Hydrology and water quality of a field and riparian buffer adjacent to a mangrove wetland in Jobos Bay Watershed, Puerto Rico

Authors: Williams, Candiss; Lowrance, Robert; Bosch, David - Dave; WILLIAMS, JIMMY - Texas A&M University; BENHAM, E - Charles E Kellogg National Soil Survey Laboratory & Research; DIEPPA, ANGEL - National Oceanic & Atmospheric Administration (NOAA); Hubbard, Robert; MAS, EDWIN - Natural Resources Conservation Service (NRCS, USDA); Potter, Thomas; SOTOMAYOR, DAVID - University Of Puerto Rico; STEGLICH, EVELYN - West Texas A & M University; Strickland, Timothy - Tim; Williams, Randall

Submitted to: Ecological Engineering
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/11/2012
Publication Date: 1/24/2013
Citation: Williams, C.O., Lowrance, R.R., Bosch, D.D., Williams, J.R., Benham, E., Dieppa, A., Hubbard, R.K., Mas, E., Potter, T.L., Sotomayor, D., Steglich, E.M., Strickland, T.C., Williams, R.G. 2013. Hydrology and water quality of a field and riparian buffer adjacent to a mangrove wetland in Jobos Bay Watershed, Puerto Rico. Ecological Engineering. http://dx.doi.org/10.1016/j.ecoleng.2012.09.005.

Interpretive Summary: The Agricultural Policy/Environmental eXtender (APEX) model and the Riparian Ecosystem Management Model (REMM) were used to simulate water quality and hydrology from agricultural lands adjacent to the Jobos Bay watershed in southern Puerto Rico. The bay is home to the Jobos Bay National Estuarine Research Reserve and potential agricultural runoff is a concern in the region. Three years of data were used to calibrate and run the models. Both models simulated water table depths that were similar to the field estimated water table. The R2 values ranged from 0.41 to 0.95. The calibrated models provided estimates of sediment and nutrient transport from both the field and riparian areas to the bay and suggest that the forested riparian zone between the farm fields and the bay reduced surface runoff by 16%; subsurface flow by 67%, sediment yield by 24%, total nitrogen loading by 31%, and total phosphorus loading by 30%. Simulations indicate that tropical storm and hurricane events may account for 63% of total loadings to the bay...primarily via surface runoff. The results of this study should aid management agencies in selecting the most effective conservation practices to reduce agricultural loading to the Reserve.

Technical Abstract: Models that estimate the effects of agricultural conservation practices on water quantity and quality have become increasingly important tools for short- and long-term assessments. In this study, we simulated the water quality and hydrology of a portion of the Jobos Bay watershed, Puerto Rico using two models, one for simulating the upland agricultural field and one for simulating the adjacent riparian zone. The model used to simulate upland field outputs was the Agricultural Policy/Environmental eXtender (APEX) model and the model used to simulate riparian functions was the Riparian Ecosystem Management Model (REMM). The field was irrigated with a center pivot system for the first two years of the study and the four quadrants of the field were treated as subareas in the APEX model. Water table depths were measured monthly in both the field and riparian areas and were used with recording well data to develop an interpolated estimate of daily water table depth in the four field and the riparian zone. Three years of water table data were available and the models were calibrated with the first two years, for four field subareas and the riparian buffer by adjusting functions within APEX and REMM. The simulated depth to water table averages for the field subareas and the riparian zone were within ±7% of the corresponding interpolated water table depths for both the calibration (2008-2009) and validation (2010) periods with the exception of field subarea 2 for the validation period. The R2 values ranged from 0.41 to 0.95. The Nash-Sutcliffe efficiencies (NSE) values ranged from 0.33 to 0.80, with the exception of field subarea 2. The percent biases were within ±9% during both the calibration and validation periods with the exception of subarea 2 which was -31.6%. The calibrated models were used to estimate nutrient and sediment transport from the fields and riparian zone. These models of the base conditions showed that over the 3-year study period, the riparian zone represented by REMM reduced surface runoff by 16%; subsurface flow by 67%, sediment yield by 24%, total nitrogen loading by 31%, and total phosphorus loading by 30%. Tropical storms and hurricanes played an important role in water and nutrient transport on this site. Tropical storm and hurricane events were predicted to represent on average 63% of total loadings. The results of this study indicate that both APEX and REMM have the ability to simulate water quality and quantity in Jobos Bay Watershed.