Heavy rainfall verification over Arunachal Pradesh during the Indian Summer Monsoon: a diagnostic study.
Saved in:
| Title: | Heavy rainfall verification over Arunachal Pradesh during the Indian Summer Monsoon: a diagnostic study. |
|---|---|
| Authors: | Sandeep, A1 (AUTHOR) drsandeepimd@yahoo.com, Saif, Rizwan1 (AUTHOR), Das, Sunit2 (AUTHOR), Mohan, K N2 (AUTHOR) |
| Source: | Journal of Earth System Science. Jun2026, Vol. 135 Issue 2, p1-18. 18p. |
| Subject Terms: | *Forecasting, *Model validation, *Monsoons, *Ensemble learning, *Rainfall, *Weather forecasting |
| Geographic Terms: | Arunachal Pradesh (India) |
| Company/Entity: | India. Meteorological Dept. |
| Abstract: | Forecasting heavy rainfall (≥64.5 mm and ≤115.5 mm in 24 hours) remains particularly challenging over complex terrains like Arunachal Pradesh (ARP). This study evaluates the performance of the IMD-Global Forecast System (GFS) model and an Operational Forecast (OPF) system based on a multi-model ensemble approach in predicting the heavy rainfall (HRF) events during the 2022–2024 monsoon seasons. Forecast verification was first conducted at state, next in sectors (western, central, and eastern ARP), and lastly at district level using four statistical metrics: Probability of Detection (POD), False Alarm Rate (FAR), Critical Success Index (CSI), and Missing Rate (MIR). Results indicate a declining trend in seasonal average rainfall but a rise in HRF event frequency. The OPF system consistently outperformed the IMD-GFS model with notable improvements in POD (11–16%) across all forecasting lead times (up to day-5). Notably, day-2 forecasts outperformed day-1 forecasts across all regions. The average POD values for all lead times were 0.76 (western ARP), 0.72 (central ARP), and 0.71 (eastern ARP). CSI values showed consistent improvements across all lead times using the OPF system. The highest gains were observed in western ARP (0.01 on day-1), central ARP (0.12 on day-1), and eastern ARP (0.12 on day-5). Moreover, CSI values declined with increasing lead time across all seasons and regions; however, 2024 recorded the highest CSI scores, followed by 2023 and 2022. Sector-wise, eastern ARP exhibited the best average CSI (0.52), with a peak of 0.59 on day-1. District-level analysis revealed high forecast skill in regions with larger sample sizes, reinforcing the robustness of the OPF system. A clear positive relationship was found between POD scores and sample size suggesting that detection skill improves with increased data volume. Overall, the OPF system offers enhanced skill and reliability for HRF forecasting, supporting early warning and disaster mitigation efforts in Northeast India. Research highlights: The study evaluates and compares the skill of IMD-GFS and OPF systems in predicting the heavy rainfall (HRF) events over Arunachal Pradesh during 2022–2024 monsoon seasons. OPF system consistently outperforms IMD-GFS across most verification metrics (POD, CSI, FAR, and MIR), with notable improvements in short-to-medium range (day-1 to day-3) forecasts. District-level analysis reveals spatial variability in model performance, highlighting that OPF shows stronger reliability in data-rich regions, while complex terrain and sparse networks remain key challenges. [ABSTRACT FROM AUTHOR] |
| Database: | Energy & Power Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
Be the first to leave a comment!
