Accuracy of satellite and reanalysis rainfall estimates over Africa: a multi-scale assessment of eight products for continental applications

Study Region: Continental Africa Study Focus: This study evaluates the accuracy of eight gauge-corrected rainfall products across Africa through direct comparisons with in situ observations for the period 2001–2020. The effect of validation datasets on the performance of the rainfall products was al...

Full description

Bibliographic Details
Main Authors: Mekonnen, Kirubel, Velpuri, Naga Manohar, Leh, Mansoor, Akpoti, Komlavi, Owusu, Afua, Tinonetsana, Primrose, Hamouda, T., Ghansah, B., Paranamana, Thilina Prabhath, Munzimi, Y.
Format: Journal Article
Language:Inglés
Published: Elsevier 2023
Subjects:
Online Access:https://hdl.handle.net/10568/131697
_version_ 1855539236307992576
author Mekonnen, Kirubel
Velpuri, Naga Manohar
Leh, Mansoor
Akpoti, Komlavi
Owusu, Afua
Tinonetsana, Primrose
Hamouda, T.
Ghansah, B.
Paranamana, Thilina Prabhath
Munzimi, Y.
author_browse Akpoti, Komlavi
Ghansah, B.
Hamouda, T.
Leh, Mansoor
Mekonnen, Kirubel
Munzimi, Y.
Owusu, Afua
Paranamana, Thilina Prabhath
Tinonetsana, Primrose
Velpuri, Naga Manohar
author_facet Mekonnen, Kirubel
Velpuri, Naga Manohar
Leh, Mansoor
Akpoti, Komlavi
Owusu, Afua
Tinonetsana, Primrose
Hamouda, T.
Ghansah, B.
Paranamana, Thilina Prabhath
Munzimi, Y.
author_sort Mekonnen, Kirubel
collection Repository of Agricultural Research Outputs (CGSpace)
description Study Region: Continental Africa Study Focus: This study evaluates the accuracy of eight gauge-corrected rainfall products across Africa through direct comparisons with in situ observations for the period 2001–2020. The effect of validation datasets on the performance of the rainfall products was also quantified in ten African countries. Four categorical and five continuous metrics were estimated at multiple spatial and temporal scales as part of the evaluation. New hydrological insights for the Region: Results indicate that the performance of the rainfall products varied in space and time. Evaluation at temporal scales revealed that, on average, most rainfall products showed poor results (KGE < 0.35) at the daily timescale. In contrast, RFE v2.0, ARC v2.0, and MSWEP v2.8 were reliable (KGE > 0.75) at the monthly and annual timescales. Among the rainfall products, the performance of TAMSATv3.1, PERSIANN-CDR, and ERA 5 was relatively poor in capturing in situ observations. Evaluation at various spatial scales revealed mixed results. The ARC v2.0 and CHIRPS v2.0 rainfall products were reliable in detecting no rains (< 1 mm/day) for all 19 spatial scales, indicating a high level of confidence for drought studies. IMERG-F v6B and RFE v2.0 were reliable in detecting heavy and high-intensity rainfall events for all spatial scales. Using the KGE performance metrics at the regional level, MSWEP v2.8 in the Northern Africa region, RFE v2.0 in the Western and Southern Africa regions, ARC v2.0 in Central Africa, and CHIRPS v2.0 in the Eastern Africa region showed better performances at monthly timescale. Moreover, the performance of the gauge-corrected rainfall datasets was reduced when compared with independent validation data (gauge data not used by rainfall products) than dependent validation data. This study provides several new insights into choosing a rainfall product for continental to regional applications and identifies the need for bias correction.
format Journal Article
id CGSpace131697
institution CGIAR Consortium
language Inglés
publishDate 2023
publishDateRange 2023
publishDateSort 2023
publisher Elsevier
publisherStr Elsevier
record_format dspace
spelling CGSpace1316972025-12-08T10:11:39Z Accuracy of satellite and reanalysis rainfall estimates over Africa: a multi-scale assessment of eight products for continental applications Mekonnen, Kirubel Velpuri, Naga Manohar Leh, Mansoor Akpoti, Komlavi Owusu, Afua Tinonetsana, Primrose Hamouda, T. Ghansah, B. Paranamana, Thilina Prabhath Munzimi, Y. rainfall estimation satellites datasets models performance assessment evaluation climatology river basins climatic zones rain gauges observation Study Region: Continental Africa Study Focus: This study evaluates the accuracy of eight gauge-corrected rainfall products across Africa through direct comparisons with in situ observations for the period 2001–2020. The effect of validation datasets on the performance of the rainfall products was also quantified in ten African countries. Four categorical and five continuous metrics were estimated at multiple spatial and temporal scales as part of the evaluation. New hydrological insights for the Region: Results indicate that the performance of the rainfall products varied in space and time. Evaluation at temporal scales revealed that, on average, most rainfall products