|Larose, Myriam -|
Submitted to: Catena
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: March 13, 2011
Publication Date: May 3, 2011
Citation: Larose, M., Heathman, G.C., Smith, D.R., Norton, L.D. 2011. Impacts of conservation buffers and grasslands on total phosphorus load using hydrologic modeling and remote sensing techniques. Catena. 86:121-129. Interpretive Summary: Natural surface and ground water systems are crucial resources for our societal well-being. However, many of these systems have been severely jeopardized due to a variety of water quality issues that continue to persist. Agrochemical losses present one such issue which, unfortunately, are expected to continue to affect these valuable resources unless sustainable and innovative approaches are adapted that achieve effective water resource management. Phosphorus (P) is one of the major macronutrients needed for optimal crop yield that can also contaminates lakes, rivers, and coastal waters. An excess of the soluble fraction of P (orthophosphate, inorganic polyphosphates, and organic phosphate compounds) in runoff can contribute to eutrophication of freshwater, a major water quality problem worldwide. The goal of this study was to assess the impacts of vegetative conservation practices on water quality using a hydrology non-point source model with a land cover dataset that includes vegetated buffers and conservation grasslands along with four precipitation datasets. The specific objectives were to: 1) develop a land cover data set to estimate VBS and conservation grasslands using object-based image analysis; 2) use the Soil and Water Assessment Tool model along with rain gauge and radar precipitation data to enhance the watershed continuous hydrological modeled responses at two different scales; and 3) assess the impacts of vegetative buffers and conservation grasslands on total phosphorus loads using the Soil and Water Assessment Tool model. Although previous studies have shown that the NEXRAD precipitation data improved hydrological predictions, this study is built on the following premises: 1) Precipitation data input in the SWAT model from a sufficiently dense network of rain gauges is adequate to improve the accuracy of hydrologic processes controlling water quality; and 2) land cover data with vegetative buffers and conservation grasslands estimates provide a more realistic approach for quantifying the environmental benefits of the program on water quality.
Technical Abstract: To better assess the impacts of conservation buffers and grasslands on water quality at large spatial scales, development and integration of novel approaches are crucial to ensure that these land management practices are functioning properly and meeting their original goals. Recent developments in remote sensing technology have greatly enriched the availability of geospatial data that can be used in hydrological modeling to assess the potential hydrological response of conservation practices over larger areas. A methodology was developed using the object-based image analysis approach with Landsat-5 TM imagery of the year 2005 to quantify conservation buffers and grasslands (OBIA-2005). The OBIA-2005 was used in the Soil and Water Assessment Tool hydrologic model, along with different precipitation datasets to assess the impacts of vegetative practices on total phosphorus (TP) loads. The results showed that precipitation data from radar were lower than that from the rain gauges, which translated into an underestimation of modeled discharge values. In general model efficiency for streamflow values was within acceptable statistical ranges. While calibration of TP loads were satisfactory for the total contributing area of two nested catchements within the SJRW, the model was not able to reproduce the observed values of a subwatershed of less than 300 km2. Vegetative buffers of 30.5 m and 61 m combined with conservation grasslands generated from the OBIA-2005 resulted in large reduction of TP loads as compared to no practices. The results also showed that conservation grassland alone would reduce TP loads by less than 2%. These findings demonstrated that the representation of vegetative practices in the geospatial land covers could be an alternative in assessing the impacts of conservation buffers and grasslands on water quality.