Applying drone and satellite data to natural vegetation monitoring for agricultural sustainability
Primary Author: Amanda Stahl
Faculty Sponsor: Alexander Fremier
Primary College/Unit: Agricultural, Human and Natural Resource Sciences
Category: Agricultural and Natural Resource Sciences
Conserving natural vegetation along streams is an important on-farm strategy to improve water quality. The Voluntary Stewardship Program (VSP, 2011) requires participating agricultural counties in Washington to monitor and report whether ecosystem functions and values are being maintained or enhanced. Emerging remote sensing technologies could provide accurate, real-time, multiscale spatial data to increase monitoring efficiency and effectiveness. We are piloting drones and analyzing Sentinel-2 satellite images to quantify streamside vegetation condition and designing procedures for seamless integration into monitoring programs to improve agricultural sustainability.
We hypothesize that drone-mounted cameras and Sentinel-2 data can accurately document vegetation condition and change for VSP reporting. To test this, we collected images with two quadcopters (3DR Solo and DJI Matrice) at 9 sites across Whitman County. Drone images were compiled into mosaics and 3D surfaces, each referenced for accurate comparison across dates to document seasonality and resolve vegetation classification. We analyzed Sentinel data seasonally and inter-annually to quantify watershed-scale change dynamics using Google Earth Engine and ArcGIS.
Differing patterns of “greenness” clearly distinguished natural vegetation from agricultural land cover in Sentinel images collected July-October 2016-2019. Drone images captured finer details, including vegetation height, volume, and species. Initial findings illustrate that these data sources can detect the changing quantity and quality of natural areas in agricultural areas. In future work we will streamline satellite data analysis in Google Earth Engine and provide guidelines for drone-based monitoring so that counties and Conservation Districts can analyze data in real-time at regional scales.