 1950. On the Influence of the Earth's Orography on the General
                Characteristics of the Westerlies, Tellus, Vol. 2, No. 3, p184-195.

            Bryan, K., F.G. Komro, S. Manabe, and M.J. Spelman. 1982. Transient Climate
                Response to Increasing Atmospheric Carbon Dioidde, Science, Vol. 215, p56-58

            Bumpus, D.F., R.E. Lynde and D.M. Shaw. 1973. Physical Oceanography, in
                "Coastal and Offshore Environmental Inventory, Cape Hatteras to Nantucket
                Shoals", Saul B. Saila (ed), Marine Publication Series No. 2, University of Rhode
                Island, Kingston, RI, pl-1 - 1-72.

            Bumpus, D.F. 1973. A Description of the Circulation on the Continental Shelf of
                the East Coast of the United States, in "Progress in Oceanography", B.A. Warren
                (ed), Vol. 6, Pergamon Press, New York, NY, p111-157.

              Bunker, A.F. 1980. Mon. Weather Review, Vol. 108. p720-723.

            California State Water Quality Control Board. 1965. An Oceanographic and
                Biological Survey of the Southern California Mainland Shelf, R.E. Stevenson
                (contributor for oceanography and climatology), Publication No. 27, The Allen
                Hancock Foundation, University of Southern California, p29-122.


                                                     63








            Caynan, D.R. and D.H. Peterson. 1989. The Influence of North Pacific
                Atmospheric Circulation on Strearnflow in the West, in "Aspects of Climate
                Variability in the Pacific and the Western Americas", D.H. Peterson (ed.),
                Geophysical Monograph 55, American Geophysical Union, p375-397.

            Chelton,D.B., P.A. Bernal, and J.A. McGowan. 1982. Large-scale Interannual
                Interactions, Journal of Marine Research, Vol. 40, No. 4, p1095-1125.

            Cochrane J.D., F.J. Kelly. 1986. Low-Frequency Circulation on the Texas-Louisiana
                Continental Shelf, Journal of Geophysical Research, Vol. 91, No. C9,
                p10645-10659.

            Conseil Permanent International Pour I'Exploration de la Mer. 1936. Bulletin
                Hydrographique Pour IAnnee 1935 (Series B1), 105p.

            Cubasch, U. 1985. The Mean Response of the ECMWF Global Model to the
                Composite El Nino Anomaly in Extended Range Prediction Experiments, Chapter
                23, "Coupled Ocean-Atmosphere Models", J.C.J. Nihoul (ed), Elsevier
                Oceanography Series 40, New York, NY.

            DeMarris, G.A., G.C. Holzworth and C.D. Hosler. 1965. Meteorological Summaries
                Pertinent to Atmospheric Transport and Dispersion Over Southern California,
                U.S. Weather Bureau Technical Paper No. 54, Washington, DC.

            Dickson, R.R. and R.E. Livezey. 1984. On the Contribution of Major Warming
                Episodes in the Tropical East Pacific to a Useful Prognostic Relationship Based
                on the Southern Oscillation, Journal of Climate and Applied Meteorology, Vol. 23,
                p194-200.

            DOE. 1979. Carbon Dioxide Effects Research and Assessment Program: Workshop
                on the Global Effects of Carbon Dioxide from Fossil Fuels, Miami Beach, FL,
                W.P. Elliot and L. Machta (eds), CONF-770385, Washington, DC, 122pp.

            Eppley, R.W. 1972. Temperature and Phytoplankton Growth in the Sea, Fisheries
                Bulletin, Vol. 70, p1063-1085.

            Finkelstein, P.L. and L.E. Truppi. 1990. On The Spatial Distribution of
                Precipitation Seasonality,EPA Research Triangle Park, NC, unpublished, 47pp.

            Flint, R.F.F. 1957. Glacial and Pleistocene Geology, John Wiley and Sons, Inc.
                 (publishers), New York, NY, Chapt. 27.

            Forry, S. 1843. Meteorology, The New World, J. Winchester Pub., New York, NY,
                p29.


                                                     64








            Fritts, H.C. 1980. Statistical Climate Reconstructions from Tree-Ring Widths, in
                "Climatic Variations and Variability: Facts and Theories", A. Berger (ed), D.
                Reidel Pub., London, p135-153.

            Gates, G., P. Roundtree and Q. Zeng. 1990. Validation of Climate Models, draft
                report "Scientific Assessment of Climate Change" of IPCC, unpublished.

