Remote sensing grassland phenology in the greater Yellowstone ecosystem : biophysical correlates, land use effects and patch dynamics

Abstract

Vegetation phenology refers to the seasonal timing of repeat biological events such as bud burst and primary-productivity and their relationship to climate. The spatial location and timing of phenology is relevant to a wide-variety of questions in ecology including the space use and population dynamics of migratory herbivores. Recent technological (remote sensing) and methodological (statistical smoothing algorithms and weighted-regression) advancement now allow for mapping spatial and temporal patterns of vegetation phenology across large spatial extents and at fine-temporal scales. It also allows for examination of vegetation response to climate. An understudied topic investigates how human activity (i.e. land use) modifies broad-scale patterns of phenology from their natural state. Land use effects on phenology is important in the context of parks and protected areas where human activity in surrounding areas can compromise biodiversity conservation goals. With this in mind, we posed the following research questions for a study-area within the Greater Yellowstone Ecosystem: 1) What are the biophysical correlates and likely drivers of landscape-scale grassland phenology under wildland conditions? 2) How do different types of land use modify grassland phenology from its wildland state? And, 3) Do spatial and temporal patterns of green forage patches produced with new tools and datasets display seasonal-dynamics that are consistent with current ecological understanding? To answer these questions we used the Normalized Difference Vegetation Index (NDVI) produced by the Moderate Resolution Imaging Spectroradiometer (MODIS) as input to the TIMESAT algorithm to produce estimates of grassland phenology. Our principle findings are that: 1) Seasonal variation in solar radiation, water availability, evaporative demand and temperature explained much of the variation in the timing of wildland grassland phenology; 2) All land use types extended the length of the growing season and agriculture increased two estimates of productivity; And, 3) New tools are capable of producing nearly-spatially and -temporally continuous maps of the pattern of green forage patches that are consistent with current ecological understanding. Results of the present study suggest that land use intensification in the Greater Yellowstone Ecosystem has the potential to alter landscape-scale ecosystem process with a variety of expected consequences for wildlife conservation and management.