We present a comparison of two ecohydrologic models commonly used for planning land management to assess the production of hydrologic ecosystem services: the Soil and Water Assessment Tool (SWAT) and the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) annual water yield model. We compare these two models at two distinct sites in the US: the Wildcat Creek Watershed in Indiana and the Upper Upatoi Creek Watershed in Georgia. The InVEST and SWAT models provide similar estimates of the spatial distribution of water yield in Wildcat Creek, but very different estimates of the spatial distribution of water yield in Upper Upatoi Creek. The InVEST model may do a poor job estimating the spatial distribution of water yield in the Upper Upatoi Creek W total water yield, which means that storage dynamics which are not modeled by InVEST may be important. We also compare the ability of these two models, as well as one newly developed set of ecosystem service indices, to deliver useful guidance for land management decisions focused on providing hydrologic ecosystem services in three particular decision contexts: environmental flow ecosystem services, ecosystem services for potable water supply, and ecosystem services for rainfed irrigation. We present a simple framework for selecting models or indices to evaluate hydrologic ecosystem services as a way to formalize where models deliver useful guidance.
KEYWORDS:ecosystem services, water yield, model comparison, SWAT, InVEST
INTRODUCTIONHydrologic ecosystem services (HESs) are beginning to influence land management decisions through both regulations and investments targeted at protecting and improving water resources [Le Maitre et al., 2007;Gordon et al., 2010;Goldman-Benner et al., 2012]. HESs are the goods and services that ecosystems provide to people related to various uses of water, and include water availability for municipal, agricultural, and commercial use, the reduction of the magnitude and frequency of flow peaks to prevent floods, the reduction of sediment and nutrients in water, and the value of natural hydrologic systems for recreation [Brauman et al., 2007; Keeler et al., 2012;Brauman, 2015]. Land managers use HESs to support investments and meet regulatory requirements around water resources in both developing countries [Goldman, 2009] and developed countries [Lautenbach et al., 2010;Logsdon and Chaubey, 2013]. To support investments or meet regulations, scientists and managers assess HESs under expected future conditions based on some model of the landscape and ecosystem response; the landscapes are then managed to restore, retain, or optimize HESs and thus improve the status of the water resources. Scientists and managers assess HES provision in distinct decision contexts-combinations of investment and regulatory goals and institutional structures-which may decide both which models are applied and how model outputs are used to make decisions.