Evaluation of CMIP6 Historical Simulations over IGAD region of Eastern Africa

Evaluation of CMIP6 Historical Simulations over IGAD region of Eastern Africa | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Evaluation of CMIP6 Historical Simulations over IGAD region of Eastern Africa Paulino Omoj Omay, Nzioka J. Muthama, Christopher Oludhe, Josiah M. Kinama, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-2747422/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Accuracy of model’s simulations are critical for climate change and its socio-economic impact. In this study, we evaluated 23 Global climate models participating in the Coupled Model Intercomparison Project phase 6 (CMIP6). The main objective was to identify top 10 best performance models in capturing patterns of rainfall for the 1981–2014 period over the Intergovernmental Authority on Development (IGAD) region of Eastern Africa. The total rainfall, annual cycle, continuous, categorical and Volumatic statistical metrics, scatter plots, Cumulative Distribution Function (CDF) and colored code portrait were used to assess the patterns of total rainfall. Results indicate that most CMIP6 models generally capture the characteristics of the observed climatology pattern of total rainfall, bimodal and unimodal rainfall regimes. The majority of models over Arid and Semi-Arid Lands (ASALs) in Kenya, Somalia, Ethiopia and Sudan scored lowest skills, highest bias and over-estimated rainfall. In addition, 21 out of 23 CMIP6 over-estimated rainfall over most parts of the region. The ACCESS-ESM1-5 and MIROC6 are the most over-estimated models opposed to CNRM-CM6-1HR as the most model under-estimated rainfall, highest bias and RMSE values. The regional and sub-national analysis showed, it is inconclusive to select best performed models based on individual metric. Out of 23 models, the INM-CM5-0, HadGEM3-GC31-MM, CMCC-CM2-HR4, IPSL-CM6A-LR, KACE-1-0-G, EC-Earth3, NorESM2-MM, GFDL-ESM4, TaiESM1 and KIOST-ESM are the best 10 performance models over IGAD region. These findings highlight the importance of selecting best performance models for mapping present and future hotspots and extreme rainfall events over the IGAD region of Eastern Africa. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 1. Introduction Anthropogenic activities have been considered as the main driver of change in climate over many parts of the world (Taylor et al.,2017). The Intergovernmental Panel on Climate Change (IPCC) six Assessment Report(AR) showed that the frequency and intensity of extreme events increased in recent years over many regions around the world due to changes in physical characteristics of drivers within climate systems(Seneviratne et al., 2021 ). Climate models are one of tools used to simulate past climate conditions (Otieno and Anyah, 2013 b), present and projection of future under different scenarios (Solomon et al.,2019). The results from these different models vary from region, country, meteorological variables or even if applied on specific location due to variation in models initial and boundary conditions (Wang et al., 2012 ). Hence, it is critical to validate the skills of these models before use in impact studies, application in a specific sector or region(Gebresellase et al.,2022). Several researchers around the world(IPCC, 2013 ),Africa(Almazroui et al.,2020), Asia(Dong et al., 2016 ), Europe(Vrac et al., 2022 ), America and the Caribbean(Almazroui et al., 2021 ) used models simulation and scenarios to assess the past and future climate signals. Also, these products used extensively at Africa continents sub-regions of West Africa(Sow et al., 2020 ), North Africa(Babaousmail et al., 2021 ), South Africa(Scoccimarro et al., 2013 ) East Africa(Yang et al.,2015). These studies showed variation in skills of different Coupled Model Intercomparison Project Phase3 (CMIP3), CMIP4, CMIP 5 used in AR 3,4 and 5 and recent CMIP6 Global climate models (GCM) simulations(Eyring et al., 2016 ) used in IPCC AR6. Even products downscaled at regional level such