            Giorgi, F. and L.O. Meams. 1991. Approaches to the Simulation of Regional
               Climate Change: A Review, Reviews of Geophysics, Vol. 29, No. 2, p191-216.

            Godshall, F.A. 1968. Intertropical Convergence Zone and Mean Cloud Amount in
                the Tropical Pacific Ocean, Monthly weather Review, Vol. 96, No. 3, p172-175.

            Godshall, F.A. and R.G. Williams (eds). 1981. A Climatology and Oceanographic
                Analysis of the California Pacific Outer Continental Shelf Region, U.S. Dept. of
                Commerce, Washington, DC, 250pp.

            Godshall, F.A. 1990. Couples of Atmospheric and Oceanic Circulation Processes
                Relevant to Climate of the United States. unpublished.

            Grassl, H. 1988. Numerical Models of Climate, in "Long and Short Term
                Variability of Climate", H. Wanner and U. Siegenthaler (eds), Lecture Notes in
                Earth Sciences 16, Springer-Verlag, New York, NY, p126.

            Hamilton, W. 1965. Cenozoic Climate Change and Its Cause, In Metero. Monog.,
                vol. 8, ed. J.M. Mitchell, Jr. Boston, Mass: American Meteorological Society.

            Hansen, J., D. Johnson, A. Lacis, S. Lebedeff, P. Lee, R. Ruedy and L. Travis.
                1983. Efficient Three-Dimensional Global Models for Climate Studies: Models I
                and II, Monthly Weather Review, Vol. 111, p609-622.

            Hansen, J. and S. Lebedeff. 1987. Gobal Trends of Surface Air Temperature,
               Journal of Geophysical Research, Vol. 92, No. D11, p13,345-13,372.

            Hanson, K, G.A. Maul, and T.R. Karl. 1989. Are Atmospheric "Greenhouse"
                Effects Apparent in the Climatic Record of Continental U.S. (1896-1987),
                Geophysical Research Letters, Vol. 16, No. 1, p49-52.

            Harper, D.E.Jr., L.D. McKinney, R.R. Salzer, and R.J. Case. 1981. The Occurrence
                of Hypoxic Bottom Water off the Upper Texas Coast and its Effect on the Benthic
                Biota, Contributions in Marine Science Vol. 24, p53-79.

            Harris, D.L. 1979. Tide and Tidal Datums for the United States, U.S. Army
                Coastal Engineering Research Center Report.


                                                     65








            Hartmann, D.L. 1984. On the Role of Global-Scale Waves in Ice-
                Albedo and Vegetation-Albedo Feedback, "Climate Processes and Climate
                Sensitivity", J.E. Hansen and T.Takahashi (eds), American Geophysical Union,
                Washington, DC p18-28.

            Hicks, D.C. and J.R. Miller. 1980. Meteorological Forcing and Bottom Water
                Movement off the Northern New Jersey Coast, Estuarine, Coastal and Marine
                Science, Vol. II, p563-571.

            Hicks, S.D., H.A. Debaugh, Jr., and L.E. Hickman, Jr. 1983. Sea Level Variations
                for the United States 1855-1980, NOAA, Rockville, AM, 170pp.

            Hopkins, T.S. and N. Garfield 111. 1979. The Existence of the Maine Intermediate
                Water, in "A Summary of an Informal Workshop of Physical Oceanography of the
                Gulf of Maine and Adjacent Seas, Woods Hole Oceanographic Institute, 28p.

            Huang, J.C.K. 1972. Recent Decadal Variation in the California Current System,
                J. Phy. Oceanography, Vol.2, No. 10, p382-390.

             Hickey, B.M. 1979. The California Current System-Hypotheses and Facts, Prog.
                Oceanogr., Vol 8, p 191

            Izrael, Yu.A. and M. Hashimoto, 1990. Potential Impacts of Climate Change, Report
              prepared for the Intergovernmental Panal on Climate Change (IPCC), World
              Meteorology Organization, United Nations, New York, NY, Chapters 1,2, and 3.

            Jenne, R.L. 1989. Data from Climate Models: theC02Warming, National Center
                for Atmospheric Research, unpublished draft, National Center for Atmospheric
                Research, Boulder, CO, 33pp.

            Jones, P.D., T.M.L. Wigley and P.M. Kelly. 1982. Variations in Surface Air
                Temperatures: Part 1. Northern Hemisphere, 1881-1980, Monthly Weather
                Review, Vol. 110, No. 2, p59-71.

