Effect of an environmental flow on vegetation growth and health using ground and remote sensing metrics

Abstract

Understanding the effectiveness of environmental flow deliveries along rivers requires monitoring vegetation. Monitoring data are often collected at multiple spatial scales. For riparian vegetation, optical remote sensing methods can estimate growth responses at the riparian corridor scale, and field-based measures can quantify species composition; however, the extent to which these different measures are duplicative or complementary is important to understand when planning monitoring programmes with limited resources. In this study, we analysed riparian vegetation growth in the delta of the Colorado River in response to an experimental pulse flow. Our goal was to compare ground-based measurements of vegetation structure and composition with satellite-based Landsat radiometric variables, such as the normalized difference vegetation index (NDVI). We made this comparison in 21 transects following the delivery of 131.8 million cubic meters (mcm) of water in the stream channel during the spring of 2014 as a pulse flow and 38.4 mcm as base flows. Vegetation cover increased 14% and NDVI increased 0.02 (15%) by October 2015, and both variables returned to pre-pulse flow values in October 2016. Observed changes in vegetation structure and composition did not persist after the second year. The highest increase in vegetation cover in October 2014 and October 2015 resulted from species that could respond rapidly to additional water such as reeds (Arundo donax and Phragmites australis), cattail (Typha domingensis), and herbaceous plants. Dominant shrubs, saltcedar (Tamarix spp.) and arrowweed (Pluchea sericea), both indicative of nonrestored habitats showed variable increases in cover, and native trees (Salicaceae family) presented low increases (1%). The strong NDVI-vegetation cover relationship indicates that NDVI is appropriate to detect changes at the riparian corridor scale but needs to be complemented with ground data to determine the contributions by different species to the observed trends.