the Coordinated Regional Downscaling Experiment (CORDEX) simulations over east Africa(Endris et al., 2013 ) showed variation in performance of models within region, countries and sub-nation levels. All these studies explained the importance of validating the accuracy of General circulation models (GCMs) and Regional Climate Models (RCMs) historical simulations before being applied in assessments of climate change signals at present time and projected climate change. IPCC AR6 noted a significant improvement in CMIP6 outputs compared to previous versions of CMIPs, whether, improvement in higher spatial resolution, parameterization schemes, biogeochemical cycles and physical processes(Li et al.,2021). Despite the progress made on modeling, representations of climate systems, physical parameterizations, high spatial resolution, dynamical downscaling and other forms, the CMIPs and CORDEX simulations still shows bias (Ayugi et al., 2021 ), variation in ability of models to reproduce observed climate variability and changes (Ongoma et al.,2018). In addition, if a model shows a good skill over specific region and location based on CMIP3 and CMIP5 simulations it does not necessarily have similar skills in reproducing the observed patterns using simulations from CMIP6 (Dosio et al., 2021 ). The risks and stresses in the future depend on the quality of current information simulated by models in the present time. In addition, the effectiveness of adaptation and mitigation actions and sustainable planning for the future rely on the ability of GCMs historical outputs to simulate the current patterns of rainfall patterns and extremes events. The analysis on a seasonal basis such as March, April, May (MAM) and June, July, August, September (JJAS) as the main rainy season over IGAD member states are rarely studied using CMIPs simulations. Therefore, the overall goal of this paper is to evaluate the skills of 23 CMIP6 GCMs historical precipitation outputs (rainfall for purpose of this study) in simulating the MAM and JJAS climatology of total rainfall patterns, then to select best 10 performance models to compute Multi Models Ensembles (MME) used in projecting wet and dry days characteristics over IGAD region of East Africa. The rest of the paper is organized as follows: details of data and techniques used are explained in Section 2 while results are presented in Section 3 . Lastly, the summary and discussions are presented in Section 4 . 2. Data And Methodology 2.1 Study area descriptions The study focuses on Intergovernmental Authority on Development (IGAD) member states of Sudan, Eritrea, Djibouti, South Sudan, Ethiopia, Kenya, Somalia and Uganda (Fig. 1 ). The geographical coordinates of the region are latitude 21.4°-51.2°E and Longitude 5 ° -23.2 ° N. The region is characterized by complex topography. The region's elevation varies from an area below sea level over Sudan to highest points of Mount Kenya at 5,199 m as the second highest mountain in Africa after Mount Kilimanjaro (5,895 m) in Tanzania, Rift Valley extended from Kenya to Ethiopia. The IGAD region climate is affected severely by these high elevation landmarks, seasonal movement of intertropical convergence zone (ITCZ) north and southward which is the one of factors determined the variation in four different rainfall seasons such as December, January, February (DJF), March, April, May (MAM), June, July, August, September (JJAS) and October, November, December (OND). Also, many studies show the climate of region influenced by El Nino/Southern Oscillation (Ogallo, 1988, Indeje et al. , 2000, Anyah and Semazzi, 2006,Anyah and Qiu, 2012 ) as well as variability of sea-surface temperature over the Indian Ocean(Williams and Funk, 2011 ). The impacts of different ENSO phases (El Niño and La Niña or neutral) have different impacts over different parts of the region (Clark et al., 2003 , Anyah and Qiu, 2012 ). The variation in climatic zones whether warm deserts or humid highland climates are mainly driven by orography, geography and micro- synoptic systems (Peel, Finlayson, and McMahon, 2007 ). These local effects offer an opportunity or could be a challenge to the CMIP6 simulations to reproduce observed climate conditions over the region. 