Rapid estimation of fractional vegetation cover in grasslands using smartphones

Abstract

Measurements of fractional vegetation cover (FVC) are important for monitoring grassland growth and predicting aboveground biomass. Thus, a method to rapidly estimate grassland FVC is highly desired. Although smartphones provide a faster and less expensive method for estimating grassland FVC than satellites and unmanned aerial vehicles, their accuracy is not well understood. Here, we evaluated the use of smartphones to accurately estimate grassland FVC by taking photos in direct and indirect sunlight at three time points (8:00 a.m., 12:00 p.m. and 4:00 p.m.). We found that smartphone photography could be applied to grassland FVC estimation by extracting a specified percentile of color vegetation index (VIC) to reduce the effect of bright lighting. The 60th percentile value was more suitable for estimating grassland FVC than the 90th percentile, which was proposed in a previous study. We also found that extracting the 60th percentile VIC value from smartphone photos taken at noon under direct sunlight efficiently minimized the influence of lighting on estimated grassland FVC. We propose a model based on the VIC percentile method that can quickly estimate grassland FVC using smartphone photography. This model is very versatile for estimating FVC in semi-arid grasslands. Our findings show that using smartphone photos to quickly estimate grassland FVC is feasible and could provide a practical solution to quickly and accurately estimate grassland FVC.