Parameter estimation of GOES precipitation index at different calibration timescales

Abstract

We examined two techniques that adjust the parameters of the GOES Precipitation Index (GPI) by combining the polar microwave and the geosynchronous infrared observations at three frequencies: daily, pentad, and monthly. The first technique is the adjusted GPI (AGPI), and the second is the universally adjusted GPI (UAGPI). The study shows that rainfall estimates can be improved by frequent calibrations providing there is sufficient superior (microwave) rainfall sampling within the calibration time and space domain. For this work, daily and pentad calibrations produce monthly rainfall estimates almost as good as monthly calibration. The daily calibration produced better daily rainfall estimates than pentad and monthly calibration, but it generates similar pentad rainfall estimates to these of the pentad calibration. The monthly calibrated scheme is not suitable for the daily and pentad rainfall estimates. Under the current twice‐per‐day sampling rate of polar‐orbiting microwave observations, the pentad calibration scheme is suggested for the monthly, pentad, and daily rainfall. The potentials of applying the UAGPI and the AGPI techniques for daily rainfall estimation are also investigated.