2.2 Data This study uses 23 daily historical simulations of CMIP6 models that participated in the new Sixth Phase of CMIP6 (Eyring et al., 2016 ).The daily precipitation for historical (1981–2014) simulations obtained from the CMIP6 website ( https://esgf-node.llnl.gov/search/cmip6/ ). The published 53 CMIP6 list and information regarding models ID, institution description, different agencies, countries and nominal resolutions found in from CMIP6 institution values (wcrp-cmip.github.io). These 53 WCRP-CMIP CMIP6 models had gone through scrutiny of availability of datasets in CIP6 database and resolution. The selected 23 CMIP6 simulations, institutions, model names, and resolution information are listed in Table 1 . In this study, first member realization outputs (r1i1p1f1) are uutilized. To overcome the challenges related to insufficient insitu data sets to be gridded for weather and climate studies over the IGAD region and eastern Africa in general(Camberlin and Okoola, 2003 , Su et al., 2008, Dinku et al., 2018 a), the High-resolution Satellite Rainfall Estimates (SRE) products selected for this study are Climate Hazards Group (CHG) Infrared Precipitation with in-situ station (CHIRPS) daily datasets from the University of California at Santa Barbara (UCSB). The CHIRPS is used as reference data due to better performance in IGAD region compared to other SRE data sets ( Kimani et al., 2017 , Cattani et al., 2018 ; Dinku et al., 2018 ; Gebrechorkos et al., 2018 , Ayugi et al., 2021 ). The CHIRPS product is developed at 0.05° spatial resolution at daily, pentadal, dekadal, and monthly temporal resolution and available from 1981 to near present (Funk et al., 2015 ). The data has been used by many researchers in previous studies in the region (Dinku et al., 2018 , Kimani et al., 2017 , Ocen et al., 2021, Ayugi et al., 2021 ). The selected 23 models and CHIRPS datasets were rescaled using from original resolutions to ten-kilometer (0.1 deg) using bilinear interpolation method used by researcher Song and Yan, ( 2022 ) to overcome challenges of differences in resolutions of CMIP6 models and CHRIPS satellite rainfall estimates over IGAD region of Eastern Africa. The Ensemble mean (EnsMean) computed to reduce systematic errors and biases. To compare models’ simulations and CHIRPS, the temporal range datasets of 1981–2014, and the mean value of all pixels over each five potential agricultural areas of Al qadarif state in Sudan, Arsi district in Ethiopia, Upper Nile state in South Sudan, Trans Nzoia County in Kenya and Arua district in Uganda and entire IGAD region were used in validation. These potential agricultural subregions were adapted based on geographical location and agriculture capacity, food security, magnitude of cash and food crops produced in these regions (see Fig. 1 ). In addition, these five regions are an important agricultural area of the IGAD region supplying other parts of the region with food consumed locally or for export. Table 1 List of 23 CMIP6 models in this study and their institutions, model names and spatial resolutions CMIP6 Model Name Institution Country Spatial resolution 1 ACCESS-ESM1-5 CSIRO-BOM Australia 1.9° × 1.2° 2 AWI-CM-1-1-MR AWI USA 0.935°×0.9375° 3 BCC-CSM2-MR BBC China 1.1° × 1.1° 4 CAMS-CSM1-0 CAMS China 1.1° × 1.1° 5 CanESM5 CCCma Canada 2.8° × 2.8° 6 CMCC-CM2-HR4 CMCC Italy 0.942°×1.25° 7 CNRM-CM6-1-HR CNRM-CERFACS France 0.5° × 0.5° 8 E3SM-1-0 E3SM-Project USA 0:94×1:25 9 EC-Earth3 EC-Earth Consortium Europe 0.7° × 0.7° 10 GFDL-ESM4 NOAA-GFDL USA 1.3° × 1° 11 GISS-E2-2-G NASA- GISS USA 1.25°×1.25° 12 HadGEM3-GC31-MM MOHC UK 0.942°×1.25° 13 IITM-ESM IITM India 1.9 × 1.9 14 INM-CM5-0 INM Russia 2° × 1.5° 15 IPSL-CM6A-LR IPSL France 2.5° × 1.3° 16 KACE-1-0-G NIMS-KMA South Korea 1.875°×1.25° 17 KIOST-ESM KIOST Korea 1.875◦ × 1.875◦ 18 MIROC6 JAMSTEC Japan 1.4° × 1.4° 19 MPI-ESM1-2-HR MPI-M