Machine Learning-Based Error Modeling to Improve GPM IMERG Precipitation Product over the Brahmaputra River Basin

Bhuiyan, Abul Ehsan; Yang, Feifei; Biswas, Nishan Kumar; Rahat, Saiful Haque; Neelam, Tahneen Jahan

doi:10.3390/forecast2030014

Cited by 53 publications

(25 citation statements)

References 85 publications

(124 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several merged SPDs (MSPDs) are developed across various regions of the globe and especially in Pakistan. Very recently, machine learning (ML) techniques have become popular in hydrological modeling, climate studies, water resources management [43,44], and precipitation sciences [45][46][47]. Different ML techniques have also been employed for merging different SPDs and SPDs with RGs.…”

Section: Introductionmentioning

confidence: 99%

Assessment of Merged Satellite Precipitation Datasets in Monitoring Meteorological Drought over Pakistan

2021

View full text Add to dashboard Cite

The current study evaluates the potential of merged satellite precipitation datasets (MSPDs) against rain gauges (RGs) and satellite precipitation datasets (SPDs) in monitoring meteorological drought over Pakistan during 2000–2015. MSPDs evaluated in the current study include Regional Weighted Average Least Square (RWALS), Weighted Average Least Square (WALS), Dynamic Clustered Bayesian model Averaging (DCBA), and Dynamic Bayesian Model Averaging (DBMA) algorithms, while the set of SPDs is Global Precipitation Measurement (GPM)-based Integrated Multi-Satellite Retrievals for GPM (IMERG-V06), Tropical Rainfall Measurement Mission (TRMM) Multi-Satellite Precipitation Analysis (TMPA 3B42 V7), Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN), and ERA-Interim (re-analyses dataset). Several standardized precipitation indices (SPIs), including SPI-1, SPI-3, and SPI-12, are used to evaluate the performances of RGs, SPDs, and MSPDs across Pakistan as well as on a regional scale. The Mann–Kendall (MK) test is used to assess the trend of meteorological drought across different climate regions of Pakistan using these SPI indices. Results revealed higher performance of MSPDs than SPDs when compared against RGs for SPI estimates. The seasonal evaluation of SPIs from RGs, MSPDs, and SPDs in a representative drought year (2008) revealed mildly to moderate wetness in monsoon season while mild to moderate drought in winter season across Pakistan. However, the drought severity ranges from mild to severe drought in different years across different climate regions. MAPD (mean absolute percentage difference) shows high accuracy (MAPD <10%) for RWALS-MSPD, good accuracy (10% < MAPD <20%) for WALS-MSPD and DCBA-MSPD, while good to reasonable accuracy (20% < MAPD < 50%) for DCBA in different climate regions. Furthermore, MSPDs show a consistent drought trend as compared with RGs, while SPDs show poor performance. Overall, this study demonstrated significantly improved performance of MSPDs in monitoring the meteorological drought.

show abstract

Section: Introductionmentioning

confidence: 99%

Assessment of Merged Satellite Precipitation Datasets in Monitoring Meteorological Drought over Pakistan

2021

View full text Add to dashboard Cite

show abstract

“…This investigation adds insights to refine strategies for combining IR and PMW precipitation estimates in the merged products such as IMERG. Therefore, this study is different from previous studies in which the final products of IMERG is assessed using ground reference [5][6][7]22]. Results of this study can improve future versions of satellite-based precipitation products and provide insight for new sensors' design and their performance over different conditions and regions.…”

Section: Introductionmentioning

confidence: 84%

Investigating Various Products of IMERG for Precipitation Retrieval over Surfaces with and without Snow and Ice Cover

