Rainfall Forecast Using Machine Learning with High Spatiotemporal Satellite Imagery Every 10 Minutes

Simanjuntak, Febryanto; Jamaluddin, Ilham; Lin, Tang-Huang; Siahaan, Hary Aprianto Wijaya; Dang, Yingnong

doi:10.3390/rs14235950

Cited by 6 publications

(4 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the precipitation from 5 mm to 10 mm, both the forecasted IMERG and gauge rainfall overestimate the raw rainfall, which may be due to the characteristic that forecasted precipitation tends to overestimate the very low precipitation (<5 mm). This result agrees well with that derived from a previous rainfall forecast study using IMERG data combined with machine learning [15]. For precipitation larger than 10 mm, both the forecasted IMERG and gauge rainfall underestimate the raw rainfall, except for the forecasted gauge rainfall at 6 h lead time.…”

Section: Forecasted Imerg Heavy Rainfall At Sub-daily and Daily Lead ...supporting

confidence: 90%

See 1 more Smart Citation

Applicability Assessment of GPM IMERG Satellite Heavy-Rainfall-Informed Reservoir Short-Term Inflow Forecast and Optimal Operation: A Case Study of Wan’an Reservoir in China

Ma,

Gui,

Xiong

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Satellite precipitation estimate (SPE) dedicated to reservoir inflow forecasting is very attractive as it can provide near-real-time information for reservoir monitoring. However, the potential of SPE retrievals with fine temporal resolution in supporting the high-quality pluvial flood inflow forecast and robust short-term operation of a reservoir remains unclear. In this study, the hydrological applicability of half-hourly Integrated Multisatellite Retrievals for Global Precipitation Measurement (GPM IMERG) heavy rainfall data was explored using a synthetic experiment of flood inflow forecast at sub-daily to daily lead times and resultant reservoir short-term operation. The event-based flood forecast was implemented via the rainfall–runoff model GR4H driven by the forecasted IMERG. Then, inflow forecast-informed reservoir multi-objective optimal operation was conducted via a numerical reservoir system and assessed by the risk-based robustness indices encompassing reliability, resilience, vulnerability for water supply, and flood risk ratio for flood prevention. Selecting the Wan’an reservoir located in eastern China as the test case, the results show that the flood forecast forced with IMERG exhibits slightly lower accuracy than that driven by the gauge rainfall across varying lead times. For a specific robustness index, its trends between IMERG and gauge rainfall inputs are comparable, while its magnitude depends on varying lead times and scale ratios (i.e., the reservoir scale). The pattern that the forecast errors in IMERG increase with the lead time is changed in the resultant inflow forecast series and dynamics in the robustness indices for the optimal operation decision. This indicates that the flood forecast model coupled with reservoir operation system could partly compensate the original SPE errors. Our study highlights the acceptable hydrological applicability of IMERG rainfall towards reservoir inflow forecast for robust operation, despite the intrinsic error in SPE.

show abstract

Section: Forecasted Imerg Heavy Rainfall At Sub-daily and Daily Lead ...supporting

confidence: 90%

“…Studies proved that merging SPE products also has benefits for improving the rainfall forecast across varying regions and basins. For example, IMERG data were used to increase the rainfall forecast in high spatiotemporal resolution in Indonesia areas [15] and Niger River basin [16].…”

Section: Introductionmentioning

confidence: 99%

Applicability Assessment of GPM IMERG Satellite Heavy-Rainfall-Informed Reservoir Short-Term Inflow Forecast and Optimal Operation: A Case Study of Wan’an Reservoir in China

Ma,

Gui,

Xiong

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Wang et al [4] proposed that the L1 norm constrained mathematical inverse problem coupled with the information entropy signal freedom was used to estimate the precipitation of typhoon "hagibis" from H8/AHI infrared data, and the estimation effect was better than the classical random forest method. Simanjuntak et al [5] carried out precipitation forecast based on 10 minute high space-time satellite image data using multivariate long-term and shortterm memory network and random forest, and achieved good accuracy. So and Shin's [6] research suggests that the relationship between shallow or warm cloud top temperature and surface rainfall may be weak.…”