Germany 0.9° × 0.9° 20 MRI-ESM2-0 MRI Japan 1.125°×1.125° 21 NorESM2-MM NCC Norway 0.94° × 1.25° 22 TaiESM1 CcliCS Taiwan 1.25◦ × 0.94◦ 23 UKESM1-0-LL MOHC UK 1.9° × 1.3° 2.3 Methodology 2.3.1 Statistical deterministic metrics The climatological mean is used to assess how CMIP6 simulations capture the annual cycles and inter-annual variability patterns. The performance of CMIP6 simulations were analyzed using 20 continuous statistical measures, categorical and volumetric indices computed over all local administrative areas in the IGAD region. The continuous indices are Correlation Coefficient (CC), Percent Bias (Pbias) ratio, Mean Error (ME), Root Mean Squared Error (RMSE), Nash-Sutcliffe Efficiency (NSE). The Categorical indices are Probability of Detection (POD), Probability of False Detection (POFD), False Alarm Ratio (FAR), Critical Success Index (CSI), Heidke Skill Score (HSS). The volumetric indices are Mean Quantile Bias (MQB), Mean Quantile Error (MQE), Volumetric Hit Index (VHI), Quantile Probability of Detection (QPOD), Volumetric False Alarm Ratio (VFAR), Quantile False Alarm Ratio (QFAR), Volumetric Miss Index (VMI), Volumetric Critical Success Index (VCSI). The CC, PBIAS and RMSE continuous indices selected to visualize the spatial patterns of model’s performance relative to CHIRPS v.2.0 reference datasets. The CC used to measure the relationship strength between each 23 individual models and CHIRPS v.2.0 reference datasets. The perfect relationship value of CC is 1.0. The PBIAS measures the model’s simulation values if it is bigger or smaller than observed. The perfect score is 0.0, therefore low-magnitude values indicate accurate model simulation. The positive values show model overestimation, whereas negative values show underestimation. The RMSE measures the differences or errors between models’ simulations and CHIRPS. The optimal value of RMSE is 0.0. The color code portrait, scatter plots, Cumulative Distribution Function (CDF) used to compare accuracy and consistency of 23 CMIP6 models with respect to CHIRPS v.2.0. The 20 continuous, categorical and Volumetric statistical metrics values and ranking the performance of each 23 CMIP6 models over Al qadarif state in Sudan, Arsi zone in Ethiopia, Arua districts in Uganda, Trans Nzioa county in Kenya and entire average area of IGAD region during MAM season and during JJAS season. The comprehensive rating, ranking and selecting best 10 performance CMIP6 to guide parts 2 of this analysis, which is projected patterns of future wet days and dry spells patterns, the CC, PR2, Pbias, ME, MAE, RMSE and NSE was computed over entire averaged areas of IGAD region, then the values scored sorted from highest to lowest performance model and color code used for each model and best 10 models consistently appeared in ranking during MAM and JJAS season selected. For mathematical expression and purpose of this paper, if O is representing CHIRPS rainfall measurements; \({O}^{-}\) representing average of the measurements; S representing CMIP6 simulations; n representing the number of data samples and, If the A, B, C and D represent Hits, false alarm, misses and correct negative respectively, the statistical skill scores are computed based on a likelihood in formulas in Eq. (5) to (8). Then continuous statistical measures (CC, Pbias, ME, MAE, RMSE, NSE, IOA), categorical (POD, POFD, FAR, CSI, and HSS) and volumetric indices (MQB, MQE, VHI, QPOD, VFAR, QFAR, VMI, VCSI) computed using formulas in Eq. (1) to 20) respectively as described in Table 2 below. Table 2 Descriptions of 20 continuous statistical, categorical and volumetric indices used in validation of CMIP6 simulations Name Formulas Perfect Score Continuous indices COR \({CC}_{GS}=\frac{\frac{1}{n}\sum _{i=1}^{n}({O}_{i}-\stackrel{-}{O}\left)\right({S}_{i}- \stackrel{-}{S} )}{\sqrt{\frac{1}{n}\sum _{i=1}^{n}{\left\{\right({O}_{i}-\stackrel{-}{O})}^{2}.\frac{1}{n}\sum _{i=1}^{n}{({S}_{i}-\stackrel{-}{S} )}^{2}\}}}\dots \dots \dots \dots \dots .