Arabzadeh

Behrangi

2021

Remote Sensing

View full text Add to dashboard Cite

Precipitation rate from various products of the integrated multisatellite retrievals for GPM (IMERG) and passive microwave (PMW) sensors are assessed with respect to near-surface wet-bulb temperature (Tw), precipitation intensity, and surface type (i.e., with and without snow and ice on the surface) over the contiguous United States (CONUS) and using ground radar product as reference precipitation. IMERG products include precipitation estimates from infrared (IR), combined PMW, and combination of PMW and IR. It was found that precipitation estimates from PMW products generally have higher skills than IR over snow- and ice-free surfaces. Over snow- and ice-covered surfaces: (1) most PMW products show higher correlation coefficients than IR, (2) at cold temperatures (e.g., Tw < −10 °C), PMW products tend to underestimate and IR product shows large overestimations, and (3) PMW sensors show higher overall skill in detecting precipitation occurrence, but not necessarily at very cold Tw. The results suggest that the current approach of IMERG (i.e., replacing PMW with IR precipitation estimates over snow- and ice-surfaces) may need to be revised.

show abstract

“…These limitations lead to higher uncertainties in SREs performance over complex terrain [31]. On one hand, few studies [38][39][40] have highlighted the potential of machine learning (ML) based error characterization methods in quantifying and correcting SREs uncertainties over the complex terrain. Similarly, a study conducted in China suggested that blended product better facilitate the hydrological modeling by blending SREs, reanalyzing, and gauging data [41].…”

Section: Introductionmentioning

confidence: 99%

Assessment of GPM-Era Satellite Products’ (IMERG and GSMaP) Ability to Detect Precipitation Extremes over Mountainous Country Nepal

et al. 2021

View full text Add to dashboard Cite

The reliability of satellite precipitation products is important in climatic and hydro-meteorological studies, which is especially true in mountainous regions because of the lack of observations in these areas. Two recent satellite rainfall estimates (SREs) from Global Precipitation Measurement (GPM)-era—Integrated Multi-Satellite Retrievals for Global Precipitation Measurement (IMERG-V06) and gauge calibrated Global Satellite Mapping of Precipitation (GSMaP-V07) are evaluated for their spatiotemporal accuracy and ability to capture extreme precipitation events using 279 gauge stations from southern slope of central Himalaya, Nepal, between 2014 and 2019. The overall result suggests that both SREs can capture the spatiotemporal precipitation variability, although they both underestimated the observed precipitation amount. Between the two, the IMERG product shows a more consistent performance with a higher correlation coefficient (0.52) and smaller bias (−2.49 mm/day) than the GSMaP product. It is worth mentioning that the monthly gauge-calibrated IMERG product yields better detection capability (higher probability of detection (POD) values) of daily precipitation events than the daily gauge calibrated GSMaP product; however, they both show similar performance in terms of false alarm ratio (FAR) and critical success index (CSI). Assessment based on extreme precipitation indices revealed that the IMERG product outperforms GSMaP in capturing daily precipitation extremes (RX1Day and RX5Day). In contrast, the GSMaP product tends to be more consistent in capturing the duration and threshold-based precipitation extremes (consecutive dry days (CDD), consecutive wet days (CWD), number of heavy precipitation days (R10mm), and number of extreme precipitation days (R25mm)). Therefore, it is suggested that the IMERG product can be a good alternative for monitoring daily extremes; meanwhile, GSMaP could be a better option for duration-based extremes in the mountainous region.

show abstract

Machine Learning-Based Error Modeling to Improve GPM IMERG Precipitation Product over the Brahmaputra River Basin

Cited by 53 publications

References 85 publications

Assessment of Merged Satellite Precipitation Datasets in Monitoring Meteorological Drought over Pakistan

Assessment of Merged Satellite Precipitation Datasets in Monitoring Meteorological Drought over Pakistan

Investigating Various Products of IMERG for Precipitation Retrieval over Surfaces with and without Snow and Ice Cover

Assessment of GPM-Era Satellite Products’ (IMERG and GSMaP) Ability to Detect Precipitation Extremes over Mountainous Country Nepal

Contact Info

Product

Resources

About