CICEET Progress Report for the period 10/1/04 Through 3/15/05

Project Title: A community model for the Chesapeake Bay
Principal Investigator(s): Ming Li
Additional Investigator(s): Raleigh R. Hood and William C. Boicourt
Project Start Date: 09/01/2002

Figures

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Tasks to meet objectives
Our work plan consists of three parts: (1) Use the coupled hydrodynamic/biogeochemical model to run hindcast simulations for the years between 1995 to 2000, covering a variety of climatic conditions. (2) Work with managers and scientists in the Chesapeake Bay region to develop management scenarios and use the model to make projections. (3) Develop an oxygen dynamics submodel.

Progress on Tasks
Despite nutrient management efforts, many estuaries like the Chesapeake Bay are experiencing deteriorating rather than improving water quality conditions. A major impediment to developing a successful restoration strategy is the complicating effects of climate variability on plankton production and water quality. Large interannual fluctuations in river runoff result in highly variable nutrient loading to and circulation within estuaries, while episodic wind events exert more subtle and poorly-understood controls on biogeochemocal processes. While hypoxia is in general correlated with winter-spring river runoff in the Chesapeake Bay, there are several years in the past 20 years that exhibit modest to strong departure from this statistical correlation. To understand the effects of interannual climate variability, we have conducted hindcast simulations over a six-year period between 1995 and 2000, spanning years of highly variable hydrological and meteorological forcing conditions. To investigate how wind affects plankton production, we have conducted model simulations with and without wind forcing. To explore how nutrient management efforts might affect plankton production and water quality, the biophysical model has been forced with nutrient concentrations at one-half and twice of those levels observed at the tributaries. The model results show interesting interannual variations in plankton biomass and regional distributions between the high and low runoff years.

To validate the coupled biophysical model, we have conducted extensive comparisons between the model results and observations. For the hydrodynamic model, we have compared the model results against time series measurements of tidal elevation at tidal gauge stations, surface and bottom salinity at Chesapeake Bay Program (EPA CBP) monitoring stations, and velocity at Chesapeake Bay Observing System (CBOS) mid-bay buoy as well as synoptic along-channel and cross-channel salinity distributions obtained during NSF-funded hydrographic surveys. For the biogeochemical model, we have compared the model results with nitrate, ammonium, chlorophyll measurements at CBP monitoring stations and surface chlorophyll maps acquired by remote-sensing aircrafts. Since the six-year simulations generate large data sets, we have not yet completed analyzing the model results.

We have made limited progress in the development of oxygen submodel. The oxygen model in the CBP water-quality model is complex and cannot be easily incorporated into ROMS. In collaboration with other biogeochemical modelers in the ROMS community, we have started developing the oxygen submodel for the Chesapeake Bay. An oxygen module is being incorporated into the biogeochemical model, but numerical results are preliminary at this moment and we have not validated the model results against oxygen measurements in the Bay.

Accomplishments
We have conducted hindcast simulations over a six-year period and investigated how climate variability affect plankton production in the Chesapeake bay. These model results will be of great value in deciphering the causes of deteriorating water quality conditions: natural climate variability versus nutrient enrichments.

Project Objectives for Next Reporting Period

Objectives
Our objective for the next reporting period is to complete the development of the coupled hydrodynamic/biogeochemical model for the Chesapeake Bay and present the model results to the community.

Tasks to Meet Objectives
To meet this objective, we need to (1) further improve the water quality model, analyze the model results over years of variable climatic forcing conditions and under different nutrient management scenarios, and (2) present the model results to the scientific and research communities in the Chesapeake Bay region.

Work Plan for Next Reporting Period
Our work plan consists of two parts: (1) Refine the biogeochemical model and complete the analysis of the hindcast simulations. (2) Work with managers and scientists to disseminate the model results.

Anticipated Success in Meeting Project Objectives
By the end of this reporting period, we expect to have a new coupled hydrodynamic-water quality model for the Chesapeake Bay. This model should provide much-needed cross-comparisons with the CBP water-quality model which has been relied on exclusively for designing nutrient management strategies.

Overall Project Timeline Update
We have requested one-year extension for this project and our request has been approved by Dr. Langan.