\left(1\right)\) 1 PBIAS \(\text{P}\text{B}\text{I}\text{A}\text{S}\left(\text{\%}\right)= \frac{\sum _{i=1}^{n}( {S}_{i}- {O}_{i})}{\sum _{i=1}^{n}{O}_{i}}100\%\dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \left(2\right)\) 0 ME \(ME= \frac{1}{N} \sum \left(S-O\right) \dots \dots \dots \dots \dots .\left(3\right)\) 0 MAE \(MAE= \frac{1}{N}\sum _{i=1}^{n}।\left({S}_{i}-{O}_{i}\right)।\dots \dots \dots .\left(4\right)\) 0 RMSE \(RMSE= \sqrt{\sum _{i=1}^{n}\frac{{({s}_{i}-{s}^{-})}^{2}}{n}} \dots \dots \dots ..\left(5\right)\) 0 NSE \(\text{N}\text{S}\text{E}=1- \frac{\sum _{i=1}^{n}{\left({S}_{i} - {O}_{i} \right)}^{2}}{\sum _{i=1}^{n}{\left( {O}_{i} - {O}^{-} \right)}^{2}}\dots \dots \dots \dots .\left(6\right)\) 1 IOA \(IOA= 1-\frac{\sum _{i=1}^{n}{\left({O}_{i}-{S}_{i}\right)}^{2}}{\sum _{i=1}^{n}{(।S}_{i}-{O}^{-}।+{{।O}_{i}-{O}^{-}।)}^{2} } , 0\le d\le 1 \dots \dots .\left(7\right)\) 1 Categorical indices POD \(POD=\frac{A}{A+C} \dots \dots \dots \dots \dots ..\left(8\right)\) 1 POFD \(POFD=\frac{B}{B+D} \dots \dots \dots \dots \dots ..\left(9\right)\) 0 FAR \(FAR= \frac{B}{A+B} \dots \dots \dots \dots \left(10\right)\) 0 CSI \(\text{C}\text{S}\text{I}= \frac{A}{A+B+C}\dots \dots \dots .\left(11\right)\) 1 HSS \(HSS= \frac{2.(A.D-B.C)}{\left(A+C).\left(C+D\right)+\left(A+B\right).(B+D\right)}\dots \dots \dots \dots .\left(12\right)\) 1 volumetric indices MQB \(MQB= \sum _{i=1}^{n}\frac{\left({P}_{S} । {P}_{S}\ge t)-({P}_{O} । {P}_{O}\ge t\right)}{n} \dots \dots \dots \dots \dots \dots ..\left(13\right)\) MQE \(MQE= \sum _{i=1}^{n}\frac{\left({P}_{S} । {P}_{S}\ge t)-({P}_{O} । {P}_{O}\ge t\right)}{n} \dots \dots \dots \dots \dots \dots ..\left(14\right)\) 0 VHI \(VHI= \frac{\sum _{i=1}^{n}\left({S}_{i}\left|({S}_{i=1}>t \& {O}_{i} >t)\right.\right)}{\sum _{i=1}^{n}({S}_{i} \left|({S}_{i}\right.>t))+\sum _{i=1}^{n}({O}_{i}⌊({S}_{i}\le t \& {O}_{i}>t))}\dots \dots \dots \left(15\right)\) 1 QPOD \(QPOD= \frac{ \sum _{i=1}^{n}।\left({P}_{S} । {P}_{S}\right)\ge t{P}_{O}\ge t }{ \sum _{i=1}^{n}। \left({P}_{S}। {P}_{S} \ge t{P}_{O} \ge t \right)+ \sum _{i=1}^{n}। ({P}_{O}। {P}_{S}t \& {O}_{i} >t)\right.\right)}{\sum _{i=1}^{n}({S}_{i} \left|({S}_{i}\right.>t \& {O}_{i} >t))+\sum _{i=1}^{n}({S}_{i}⌊({S}_{i}>t \& {O}_{i}>t))+\sum _{i=1}^{n}\left({S}_{i}\left|({S}_{i=1}>t \& {O}_{i}\le t)\right.\right) }\left(17\right)\) 0 QFAR \(QFAR= \frac{\sum _{i=1}^{n}\sum _{i=1}^{n}।\left({P}_{S} । {P}_{S}\right)\ge t{P}_{O}\ge t }{\sum _{i=1}^{n} \sum _{i=1}^{n}। \left({P}_{S}। {P}_{S} \ge t{P}_{O} \ge t \right)+ \sum _{i=1}^{n}। ({P}_{O}। {P}_{S}\ge t{P}_{O}t\left)\right)}{\sum _{i=1}^{n}\left({S}_{i}। \left({S}_{i}>t\&{O}_{i}>t\right)\right)+\sum _{i=1}^{n}\left({O}_{i}। \left({S}_{i}\le t\&{O}_{i}>t\right)\right)}\dots \dots ..\left(19\right)\) 0 VCSI \(VCSI= \frac{\sum _{i=1}^{n}\left({S}_{i}\left|({S}_{i=1}>t \& {O}_{i} >t)\right.\right)}{\sum _{i=1}^{n}({S}_{i} \left|({S}_{i}\right.>t \& {O}_{i} >t))+\sum _{i=1}^{n}({S}_{i}⌊({S}_{i}>t \& {O}_{i}>t))}\dots \dots \dots \dots ..\left(20\right)\) 1 3. Result And Discussions 3.1 Seasonal rainfall climatology The spatial climatology patterns of total rainfall of 23 CMIP6 historical simulations, ensemble mean (Ensmean) and observation (CHIRPS v2.0) of the 1981–2014 average is presented in Figs. 2 and 3 . Most models successfully reproduce the spatial patterns of total rainfall over highlands of western Ethiopia, western South Sudan, dry conditions over extreme northern parts of Sudan, arid and semi-arid climate over southern, northeastern Kenya, south-eastern Ethiopia, and most parts of Somalia during MAM (Fig. 2 )and JJAS(Fig. 3 ). The majority of Models reproduced highest total rainfall amount (200–600 mm) and 800-1200mm during MAM and JJAS respectively. Total rainfall increased from MAM to JJAS which is associated with the northward movement of ITCZ as main driver of simulated and projected rainfall over the region(Souverijns et al., 2016). Despite the important of MAM for parts of region close to Equator and JJAS seasonal rainfall over northern sector of IGAD region, models are able to capture the highest amount