            Karl, T.R. and W.E. Riebsame. 1989. The Impact of Decadal Fluctuations in Mean
                Precipitation' and Temperature on Runoff. A Sensitivity Study Over the United
                States, Climate Change, Vol. 15, Kluwer Academic Pubs., Netherlands, p423-447.

            Keigwin, L.D., Jr. 1976. Late Cenozoic Planktonic Foraminiferal Biostratigraphy
                and Paleoceanography of the Panama Basin, Micropaleo. 22: 419-42.






                                                    66








           King, G.A., R.L. DeVelice, R.P. Neilson, and R.C. Worrest. 1990. Potential Impacts
               of Climatic Warming on California, Smith, J.B. and D.A. Tirpak (eds), The
               Potential Effects of Global Climate Change on the United States, Hemisphere
               Publishing Corp., New York, NY, Chapter 4.

           Kolpack, R.L. (ed). 1971. Biological and Oceanographical Survey of the Santa
               Barbara Channel Oil Spill, 1969-1970, Vol. II, Allan Hancock Foundation,
               University of Southern California, p7-180.

           Koppen, W. 1899. Grundlinien der Maritimen Meteorologie, G.W. Niemeyer
               Nachfolger (publishers), Hamburg, Germany.

           Kres,C.J. 1972. Ecology, Harper & Row, Pub., New York, NY, p452-453.

           Kutzbach, J.E. and B.L. Otto-Bliesner. 1982. The Sensitivity of the African-Asian
               Monsoonal Climate to Orbital Parameter Changes for 9000 Years BP in a
               Low-Resolution General Circulation Model, Journal of Atmospheric Science, Vol
               39, p1177-1188.

           La Brecque, M. 1989. Detecting Climate Change I, MOSAIC, Vol. 20, No. 4,
              NASA, p2-9.

           Lamb, H.H. and A.J. Johnson. 1959. Climatic Variation and Observed Changes in
               the General Circulation, Geografiska Annaler, Vol. 41, No. 2-3, p94-134.

           Latham, J. and M.H. Smith. 1990. Effect on Global Warming of Wind-Dependent
               Aerosol Generation at the Ocean Surface, Nature, Vol. 347, Sept. 27, p372-373.

           Lau, N. and A.H. Oort. 1985. Response of a GFDL General Circulation Model to
               SST Fluctuations Observed in the Tropical Pacific Ocean During the Period
               1962-1976, Chapter 21, "Coupled Ocean-Atmosphere Models", J.C.J. Nihoul (ed),
               Elsevier Oceanography Series 40, Elsevier, New York, NY

           Leipper, D.F. 1955. Sea Temperature Variations Associated with Tidal Currents
               in Shallow Water Over an Irregular Bottom, Journal of Marine Research, Vol. 14,
               p234-252.

           Lutgens, F. 1989. The Atmosphere:an Introduction to Meteorology, Prentice Hall,
              Inc., p192.

           Manabe, S. and R.T. Wetherald. 1980. On the Distribution of Climate Change
               Resulting From Increase of C02 Content of the Atmosphere, Journal of
               Atmospheric Science, Vol. 37, p99-118.



                                                    67








            Manabe, S. and R.T. Wetherald. 1987. Large-Scale Changes in Soil Wetness Induced
                by an Increase in Carbon Dioxide, Journal of Atmospheric Science, Vol. 44,
                p1211-1235.

             Marner, H.A. 1951. Tidal Datum Planes, Special Publication No. 135, U.S.
                Coast and Geodetic Survey, Washington, DC.

            Meade, R.H. and K.O. Emery. 1972. Sea level as affected by River Runoff, Eastern
                United States, Science, Vol. 173, p425-428.

            Montgomery, R.B. 1938. Fluctuations in Monthly Sea Level on Eastern U.S. Coast
                as Related to Dynamics of Western North Atlantic Ocean, Journal of Marine
                Research, Vol. 1, p165-185.

            Namias, J. 1960. The Meteorological Picture, 1957-1958: CaICOFI Reports, Vol.
                VII, California Dept. of Fish and Game, p31-42.

            Namias, J. 1980. Case Studies of Exceptional Climate in United States 1975-1979
                and Air-Sea Interactions, (in) "Climatic Variations and Variability: Facts and
                Theories", A. Berger (ed), D.Reidel Publishing Company, Boston, MA, p369-398.

            Namias, J. 1985. New Evidence for Relationships Between North Pacific
                Atmospheric Circulation and El Nino, Tropical Ocean-Atmosphere News-letter,
                March 1985, p2-3.

