LAUSR

This analysis is the first global validation of the Moderate Resolution Imaging Spectroradiometer (MODIS)-derived near-surface air temperature and dew point estimates, which both serve as crucial input data in models of energy, water, and carbon exchange between terrestrial ecosystems and the atmosphere. By hypsometrically interpolating the MOD07 Level-2 atmospheric profile product to surface pressure level, we obtained near-surface air temperature and dew point observations at 5 km pixel resolution. We compared these daily data, retrieved over a 14-year record, to corresponding measurements from 109 ground meteorological stations (FLUXNET). Our results show strong agreement between satellite and in situ near-surface air temperature measurements (R = 0.89, root-mean-square error = 3.47°C, and bias = 0.19°C) and dew point observations (R = 0.76, root-mean-square error = 5.04°C, and bias = 0.79°C) with insignificant differences in error across climate zones. This validation is among the earliest assessments of the reprocessed, crosstalk-corrected Collection 6.1 Terra MODIS data and provides support for widespread applications of near-surface atmospheric data. Plain Language Summary Scientists often use complex models to study the Earth’s land, water, and atmosphere. Most models require various types of data that describe different processes critical to climate. Two common ingredients in these models are air temperature and dew point temperature, the latter a measure of the moisture in the air, near the ground. Although weather stations can report the two temperatures at precise locations, satellites can make measurements that are uniformly distributed over the entire globe. For this reason, many models use satellite data, particularly those with moderate to fine resolution, to represent near-surface air temperature and dew point. One common source of such data is a product from the Moderate Resolution Imaging Spectroradiometer (MODIS) satellite instrument. Satellite data require comparison against ground measurements, and, until now, the uncertainties in near-surface air temperature and dew point from the MODIS product had not been thoroughly studied. In this analysis, we compared daily MODIS data over a 14-year record to corresponding measurements from over 100 ground meteorological stations worldwide. The high accuracy we uncovered will allow other scientists to confidently use near-surface air temperature and dew point estimates from the satellite product in their models.