of rainfall over highlands of western Ethiopia, while northern parts of Sudan, Northeastern Kenya, southeastern Ethiopia, central and northern Somalia received lowest amount during MAM and JJAS. These results are in agreement with results from CORDEX Regional Climate Models carried out by Endris et al., ( 2013 b) over East African. Additionally, Majority of models under-estimated MAM and JJAS rainfall over highlands of western Kenya and well reproduced the patterns over costal. The results also reveal that CNRM-CM6-1-HR, GISS-E2-2-G, KIOST-ESM and CAMS-CSM1-0 tend to under-estimated rainfall, contrary to ACCESS-ESM1-5 over-estimated rainfall over South Sudan, central and highland of western Ethiopia and most parts of Uganda during JJAS. The Ensmean well represented MAM and JJAS rainfall compared to individual models, however, the arid and semi-arid regions in Kenya, Ethiopia and Somalia over-estimated the total rainfall over most parts of IGAD region. The total rainfall annual cycle patterns of 23 CMIP6 GCMs, ensemble mean and CHRIPS v2.0 presented in Fig. 4 . The datasets extracted over five potential food and cash crops cultivation areas of Al qadaref state in Sudan, Upper Nile state in South Sudan, Arsi Zone in Ethiopia, Aura district in Uganda and Trans Nzoia County in Kenya. The results show the models well reproduced the annual rainfall cycle, unimodal rainfall regimes over Al qadaref state in Sudan (Fig. 4 a) and upper Nile state (Fig. 4 b), peak in April and May, then July to October over Arsi zone in Ethiopia (Fig. 4 c) and biomodal over Arua districts (Fig. 4 d) and Trans Nzoia County (Fig. 4 e). Annual cycle, explained the importance of MAM and OND seasons for equatorial eastern Africa countries, JJAS for the northern sector of GHA. It is clear that, most models under-estimate peak of rainfall in August over Al qadaref, June-August over Arua, February-May over Arsi, February-September over Trans Nzoia, while over-estimate April-September rainfall over Upper Nile, October-December over Arzi, Arua and Trans Nzoia. On a seasonal timescale, the MAM and JJAS over the Highlands of western Kenya, JJAS over Central Ethiopia and northeastern Sudan under-estimated rainfall, while northern Uganda during MAM and OND, central Ethiopia during OND, northeastern South Sudan during JJAS over-estimated rainfall. The MAM under-estimated cycle pattern in Kenya, JJAS in Sudan, over-estimate during OND season over Kenya and Uganda are consistent with CMIP5 results by Ongoma et al., ( 2018 ), also, in agreement with most CMIP3 simulations(Anyah & Qiu, 2012 ). The ensemble means of the models reproduce an annual cycle to a large extent compared to most individual models. ACCESS-ESM1-5 is only model over-estimates interannual variability from January to December over Al qadarif state, while CNRM-CM6-1-HR failed to simulate annual cycle over Al qadarif state, Uper Nile state, Arsi zone, and Arua district. The results also show the model's patterns confirmed influence of north–south passage of the ITCZ as reported in study by Clark et al., ( 2003 ) and possible influence of ENSO phenomena among others climate drivers during OND rainfall(Ongoma et al., 2019). Seems, the effect of EA–Indian Ocean, –Asian monsoon, atmosphere–ocean–monsoon interactions are the main driver of bias in MAM and OND rains as reported by Yang et al., ( 2015 ) in his study of Annual Cycle Bias over East Africa in CMIP5. Generally, the performance of the CMIP6 models is characterized by low skills in reproducing MAM rainfall patterns compared to peaks in OND season. 