             Nelson, C.S. 1977. Wind Stress and Wind Stress Curl Over the California
                Current, NOAA Tech. Rept. NMFS-SSRF-714.

            NOAA. 1985. Comprehensive Ocean-Atmosphere Data Set, Release 1, Boulder,
                CO, 216pp.

            NOAA GFDL. 1988. Q-Flux General Climate Model, Unpublished Communication
                With EPA.

            Nuzzi, R. 1971 The Oceanography of the New York Bight: Physical, Chemical,
                Biological, Tech. Rpt. 0017, New York Ocean Science Laboratory, Vol.I, 108pp.

            Nyberg, A. 1951. On the Variation of the General Circulation of the Atmosphere
                During Past Ages, in "Reports of the Swedish Deep-Sea Expedition 1947-1948",
                Vol. V/II, Royal Society of Arts and Sciences, Goteborg, 1951-1960.

            Officer, C.B., R.B. Biggs, J.L.Taft, L.E. Cronin, M.A. Tayler, and W.R. Boynton,
                1984. Chesapeake Bay Anoxia: Origin, Development, and Significance, Science,
                Vol 223, p22-27.


                                                   68








            Olausson, E. and C. Ingernar (eds). 1980. Chemistry and Biogeochernistry of
                Estuaries, John Wiley & Sons, New York, NY, 452pp.

            Pfister,C. 1988. Variations in the Spring-Siimmer Climate of Central Europe from
                the High Middle Ages to 1850, Lecture Notes in Earth Sciences, p57-82.

            Palutikof, J.P., T.M.L. Wigley, and J.M. Lough. 1984. Seasonal Scenarios for
                Europe and North America in a High-C02 Warmer World, U.S. Department of
                Energy, C02 Research Division Technical Report TRO12, Washington, DC, 70pp.

            Reid, J.L. and A.W. Mantyla. 1976. The Effect of Geostrophic Flow Upon Coastal
                Sea Level Variations in the Pacific Ocean, J. Geophys. Res., Vol. 81, p3100-3110.

            Reiter, E.R. 1963. Jet-Stream Meteorology, English Translation, The University of
                Chicago Press, Chicago, Chapter 7.

            Segar, D.A. and G.A. Berberian. 1976. Oxygen Depletion in the New York Bight
                Apex: Causes and Consequences, (in) Middle Atlantic Continental Shelf and the
                New York Bight, M.G. Gross (ed), American Society of Liminology and
                Oceanography, Special Symposium 2, p220-239.

            Siegenthaler, U. 1988. Causes and Effects of Natural C02 Variations During the
                Glacial-Interglacial Cycles, Lecture Notes in Earth Sciences 16, p166.

            Signal Office. 1891. Charts Showing the Isobars, Isotherms, and Winds in the
                United States for Each Month from January 1871 to December 1873, U.S. War
                Dept., Washington DC.

            Shngo, A. and D.W. Pearson. 1987. A Comparison of the Impact of an Envelope
                Orography and of a Parameterization of Orographic Gravity Wave Drag on Model
                Simulations, Quarterly Journal of the Royal Meteorological Society, Vol. 113,
                p847-870.

            Smith, R.A. and R.J. Leffler. 1980. Water Level Variations Along California Coast,
                Journal of the Waterway, Port, Coastal and Ocean Division, Proceedings of the
                American Society of Civil Engineers, Vol. 106, No. WW3, p335-348.

            Snedden, J.W., Dag Nummedal and A.F. Amos. 1988. Storm and Fair Weather
                Combined Flow on the Central Texas Continental Shelf, Journal of Sedimentary
                Petrology, Vol. 58, No. 4, p580-595.

            Stewart, H.B. Jr. 1960. Coastal Water Temperature and Sea Level, California to
                Alaska: CalCOFI Reports, Vol. VII, California Dept. of Fish and Game, p97-102.



                                                     69








            Stoddard, A., 1989: Eutrophication and Recurrent Hypoxia in the New York Bight:
                 A Synthesis of Historical Data and a Numerical Model of the 1976 Anoxic Event,
                 in "Estuarine and Coastal Modeling", M.L. Spaulding (ed), American Society of
                 Civil Engineers, New York, NY, p361-370.

            Swanson, R.L. and C.J. Sindermann (eds). 1979. Oxygen Depletion and
                 Associated Benthic Mortalities in New York Bight, 1976, NOAA Professional
                 Paper 11, National Oceanic and Atmospheric Admin., Washington, DC, 345pp.

