Publications

Scarino, B; Doelling, DR; Haney, C; Bedka, K; Minnis, P; Gopalan, A; Bhatt, R (2017). Utilizing the precessing orbit of TRMM to produce hourly corrections of geostationary infrared imager data with the VIRS sensor. INFRARED REMOTE SENSING AND INSTRUMENTATION XXV, 10403, UNSP 104030H.

Abstract
Accurate characterization of the Earth's radiant energy is critical for many climate monitoring and weather forecasting applications. For example, groups at the NASA Langley Research Center rely on stable visible-and infrared-channel calibrations in order to understand the temporal/spatial distribution of hazardous storms, as determined from an automated overshooting convective top detection algorithm. Therefore, in order to facilitate reliable, climate-quality retrievals, it is important that consistent calibration coefficients across satellite platforms are made available to the remote sensing community, and that calibration anomalies are recognized and mitigated. One such anomaly is the infrared imager brightness temperature (BT) drift that occurs for some Geostationary Earth Orbit satellite (GEOsat) instruments near local midnight. Currently the Global Space-Based Inter-Calibration System (GSICS) community uses the hyperspectral Infrared Atmospheric Sounding Interferometer (IASI) sensor as a common reference to uniformly calibrate GEOsat IR imagers. However, the combination of IASI, which has a 21:30 local equator crossing time (LECT), and hyperspectral Atmospheric Infrared Sounder (AIRS; 01:30 LECT) observations are unable to completely resolve the GEOsat midnight BT bias. The precessing orbit of the Tropical Rainfall Measuring Mission (TRMM) Visible and Infrared Scanner (VIRS), however, allows sampling of all local hours every 46 days. Thus, VIRS has the capability to quantify the BT midnight effect observed in concurrent GEOsat imagers. First, the VIRS IR measurements are evaluated for long-term temporal stability between 2002 and 2012 by inter-calibrating with Aqua-MODIS. Second, the VIRS IR measurements are assessed for diurnal stability by inter-calibrating with Meteosat-9 (Met-9), a spin-stabilized GEOsat imager that does not manifest any diurnal dependency. In this case, the Met-9 IR imager is first adjusted with the official GSICS calibration coefficients. Then VIRS is used as a diurnal calibration reference transfer to produce hourly corrections of GEOsat IR imager BT. For the 9 three-axis stabilized GEO imagers concurrent with VIRS, the midnight effect increased the BT on average by 0.5 K (11 mu m) and 0.4 K (12 mu m), with a peak at similar to 01:00 local time. As expected, the spin-stabilized GEOsats revealed a smaller diurnal temperature cycle (mostly < 0.2 K) with inconsistent peak hours.

DOI:
10.1117/12.2273883

ISSN:
0277-786X