3.2 Spatial Patterns Of Statistical Deterministic Metrics The spatial patterns of CC, PBIAS and RMSE statistical deterministic metrics during MAM and JJAS seasons over each local administrative unit in the IGAD region presented in Figs. 5 –7. The patterns of correlation between 23 CMIP6 models and EnsMean simulations relative to CHIRPS v2.0 reference datasets for the period from 1981 to 2014. The results revealed that all individual models during MAM (Figs. 5 a-w), and JJAS seasons (Figs. 5 a-w) and EnsMean (Figs. 5 x) observed the positive correlation over the entire IGAD region. Most parts of the region recorded weak CC values between 0-0.2, while the highest values not exceeding 0.8. Majority of zones in western Ethiopia recorded the highest CC, followed by South Sudan counties and southern parts of Sudan. The ACCESS-ESM1-5, CanESM5, CMCC-CM2-HR4, IPSL-CM6A-LR, MRI-ESM2-0 and NorESM2-MM are the most performed CMIP6 simulations over South Sudan with CC values exceeding 0.4. The majority of models recorded higher correlation over southern and central parts of Somalia compared to northern parts of the country. The majority of sub-counties and Parishes in northern Uganda recorded higher CC compared to central and southern parts. The GFDL-ESM4 performed better than other models over most parts of Uganda (Figs. 5 j). Majority of consistencies in Kenya, zones in central and northern Ethiopia recorded lowest correlation. All 23 models recorded less than 0.1 CC values over all districts in Djibouti, while the districts in northwestern Eritrea recorded better CC compared to other parts of the country. All models recorded lower CC over northern parts of Sudan. Compared to individual models, the EnsMean patterns showed the improved CC values over all parts of the region, with western Ethiopia, southern parts of Sudan and counties in South Sudan showing a remarkable correlation ranging between 0.5–0.8(Figs. 5 x). The Pbias patterns during JJAS showed all 23 models over-estimated rainfall over all wards in southern and northeastern Kenya, southeastern Ethiopia, most districts in Somalia and Djibouti. Again, the ACCESS-ESM1-5 and MIROC6 the two models of over-estimated rainfall over the IGAD region. Also, with exception of ACCESS-ESM1-5 and MIROC6, all zones in highlands of western Ethiopia, all districts in Sudan under-estimated rainfall (Figs. 6 ). The ACCESS-ESM1-5 and MIROC6 are most model over-estimated (Figs. 6 a, r) and CNRM-CM6-1-HR is the most under-estimated rainfall over most parts of the region (Figs. 6 g). The amount of rainfall over-estimated(under-estimated) for both MAM and JJAS exceeded 80mm per season over these areas. The patterns of under-estimation (over-estimation) are closely with seasons and dry or wet climate of the region. For instance, MAM and JJAS which are not main rainy seasons in Sudan and most parts of Kenya observed significant over-estimation of rainfall. The EnsMean simulated lowest Pbias compared to individual models over South Sudan compared to other countries. Generally, the Pbias has reduced over Sudan, Ethiopia and South Sudan, while northeastern Kenya, southeastern Ethiopia and most parts of Somalia over-estimation of rainfall amount which exceeded 70mm per/dekad (Figs. 6 x). 3.3 Regional and sub-national statistical metrics Table 3 illustrates colored code portraits of 23 CMIP6 GCMs historical simulation with respect to CHIRPS-2.0 over Trans Nzioa county in Kenya during MAM and Arsi zone in Ethiopia during JJAS season. Assessment done using 20 different continuous, categorical and Volumatic indexes. The results show it is premature to conclude the best performed models based on limited indexes. In other words, it is not inclusive to selected the best performed models based on continuous indexes such as correlation, bias and RMSE, or categorical indexes (FAR, CSI, HSS) and Volumatic indexes such as VHI, VFAR, VMI, VCSI indexes. The individual models show variation in skills within 20 indexes considered in analysis. Some models performed very well based on continuous indexes and poorly under categorical and Volumatic indexes and vice versa. For example, MIROC6 performed very well using Volumatic indexes and poorly using continuous over Trans Nzoia. Similarly, the EC-Earth3 and CNRM-CM6-1-HR performed poorly under continuous indexes and better under both categorical and Volumatic indexes. The EnsMean scored best skill based on continuous indexes and poorly under Volumatic. The INM-CM5-0, HadGEM3-GC31-MM, CMCC-CM2-HR4, AWI-CM-1-1-MR, TaiESM1, NorESM2-MM, IITM-ESM, IPSL-CM6A-LR, GFDL-ESM4 