            Thompson, K.R. 1990. North Atlantic Sea Level and Circulation, in "Studies in
                 Geophysics", R.R. Revelle (chairman), National Academy Press, Washington, DC,
                 p52-62.

            Tibby, R.B. and R.D. Terry. 1959. Physical and Chemical Characteristics of the
                 Water Over the Southern California Shelf, 1959-1958, in "Oceanographic Survey
                 of the Continental Shelf Area of Southern California", Allan Hancock Foundation,
                 University of Southern California, p119.

            Tollner, H. 1956. Die Meteorologisch-lGimatischen Ursachen der
                 Gletscherschwankungen in der Ostatpen Wahrend der Letzten Zwei
                 Jahrhunderte (the meteorological climate under the general circulation of the
                 westerlies within the last 200 years), Mitt. der Geogr. Ges. 96 (1-4) p31-74.

            U.S. Weather Bureau. 1930. Monthly Weather Review, Supplement No. 34, The
                 Daily, Monthly and Annual Normals of Precipitation in the United States, Based
                 on the 50-year Period 1878 to 1927 Inclusive, P.C. Day (ed), Washington, DC ,
                 101pp.

            Wallace, J.M. and D.S. Gutzler. 1981. Teleconnections in the Geopotential Height
                 Field During the Northern Hemisphere Winter, Monthly Weather Review,
                 Vol.109, p784-812.

            Washington, W.M. and G.A. Meehl. 1983. General Circulation Model Experiments
                 on the Climatic Effects Due to a Doubling and Quadrupling of Carbon Dioxide
                 Concentration, Journal of Geophysical Research, Vol. 88, No. C11, p6600-6610.

            Washington, W.M. and G.A. Meehl. 1984. Seasonal Cycle Experiment on the
                 Climate Sensitivity Due to a Doubling of C02 With an Atmospheric General
                 Circulation Model Coupled to a Simple Mixed-Layer Ocean Model, Journal of
                 Geophysical Research, Vol. 89, No. D6, p9475-9503.

            Wetherald, R.T. and S. Manabe. 1986. An Investigation of Cloud Cover Change in
               Response to Thermal Forcing, Climate Change, Vol. 8, p5-24.



                                                     70








            Wigley, T.M.L., P.D. Jones, and P.M. Kelly. 1986. Empirical Climate Studies,
              in "The Greenhouse Effect, Climatic Change, and Ecosystems, SCOPE 29, B. Bolin
              et al. (eds), John Wiley and Sons, NY, p271-322.

            Williams, R.G. and F.A. Godshall. 1977. Summarization and Interpretation of
               Historical Physical Oceanographic and Meteorological Information for the
               Mid-Atlantic Region, NOAA U.S. Dept. of Commerce, Washington, DC, 295pp.

            * Whysall, K.D.B., N.S. Cooper and G.R. Bigg. 1987. Nature, Vol. 327, p216.

            * Wyllie, J.G. 1966. Geostrophic Flow of the California Current at the Surface and
               at 200m, California Cooperative Oceanic Fisheries Investigation Atlas No.4,
               288pp.


































                                                   71




 I
 I                            APPENDICKA
 I          MONTHLY MEANS OF ATMOSPHERIC SURFACE PRESSURE
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 1                               72
 1















                  MONTHLY MEANS OF ATMOSPHERIC SURFACE PRESSURE




           INTRODUCTION:

               The months of August and January were chosen to represent the summer and
           winter seasons of the northern hemisphere for the longitudes of the United States
           because the mean position of the Intertropical Convergence (ITC) zone (Godshall,
           1968) is furthest north in August and at the southern most position in January.
           COADS (NOAA, 1985) 20 square average surface data in the period from 1854 to 1979
           were averaged to produce the January surface analysis (Figure A-1.) and August
           analysis (Figure A-2.).




























                                                  73






































                                                                                                                                                                                                -101
                              02                          -lod,004
                                                                                   1008
                                                                            loto
                                                                                                                         22
                                                                  jG12
                                                                                                ol                             22
                                                         &14                        14                                                                          20
                                                                                                                         tole                                                      Iola
                                                                                                                 1016
                                                                                                                                                                                                 1016
                                                                                                                        1034
                                      1010                                                                               1012


                                                                                       08
                                        a                       Im            Ooe












                                                                         Figure A-1. January Mean Atmospheric Surface Air pressure.


