Evaluation of TEMPEST STP-H8 Surface Rain Rate Estimates for Tropical Cyclones, Hurricanes and Typhoons

This study evaluates surface rain rate estimates derived from the Temporal Experiment for Storms and Tropical Systems (TEMPEST) Space Test Program-Houston 8 (STP-H8) observations, focusing on tropical cyclone events all over the globe. The TEMPEST STP-H8 mission enhances the temporal coverage of microwave atmospheric measurements by leveraging small satellite technology within low Earth orbit (LEO) constellations. As a near-identical successor to the TEMPEST-D (where ‘D’ stands for Demonstration) sensor, TEMPEST STP-H8 was launched from Kennedy Space Center on December 21, 2021, and subsequently deployed to the Japanese Experiment Module Exposed Facility (JEM-EF) on the International Space Station (ISS) on January 8, 2022, beginning its planned 3.5-year period of operations. The instrument continuously collects atmospheric data in five microwave channels (89, 165, 176, 180, and 182 GHz).

Prior research [1] established the efficacy of machine learning (ML) models for surface rain rate retrievals using TEMPEST-D brightness temperature (TB) observations. Considering that the TEMPEST-D and TEMPEST STP-H8 sensors are very similar to each other, aside from minor channel frequency differences, this study applies the previously developed ML-model rain rate estimation to TEMPEST STP-H8 TB data. The retrieved rain rates are cross-validated using Global Precipitation Measurement (GPM) Integrated Multi-satellite Retrievals for GPM (IMERG) products and other established satellite-based precipitation datasets.

Preliminary results using TEMPEST STP-H8 data over Hurricane Hilary in 2023 and Tropical Cyclone Batsirai in 2022 demonstrate that the ML-driven estimation approach effectively captures both the variability and intensity of rain rates, achieving a correlation coefficient (r) of 0.81 and a mean absolute error (MAE) of 7.0 mm/hr. This study extends the evaluation to the extensive dataset now available from TEMPEST STP-H8, assessing the model’s performance across oceanic regions across the globe. Additional findings, including an extensive cross-comparison analysis and further case studies, are also presented.

References:

[1] C. Radhakrishnan, V. Chandrasekar, S. C. Reising and S. T. Brown, "Development of Surface Rain Estimates from TEMPEST-D Observations," IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Pasadena, CA, USA, 2023, pp. 666-669, doi: 10.1109/IGARSS52108.2023.10282117.