CICEET Progress Report for the period 9/01/08 Through 2/15/09

Project Title: Applying multiple models to inform watershed management and land use planning a coastal community
Principal Investigator(s): Kathleen B. Boomer, Xuyong Li, Latif Kalin, Donald E. Weller
Project Start Date: 09/01/07
Report Compiled By: K. Boomer

Figures

Figure 1

Figure 2

Figure 3

Figure 4

Tables

Table 1

Table 2

Project Objectives for This Reporting Period
Objectives
Technical objectives and tasks: During the project’s third quarter, we planned to complete the model calibrations and apply the resulting models in order to estimate non-point source discharge under current and future land cover conditions throughout the Queenstown study area. We also expected to begin formally comparing the model predictions in an ensemble framework.

Outreach objectives and tasks: We planned to continue contributing project updates and attend semi-monthly Planning Commission meetings on an as-needed basis. We also planned to pursue an opportunity to work with the Town of Greensboro, Queen Anne’s County, and the Corsica River Restoration group.

Progress on Tasks
Model set-up, calibration, and application: At this time, our models are at different stages in their development and application. The GWLF model is fully calibrated and we have completed our first evaluation of non-point sources in the Queenstown area under current land cover conditions. The SWAT model is also now fully calibrated and ready to evaluate current and future land cover scenarios. The LTHIA model has been modified to accomplish the objectives of our study, but is not yet fully calibrated. Results to date for all of these models are presented in the Section IV of this report (Additional Information). Completion of the HSPF Phase 5 model calibration by the Chesapeake Bay Program has been delayed from December, 2008 to spring 2009.

Assessment of Alternative Future Land Cover Scenarios: During the past six months, we have worked with the Queenstown Planning Commission to refine the three alternate growth scenarios. Information from an economic analysis was used to predict the amount of future commercial development and the housing density. Resulting estimates of impervious surface area can be incorporated in each of the models to evaluate impacts from development.

Outreach tasks: We have continued to provide regular project updates during semi-monthly Planning Commission meetings.

Difficulties
Overall, calibration continued to be a challenging aspect of using multiple models to inform land management decisions. We had hoped to have our entire model set calibrated, but there remains some additional work on the LTHIA adaptation as well as the HSPF application, which is being developed through the Chesapeake Bay Program. Due to these difficulties, we anticipate requesting a one year, no cost extension on the grant.

Pertaining to outreach, it has continued to be challenging to engage other municipalities in our study area. This is largely because we do not have the capacity to interact with them as frequently and as closely as we can with the Queenstown Planning Commission.

Data Generated to date
The watershed delineations based on the 2 m DEM were used to summarize land cover proportions across the study area, which are included in Section IV of this report. In addition, initial results from the GWLF application estimating the amount of non-point source pollution under current conditions are included along with results from the SWAT calibration and the LTHIA adaptation.

Project Objectives for Next Reporting Period

Objectives
Technical objectives and tasks: With the calibration and validations completed for the six nearby watersheds with monitoring data, we will use the models to assess impacts from alternative land cover scenarios across the Queenstown area as well as different climate scenarios. We also will begin combining the predictions from the four watershed models using a model averaging framework and developing the source prioritization analysis.

Outreach objectives and tasks: We will continue to contribute project updates and attend semi-monthly Planning Commission meetings on an as-needed basis. We also will continue to engage the Town of Greensboro and Queen Anne’s County when we have results to share with them.

Work Plan to Meet Objectives
Continue Monitoring of sites as possible (exceeds project expectations)
Continue to assemble past and current data into database to support Task 3,4,5

Expenditures
To date, expenditures for personnel compensation at SERC amount to $52,626 (including salaries, benefits, and overhead). We also provided Auburn University with $77,961 for both contract years. Approximately $90,000 remains to be spent at SERC.

Additional Information
Land use of the study watersheds - We selected twelve watersheds in the Queenstown area (Figure 1). The twelve watersheds considered in this study range in size from 75.5 to 518.6 ha and differed in land cover (Table 1).

Watershed land cover was calculated from RESAC 2000 dataset (Mid-Atlantic RESAC, 2004). The proportion of cropland varies from 5.8% to 61.6%. Six of twelve watersheds have more than forty percent cropland. The proportions of other land cover categories are relatively lower. For examples, the proportion of forest varies from 16.3% to 49.0%, and developed land varies from 1.9% to 30.3%.

GWLF model prediction of flow and nutrient discharges

Water flow and nutrient discharges in average annual, wet year and dry year
We applied GWLF to predict water flow and nutrient discharges under current land cover during 1995-2004. During 1995-2004, the average total annual water discharge varied with watershed size, ranging from 265,647 m³ to 1794,519 m³; but the average water discharge per unit area varied little from 34.4 cm to 35.3 cm. Water discharge was more than double in a wet year (2003) with 56.1-57.0 cm compared with dry year (2001) with 20.5-21.1 cm (Table 2).

Watersheds QT04 and QT01 had highest and lowest discharges, respectively, of TN, TP, DIN and DIP (Table 2). QT08 had the highest nutrient discharges per unit area: 18.9 kg N ha^-1 for TN, 1.5 kg P ha^-1 for TP, 18.5 kg N ha^-1 for DIN, and 1.4 kg P ha^-1. QT11 had the lowest TN discharge per unit area, and QT09 had the lowest discharges of per unit area of TP, DIN, and DIP.

Changes in annual water flow and nutrient discharges during 1995-2004 - Overall, the trends of annual TN and TP discharges followed the trend of annual water flow discharge (Figure 2). In few years, the trends were slightly different. For example, the highest flow (QT01) appeared in 2003, but the highest nutrient discharge (QT01, QT02, and QT03) appeared in 1999.

Changes in monthly water flow and nutrient discharges during 1995-2004: The watersheds discharged more water, TN and TP during winter, spring and September. Discharges were lowest n summer (Figure 3).