showed poor results (KGE < 0.35) at the daily timescale. In contrast, RFE v2.0, ARC v2.0, and MSWEP v2.8 were reliable (KGE > 0.75) at the monthly and annual timescales. Among the rainfall products, the performance of TAMSATv3.1, PERSIANN-CDR, and ERA 5 was relatively poor in capturing in situ observations. Evaluation at various spatial scales revealed mixed results. The ARC v2.0 and CHIRPS v2.0 rainfall products were reliable in detecting no rains (< 1 mm/day) for all 19 spatial scales, indicating a high level of confidence for drought studies. IMERG-F v6B and RFE v2.0 were reliable in detecting heavy and high-intensity rainfall events for all spatial scales. Using the KGE performance metrics at the regional level, MSWEP v2.8 in the Northern Africa region, RFE v2.0 in the Western and Southern Africa regions, ARC v2.0 in Central Africa, and CHIRPS v2.0 in the Eastern Africa region showed better performances at monthly timescale. Moreover, the performance of the gauge-corrected rainfall datasets was reduced when compared with independent validation data (gauge data not used by rainfall products) than dependent validation data. This study provides several new insights into choosing a rainfall product for continental to regional applications and identifies the need for bias correction. 2023-10 2023-08-31T15:51:37Z 2023-08-31T15:51:37Z Journal Article https://hdl.handle.net/10568/131697 en Open Access Elsevier Mekonnen, Kirubel; Velpuri, Naga Manohar; Leh, Mansoor; Akpoti, Komlavi; Owusu, Afua; Tinonetsana, Primrose; Hamouda, T.; Ghansah, B.; Paranamana, Thilina Prabhath; Munzimi, Y. 2023. Accuracy of satellite and reanalysis rainfall estimates over Africa: a multi-scale assessment of eight products for continental applications. Journal of Hydrology: Regional Studies, 49:101514. [doi: https://doi.org/10.1016/j.ejrh.2023.101514]
spellingShingle rainfall
estimation
satellites
datasets
models
performance assessment
evaluation
climatology
river basins
climatic zones
rain gauges
observation
Mekonnen, Kirubel
Velpuri, Naga Manohar
Leh, Mansoor
Akpoti, Komlavi
Owusu, Afua
Tinonetsana, Primrose
Hamouda, T.
Ghansah, B.
Paranamana, Thilina Prabhath
Munzimi, Y.
Accuracy of satellite and reanalysis rainfall estimates over Africa: a multi-scale assessment of eight products for continental applications
title Accuracy of satellite and reanalysis rainfall estimates over Africa: a multi-scale assessment of eight products for continental applications
title_full Accuracy of satellite and reanalysis rainfall estimates over Africa: a multi-scale assessment of eight products for continental applications
title_fullStr Accuracy of satellite and reanalysis rainfall estimates over Africa: a multi-scale assessment of eight products for continental applications
title_full_unstemmed Accuracy of satellite and reanalysis rainfall estimates over Africa: a multi-scale assessment of eight products for continental applications
title_short Accuracy of satellite and reanalysis rainfall estimates over Africa: a multi-scale assessment of eight products for continental applications
title_sort accuracy of satellite and reanalysis rainfall estimates over africa a multi scale assessment of eight products for continental applications
topic rainfall
estimation
satellites
datasets
models
performance assessment
evaluation
climatology
river basins
climatic zones
rain gauges
observation
url https://hdl.handle.net/10568/131697
work_keys_str_mv AT mekonnenkirubel accuracyofsatelliteandreanalysisrainfallestimatesoverafricaamultiscaleassessmentofeightproductsforcontinentalapplications
AT velpurinagamanohar accuracyofsatelliteandreanalysisrainfallestimatesoverafricaamultiscaleassessmentofeightproductsforcontinentalapplications
AT lehmansoor accuracyofsatelliteandreanalysisrainfallestimatesoverafricaamultiscaleassessmentofeightproductsforcontinentalapplications
AT akpotikomlavi accuracyofsatelliteandreanalysisrainfallestimatesoverafricaamultiscaleassessmentofeightproductsforcontinentalapplications
AT owusuafua accuracyofsatelliteandreanalysisrainfallestimatesoverafricaamultiscaleassessmentofeightproductsforcontinentalapplications
AT tinonetsanaprimrose accuracyofsatelliteandreanalysisrainfallestimatesoverafricaamultiscaleassessmentofeightproductsforcontinentalapplications
AT hamoudat accuracyofsatelliteandreanalysisrainfallestimatesoverafricaamultiscaleassessmentofeightproductsforcontinentalapplications
AT ghansahb accuracyofsatelliteandreanalysisrainfallestimatesoverafricaamultiscaleassessmentofeightproductsforcontinentalapplications
AT paranamanathilinaprabhath accuracyofsatelliteandreanalysisrainfallestimatesoverafricaamultiscaleassessmentofeightproductsforcontinentalapplications
AT munzimiy accuracyofsatelliteandreanalysisrainfallestimatesoverafricaamultiscaleassessmentofeightproductsforcontinentalapplications