The PERSIANN system is an automated method for estimating rainfall using artificial neural networks (ANNs) from geosynchronous satellite longwave infrared imagery (GOES-IR) at a resolution of 0.25° × 0.25° every half-hour. The system improves accuracy by adaptively adjusting network parameters using TRMM microwave imager (TMI) data and reduces random errors by aggregating data to a daily 1° × 1° resolution. The PERSIANN-GT product, based on GOES-IR and TRMM TMI data, was evaluated over the tropical Pacific Ocean and parts of Asia, Australia, and the Americas. It showed good agreement with radar-gauge composite data in Florida and Texas (correlation coefficient > 0.7) and with monthly WMO gauge measurements in high-gauge-density areas (ρ ~ 0.77–0.90). The product also compared well with TRMM 3B43 monthly data but indicated higher rainfall over the western Pacific compared to the adjusted TRMM 3B42 product. The study evaluated the PERSIANN-GT product against global rain gauge observations, NEXRAD gauge composite data over Florida and Texas, and various TRMM rainfall products. The product showed good agreement with rain gauge data, with correlation coefficients greater than 0.77 for grid cells with more than five gauges. The product also performed well against NEXRAD data, with an average correlation of 0.73 and root-mean-square error of 6.43 mm/day. The PERSIANN-GT product was found to overestimate rainfall over the ocean but underestimate it over land when compared to TRMM data. The product also showed good agreement with TRMM 3B43 data, with correlation coefficients of approximately 0.9 for grid cells with more than 10 gauges. The study concluded that the PERSIANN-GT product provides improved tropical rainfall estimates with high spatial and temporal resolution. The product's performance was consistent with other TRMM products, and it showed good agreement with rain gauge data. The study also highlighted the need for further improvements in the algorithm to better distinguish convective clouds from no-rain cirrus clouds based on infrared data. The PERSIANN-GT product is available for use and further testing, covering a 1-year period from August 1998 to July 1999. The study emphasizes the importance of high-quality TRMM rainfall data for improving global precipitation estimates for weather/climate, hydrology, and environmental studies.The PERSIANN system is an automated method for estimating rainfall using artificial neural networks (ANNs) from geosynchronous satellite longwave infrared imagery (GOES-IR) at a resolution of 0.25° × 0.25° every half-hour. The system improves accuracy by adaptively adjusting network parameters using TRMM microwave imager (TMI) data and reduces random errors by aggregating data to a daily 1° × 1° resolution. The PERSIANN-GT product, based on GOES-IR and TRMM TMI data, was evaluated over the tropical Pacific Ocean and parts of Asia, Australia, and the Americas. It showed good agreement with radar-gauge composite data in Florida and Texas (correlation coefficient > 0.7) and with monthly WMO gauge measurements in high-gauge-density areas (ρ ~ 0.77–0.90). The product also compared well with TRMM 3B43 monthly data but indicated higher rainfall over the western Pacific compared to the adjusted TRMM 3B42 product. The study evaluated the PERSIANN-GT product against global rain gauge observations, NEXRAD gauge composite data over Florida and Texas, and various TRMM rainfall products. The product showed good agreement with rain gauge data, with correlation coefficients greater than 0.77 for grid cells with more than five gauges. The product also performed well against NEXRAD data, with an average correlation of 0.73 and root-mean-square error of 6.43 mm/day. The PERSIANN-GT product was found to overestimate rainfall over the ocean but underestimate it over land when compared to TRMM data. The product also showed good agreement with TRMM 3B43 data, with correlation coefficients of approximately 0.9 for grid cells with more than 10 gauges. The study concluded that the PERSIANN-GT product provides improved tropical rainfall estimates with high spatial and temporal resolution. The product's performance was consistent with other TRMM products, and it showed good agreement with rain gauge data. The study also highlighted the need for further improvements in the algorithm to better distinguish convective clouds from no-rain cirrus clouds based on infrared data. The PERSIANN-GT product is available for use and further testing, covering a 1-year period from August 1998 to July 1999. The study emphasizes the importance of high-quality TRMM rainfall data for improving global precipitation estimates for weather/climate, hydrology, and environmental studies.