Fyock, Ellen L.. Using Geographical Information System (GIS)-based watershed characteristics to predict stream visual assessment scores. Retrieved from https://doi.org/doi:10.7282/T36Q1XMT
DescriptionThe Stream Visual Assessment Protocol (SVAP) was designed as a cost-effective measurement of stream condition that can be used by non-scientific individuals with minimal training. It is intended as a preliminary assessment to determine problem areas where further study is needed. Although the SVAP is a simple assessment, resources are still required to train staff and volunteers on the protocol and to spend days in the field collecting data. However, if existing data can be used to obtain comparable stream health information without requiring field work, the cost associated with these resources can be reduced. This research investigated whether models can be created using previously collected SVAP data and GIS-based watershed characteristics to evaluate stream conditions to eliminate field work.
Using GIS, characteristics such as basin area and stream size were calculated for the areas draining into SVAP assessment locations. Digital data was used to determine characteristics based on land use/cover and soils. Statistical models were created using SVAP data from the Ramapo watershed in northeastern New Jersey. Models significant at α = 0.10 or lower were applied to the Wanaque watershed to determine whether the SVAP scores can be predicted in another location. While the regression models generally explained a high amount of variance in the Ramapo SVAP scores, the SVAP scores for Wanaque could not be accurately predicted from the test models.