Land surface phenology along urban to rural gradients in the U.S. Great Plains

The elevated surface and air temperatures of urban environments can influence the timing of vegetation growth dynamics within and across city boundaries. We examined patterns of land surface phenology (LSP) throughout the U.S. Great Plains region, which contains diverse metropolitan areas embedded within a predominately agricultural landscape. We assembled a time series (2002-2012) of Moderate Resolution Imaging Spectroradiometer (MODIS) Nadir BRDF-Adjusted Reflectance (NBAR) data and land surface temperature data at 500 m and 1000 m spatial resolution, respectively. We derived measures of the vegetated land surface and the thermal regime of the growing season at 8-day intervals using the Normalized Difference Vegetation Index (NDVI) and Accumulated Growing Degree-Days (AGDD). Fitting the convex quadratic LSP model of NDVI as a function of AGDD yielded two phenometrics - Peak NDVI and Thermal Time to Peak - and one model fit metric - the adjusted coefficient of determination (r2adj) - for each pixel per growing season. We linked the phenometrics with impervious surface area (ISA) data extracted from the U.S. Geological Survey National Land Cover Dataset (NLCD) to characterize differences in timing and amplitude of peak greenness between urban areas and their surrounding landscapes. Our results reveal the broad control of climatic conditions and moisture availability on phenological patterns across urban to rural gradients, with drier, southern cities displaying more varied responses of peak greenness timing and amplitude to urban intensity.