Preliminary Data
National Ocean Survey (NOS) maintains water-level gauges in the Chesapeake Bay. In Figure 3 we show a comparison of the modeled and observed tidal elevations at three representative stations in the Bay. The Baltimore station is located in the Upper Bay, the Lewisetta station in the Mid Bay and the CBBT (Chesapeake Bay Tunnel Bridge) station is located near the mouth of the Bay (see Figure 2 for their locations). The model provides an accurate prediction for tidal elevations in the Chesapeake Bay.The rms error is 4.5 cm, 2.1 cm and 5.4 cm, and the relative average error is 4.5%, 1.2% and 2.0% at the three stations. The correlation coefficient is 0.98, 0.99 and 0.99 and the model skill is 0.98, 0.99 and 0.99.

To evaluate how the model captures the temporal variability in salinity, we located four stations in the main stem: Station CB3.3C, CB4.4, CB5.4 and CB8.1 (see Figure 2 for their locations). They occupy different salinity regimes along the main axis of the Bay, ranging from nearly-fresh water in the Upper Bay to shelf salinity at the Bay mouth. In addition, we located one station in each of the two large tributaries: LE2.3 in the Potomac River and LE5.5 in the James River. We compare the salinity time series of 1996 in Figure 4 and those of 1997 in Figure 5. The model appears to have captured the seasonal salinity variations well. The model does a better job in hindcasting the normal runoff year of 1997 than the high runoff year of 1996. This result is expected because turbulence mixing parameterization schemes perform better under lower runoff and weaker stratification conditions.

Time-series current observations were made at the Chesapeake Bay Observing System (CBOS) Mid-Bay Station located in the deep channel of the Bay, approximately 100 km seaward of the Susquehanna River mouth (see Figure 2 for its location). Typically, two fixed-depth conventional current meters were employed to obtain flow measurements at 2.4 and 19 m depths. We carried out a comparison of the subtidal velocity over a forty-day period during 1997 (Figure 6). There were a series of strong wind events, each lasting for 2 to 5 days, as shown in both the original and filtered winds. The Chesapeake Bay responded to this local longitudinal wind forcing, with amplitude and duration of currents matching those in the wind record reasonably well. Figure 6 shows that the modeled currents track the observed current. When averaged over the monthly record shown in Figure 6, the observed current shows a net seaward flow of 0.08 ms^-1 at 2 m depth and a net landward flow of 0.02 ms^-1 at 19 m depth, consistent with the two-layer gravitational circulation.

To understand how interannual variability in river runoff affects plankton production in the Bay, we compare the surface distribution of spring phytoplankton biomass between the high runoff year of 1996 and normal runoff year of 1997, as shown in Figure 7. The spring bloom shifts to lower and mid Bay during 1996 as the fresh water carries nutrient further downstream while high sediment loading inhibits phytoplankton production in the upper bay. In contrast, the spring bloom occurs in the upper and mid bay during 1997, since nutrients are exhausted before reaching the lower bay and light field is favorable in the upper reaches of the Bay. These model predictions are in agreement with observed interannual variations of plankton biomass distributions.

Figure 8 compares the time series of nitrate, phytoplankton and zooplankton concentrations at the mid-bay station CB5.1 between the high runoff year of 1996 and normal runoff year of 1997. Red cross symbols represent data collected by Chesapeake Bay Program (CBP) measurements while blue lines are model predictions. The model captured some aspects of the observed temporal variability but misses other aspects. Further improvements on the biogeochemical model are clearly needed.

Dissemination
Publications:
Li, M., L. Zhong and W.C. Boicourt. 2004. ROMS simulations of the Chesapeake Bay estuary: sensitivity to turbulence mixing parameterization and comparison with hydrographic observations. J. Geophys. Res., in review.

Workshops:
L. Zhong, M. Li, S. Zhang and D. Zhang. 2004. Hindcast simulations of Chesapeake Bay circulation during Hurricane Isabel using a mesoscale coupled atmosphere-ocean model. Hurricane Isabel in Perspective Conference. November 15 - 17, 2004. Maritime Institute, Linthicum Heights, Maryland.

Expenditures
Each institution’s grants office is responsible for submitting financial reports. Please state whether or not expenditures are in the range anticipated for the work accomplished to date.
Yes.