H52E-03
Relating watershed nutrient loads to satellite derived estuarine water quality
Friday, 18 December 2015: 10:50
3022 (Moscone West)
John C Lehrter, US EPA, Gulf Breeze, FL, United States and Chengfeng Le, Zhejiang University, Ocean College, Zhejiang, China
Abstract:
Nutrient enhanced phytoplankton production is a cause of degraded estuarine water quality. Yet, relationships between watershed nutrient loads and the spatial and temporal scales of phytoplankton blooms and subsequent water quality impairments remain unquantified for most systems. This is partially due to a lack of observations. In many systems, satellite remote sensing of water quality variables may be used to supplement limited field observations and improve understanding of linkages to nutrients. Here, we present the results from a field and satellite ocean color study that quantitatively links nutrients to variations in estuarine water quality endpoints. The study was conducted in Pensacola Bay, Florida, an estuary in the northern Gulf of Mexico that is impacted by watershed nutrients. We developed new empirical band ratio algorithms to retrieve phytoplankton biomass as chlorophyll a (chla), colored dissolved organic matter (CDOM), and suspended particulate matter (SPM) from the MEdium Resolution Imaging Spectrometer (MERIS). MERIS had suitable spatial resolution (300-m) for the scale of Pensacola Bay (area = 370 km2, mean depth = 3.4 m) and a spectral band centered at wavelength 709 nm that was used to minimize the effect of organic matter on chla retrieval. The algorithms were applied to daily MERIS remote sensing reflectance (level 2) data acquired from 2003 to 2011 to calculate nine-year time-series of mean monthly chla, CDOM, and SPM concentrations. The MERIS derived time-series were then analyzed for statistical relations with time-series of mean monthly river discharge and river loads of nitrogen, phosphorus, dissolved organic carbon, and SPM. Regression analyses revealed significant relationships between river loads and MERIS water quality variables. The simple regression models provide quantitative predictions about how much chla, CDOM, and SPM concentrations in Pensacola Bay will increase with increased river loading, which is necessary information for nutrient, land-use, and climate management decisions. We will discuss the lessons learned for management of Pensacola Bay and about communicating the value of satellite observations to water quality managers.