Using NDVI and EVI to Map Spatiotemporal Variation in the Biomass and Quality of Forage for Migratory Elk in the Greater Yellowstone Ecosystem

Abstract

Abstract:The Normalized Difference Vegetation Index (NDVI) and the Enhanced Vegetation Index (EVI) have gained considerable attention in ecological research and management as proxies for landscape-scale vegetation quantity and quality. In the Greater Yellowstone Ecosystem (GYE), these indices are especially important for mapping spatiotemporal variation in the forage available to migratory elk (Cervus elaphus). Here, we examined how the accuracy of using MODIS-derived NDVI and EVI as proxies for forage biomass and quality differed across elevation-related phenology and land use gradients, determined if polynomial NDVI/EVI, site, and season effects improved these models, and then mapped spatiotemporal variation in the abundance of high quality forage available to elk across the Upper Yellowstone River Basin (UYRB) of the GYE. Models with a polynomial NDVI effect explained 19%-55% more variation in biomass than the linear NDVI and EVI models. Models with linear season effect explained 14%-20% more variation in chlorophyll, 37%-69% more variation in crude protein, and 26%-50% more variation in in vitro dry matter digestibility (IVDMD) than the linear NDVI and EVI models. Linear NDVI models explained more variation in biomass and quality across the UYRB than the linear EVI models. The accuracy of these models was lowest in grasslands with late onset of growth, in irrigated agriculture, and after the peak in biomass. Forage biomass and quality varied across the elevation-related phenology and land use gradients in the UYRB throughout the season. At their seasonal peak, the abundance of high quality forage for elk was 50% greater in grasslands with late onset of growth and 200% greater in irrigated agriculture than in all other grasslands, suggesting that these grasslands play an especially important role in the movement and fitness of migratory elk. These results provide novel information on the utility of NDVI and EVI for mapping spatiotemporal patterns of forage biomass and quality.

NDVI exhibits mixed success in predicting spatiotemporal variation in caribou summer forage quality and quantity

“…3. Although linear models of summer forage-NDVI relationships performed poorly, we found that NDVI variables (NDVI and DaysToMax) could be used in a broader modeling effort to depict spatiotemporal variation in biomass, N, DN, and DE by incorporating nonlinear forage-NDVI relationships (Kawamura et al 2005, Santin-Janin et al 2009, Garroutte et al 2016) and other covariates known to influence forage characteristics (i.e., Veg, Coast). Notes: Forage conditions were assessed relative to the Normalized Difference Vegetation Index (NDVI), the difference in the number of days between the sampling date and the annual maximum NDVI value (DaysToMax), distance to the coast (Coast; calculated per 10 km), and vegetation type (reference class = tussock tundra).…”

“…In addition to NDVI, we included the variable DaysToMax to account for temporally varying forage-NDVI relationships (Santin-Janin et al 2009, Garroutte et al 2016. To meet this objective, we modeled variation in field measurements of forage biomass, N, DN, and DE across the growing season as a function of NDVI variables and base habitat covariates.…”

“…Meanwhile, NDVI values increase with photosynthetic activity at the start of the growing season, tend to peak in mid-to late summer, and then decline with the onset of senescence (Pettorelli et al 2005). Despite this complexity, most ecologists assume that the relationship between forage conditions and NDVI is temporally consistent (i.e., linear) across the growing season (but see Santin-Janin et al 2009, Garroutte et al 2016. Investigators have hypothesized that NDVI values acquired early in the growing season are more likely correlated with forage nutrition, while values acquired later in the growing season are more likely correlated with biomass (Thoma et al 2002, Parrett 2007.…”

Normalized Difference Vegetation Index as an Estimator for Abundance and Quality of Avian Herbivore Forage in Arctic Alaska

“…Remotely sensed vegetation indices are well established tools of wildlife research and management that can serve as proxies for landscape level changes in biomass and nutrient quality without the repeated destructive sampling of vegetation [12][13][14][15][16]. Of these indices, the Normalized Difference Vegetation Index (NDVI) is one of easiest to calculate, most commonly available and has been proven to be an effective tool for monitoring vegetative changes across a wide range of terrestrial ecosystems, including alpine and arctic tundra, grasslands, and temperate and tropical forests [14][15][16][17][18][19][20], see [21] for a review. The index is a simple ratio calculation of the red and near-infrared (NIR) reflectance bands that is sensitive to the reflected photosynthetically active radiation of plants [22,23].…”

Preface: Remote Sensing of Biodiversity

“…Wang et al [21] also address grassland productivity in a Southern Alberta prairie using airborne imaging spectrometry combined with ground sampling and eddy covariance data, showing greater productivity in sites with higher biodiversity based on species richness and the Shannon Index. Garroutte and Hansen [22] evaluated the quality of grasslands for elk habitat in the Yellowstone River Basin using seasonal MODIS EVI and NDVI. Zhao et al [23] address the optimal detection of biochemical indicators for species mapping, and two papers show the potential of mapping foliar traits related to ecosystem functionality.…”