                                                                                                                                74

































                                                                                                                                                                                                  -Jv


                     :2                                                                         d2             jD24
                                                                               6P20
                                                                                                            j4)22
                                                                                                        j020

                                                                                                 t&               10-6                                       1014
                                                                                                                                                                       ioli
                                                                              1014                              1014
                                                                            ID12                               1012                                                            1012       101,j


                                                                                 1010
                                                                          ffo                               010
                                                               P010,;,










                                                                         Figure A-2. August Mean Atmospheric Surface Air Pressure.


                                                                                                                                75




 I
 I                            APPENDEKB
 I                     VECTOR ANALYSIS OF WINDS
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 1                                76
 1









                                                APPENDEKB


                                        WIND VECTOR ANALYSIS



            INTRODUCTION:

                Wind observations are usually reported as vector quantities with magnitude and
            direction. This analysis procedure uses sets of pairs of wind observations. The
            procedure seeks a wind speed factor nu M and a wind direction factor phi (p)
            that will adjust a set of the paired wind observations so there will be a minimum sum
            of squared vector differences for the set. In this climate variation study, the sets
            consist of long-period monthly mean wind vectors from each 20 geographic area of the
            COADS summary paired with the average monthly winds of the square area. The
            nu. and phi factors are the vector characteristic differences of the 2* square area
            yearly winds from the long-term means.

            Vector Differences:

                The wind vectors (R) from one set, R, = p, exp (i 01) are compared to a second set
            of vectors, R2= V P2 exp(i (02+0)). With 0) wind vectors in each of the sets, pU and
            p2, are the wind speeds, 0,, and 02j are wind directions, and (i) is the imaginary
            number exponent of the complex numbers representing the wind vectors. With E
            representing the amount of squared inequality between R, and R2pairs in a group
            of N observations,

                                  N                                            N
                                j=1 E p. exp(i 61) - vp2, exp(i (02j + 0)) 2 = j=J E Ej


            After expanding the squared difference, differentiating E sum with respect to nu. and
            then with respect to phi, and setting the differentials equal to zero, the differentials
            may be solved algebraically to produce equations for nu and phi. These nu and phi
            are the factors to make the squared vector differences a
                 mum,






                                          N
                                        j=1 E (Pli P2i) COS(eli - 02i
                          NTJ
                                                     E (p2j) 2
                                                  j=1



                                                       77




 I
 I
 I
 I                                 N
                                  j=l E SIN (61i - 02i)
                 PHI = ARC TAN
 I                                N
                                  j=l E Cos (01i - N)
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 1                                       78
 1
                                    1




 I
 I
 I                             APPENDIXC
 I            MONTHLY WIND FIELDS OF LOW IIATITUDE REGIONS
 I                 OF THE ATLANTIC AND PACIFIC OCEANS
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 1                                 79
 1










                      MONTHLY WIND FIELDS OF LOW LATITUDE REGIONS
                              OF THE ATLANTIC AND PACIFIC OCEANS



            INTRODUCTION:



                Monthly vector-averaged wind fields were computed from COADS (NOAA, 1985)
            for the area 130* E eastward to 10' W longitude, 0' to 40' N latitude. Based on
            climatology of the Intertropical Convergence (ITC) zone (Godshall, 1968), August was
            selected to represent the summer season and January was selected to represent
            winter. Monthly vector-mean wind was computed from the COADS 2' square
            averaged wind components and analyzed to illustrate the broad-scale surface wind
            regimes. During January (Figure C-1.), the westerlies are found over the northern
            part of the study region, the subtropical anticyclones and trade winds are relatively
            weak, and the ITC is at it's southern most position in the Atlantic and Pacific oceans.
            During August (Figure C-2.), the westerlies barely get into the study area, the trade
            winds are relatively strong, and the ITC is at the northern most position.



























                                                       80








                                                                 'ji I I Ilar -Y wl Ild V(@(-tors




                                                                                            -5 .1":


                                                 jr,












                                                                                                             . . . . . . . . . .






                                           ...            --------------


                  - - - - - - - - - - - - -
                                                                                  ........              43, ;V
                         ..........

                              I@Iklw
                ............


















                                                   Figure C-1. January Vector Mean Surface Winds.


                                                                               81












                                                                 AugusL wind vectors














                                         ...........
                                                  -------------------

                                                      ------  ------

                                                                                                    - ------------------
                                         ------------------------                                ---- ---
                                                                        . . ...... .
                                              ---------- --- -                .. ......
                                                                                .........
                                  ..........      ----------











                                                   Figure C-2. August Vector Mean Surface Winds.

                                                                              82



                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  I
                                                  1
                                  3 6668 14100 74781