Section: Introductionmentioning

confidence: 99%

Precipitation estimation by infrared brightness temperature measurement of FengYun-4A imager

Wang

Song

Shen

et al. 2023

First International Conference on Spatial Atmospheric Marine Environmental Optics (SAME 2023)

View full text Add to dashboard Cite

In this paper, based on the infrared channel brightness temperature from the advanced geosynchronous radiation imager (AGRI) of FengYun-4A satellite, the research on quantitative estimation of precipitation is carried out. The algorithm of precipitation estimation can be divided into three steps. Firstly, the dictionary of brightness temperature of FY-4A/AGRI infrared channel brightness temperature and the integrated multi-satellite retrievals for GPM (IMERG) precipitation is constructed as the historical training sample library. Secondly, the precipitation FOVs are identified. As prior information, the IMERG and ice cloud products are coupled to classification models of the K-nearest neighbor (KNN) and random forest to determine whether there is precipitation at the FOV to be estimated. Finally, the precipitation estimation is performed. Inverse problem regularization method and random forest regression model are used for precipitation estimation, respectively. On this basis, the preliminary experiments for precipitation estimation of and "Ampil (2018)" are carried out. The results show that the precipitation estimation accuracy with ice cloud products as prior information through the inverse problem regularization is better than that with the IMERG products as priori information, while the conclusion is the opposite for the random forest method. The accuracy of precipitation estimation based on the random forest method is better than that of the inverse problem regularization, especially in the "extreme" precipitation center.

show abstract

“…In water resources management, GPM IMERG has often been used globally as an alternative to predicting ground station rainfall data, such as by [1], [9], [10], [11], and [12]. In Indonesia, [13], [14], [15], [16], [17], [18], [19], and [20] have also conducted the same study. Especially for the island of Sumatra, [21], [22], [23], and [24] showed that GPM IMERG is one of the possible rainfall satellite data that can be used in imitating the ground station data.…”

Section: Introductionmentioning

confidence: 99%

The Evaluation of GPM IMERG v.06 Rainfall Product over the Lau Simeme Watershed in Indonesia

Hutagaol¹,

Ball²,

Suhartanto³

et al. 2023

CIVENSE

View full text Add to dashboard Cite

In Indonesia, rainfall is still significant spatially and temporally. In order to gain optimal results from utilising water resources, we have to ensure that the precipitation data is provided in good quality and quantity. Several spatial rainfall measurement sources have become available in recent years, such as GPM data (Global Precipitation Measure). In this study, the GPM IMERG V.06 product was evaluated using rain gauge measurements in the Lau Simeme watershed in North Sumatra Province, Indonesia. The relevance of the GPM IMERG was tested by direct comparison with observations at different time scales (daily, monthly, annual and seasonal) between 2005 and 2019. Results show that the satellite product provides poor rainfall estimations at the daily and annual time scales. However, the accuracy of GPM IMERG Final datasets is improved when temporally average to monthly timescale (R2 of 0.728, RMSE of 68.318 mm and NSE of 0.725), wet seasonal time scale (R2 of 0.673, RMSE of 79.287 mm and NSE of 0.658) and dry seasonal time scale (R2 of 0.947, RMSE of 20.356 mm and NSE of 0.924).

show abstract

Rainfall Forecast Using Machine Learning with High Spatiotemporal Satellite Imagery Every 10 Minutes

Cited by 6 publications

References 57 publications

Applicability Assessment of GPM IMERG Satellite Heavy-Rainfall-Informed Reservoir Short-Term Inflow Forecast and Optimal Operation: A Case Study of Wan’an Reservoir in China

Applicability Assessment of GPM IMERG Satellite Heavy-Rainfall-Informed Reservoir Short-Term Inflow Forecast and Optimal Operation: A Case Study of Wan’an Reservoir in China

Precipitation estimation by infrared brightness temperature measurement of FengYun-4A imager

The Evaluation of GPM IMERG v.06 Rainfall Product over the Lau Simeme Watershed in Indonesia

Contact Info

Product

Resources

About