and EC-Earth3 are the best 10 performed models over Trans Nzoia in Kenya during MAM season. Ranking individual models show INM-CM5-0 scored the best skills based on continuous indexes (Pbias, ME, MAE, RMSE, NSE and IOA), POD and CSI categorical indexes. Again, CMCC-CM2-HR4 scored better based on volumatic indexes (MQB, VHI, VFAR, VMI and VCSI). The colored code portraits of 23 CMIP6 over Arsi zone in Ethiopia during JJAS season shows failed assessment using categorical indexes because all values are the same for all models. Within one categorical or continuous index, there is variation in the way models perform. For instance, the EC-Earth3 performed the best using correlation and poorly under PBIAS, ME, MAE, RMSE. The ACCESS-ESM1-5, EC-Earth3, IPSL-CM6A-LR, MIROC6 and MRI-ESM2-0 performed well under categorical and Volumatic and poorly under continuous indexes. The BCC-CSM2-MR, INM-CM5-0, CAMS-CSM1-0, HadGEM3-GC31-MM, GFDL-ESM4, KACE-1-0-G, UKESM1-0-LL, NorESM2-MM, AWI-CM-1-1-MR and MPI-ESM1-2-HR are the best 10 performed models over Arsi in Ethiopia during. Again, the EnsMean scored best skill based on continuous indexes and poorly under Volumatic during JJAS over Arsi. It is worthy to mention that the validation using continuous indexes are much better and representative compared to categorical and Volumatic indexes. The scatter plots in Figs. 8 compare the 23 CMIP6 simulation, EnsMean and CHIRPS products at dekadal time-scales. The sample of results used are for Arua district in Uganda and El Qadaref in Sudan during MAM during JJAS respectively. There is wide scatter for all 23 CMIP6 products and EnsMean over Arua compared to El Qadaref. The values toward models’ simulations (toward CHIRPS) explained the overestimated(underestimation) of the rainfall patterns. The majority of models overestimated rainfall, especially rainfall values over 50mm.The CNRM-CM6-1-HR and EC-Earth3 are the models that underestimated the rainfall over Arua and El Qadaref. The BCC-CSM2-MR, ACCESS-ESM1-5, CMCC-CM2-HR4, MIROC6, TaiESM1, MRI-ESM2-0 and GFDL-ESM4 are the most models with wider scattered rainfall values of overestimated rainfall over Arua during MAM season. The majority of models are much better performed over El Qadaref state compared to Arua district. The ACCESS-ESM1-5, AWI-CM-1-1-MR and MPI-ESM1-2-HR are the most common models with wide scatter over El Qadaref. The HadGEM3-GC31-MM shows the least scatter and best performed models (rainfall values are in agreement with model simulation). On the other hand, the EnsMean showed substantial difference scatter from individual models which exhibits systematic underestimation of rainfall amount. These due to high variability and number of Models overestimating/underestimating the rainfall amount, especially rainfall values exceeded 50mm. This scatter may be attributed to uncertainty in standard calendar, original resolution and variation associated with CHIRPS satellite estimates. The Cumulative Distribution Function comparing each 23 CMIP6 simulation against CHIRPS during MAM and JJAS seasons presented in Figs. 9. The MAM season is represented by Trans Nzaia in Kenya and the JJAS season by Upper Nile in South Sudan. The results show that CDF the model’s performance varies from each model and rainfall amounts categories (0–50 and 100-150mm). Majority of models over Trans Nzoia show a tight agreement with observation at lower rainfall compared to high amounts. The CMCC-CM2-HR4, BCC-CSM2-MR, IPSL-CM6A-LR, IITM-ESM, NorESM2-MM, MRI-ESM2-0 are the best performing models (shows a tight simulation). The AWI-CM-1-1-MR, CNRM-CM6-1-HR, EC-Earth3, MIROC6, MPI-ESM1-2-HR, UKESM1-0-LL and GISS-E2-2-G are most models that show a loose pattern. The AWI-CM-1-1-MR, CAMS-CSM1-0, CNRM-CM6-1-HR, EC-Earth3, MPI-ESM1-2-HR and UKESM1-0-LL underestimated the rainfall patterns. The performance of all models improved during JJAS in Upper Nile compared to Trans Nzoia, with ACCESS-ESM1-5 and BCC-CSM2-MR as the most model with loose simulation patterns for all rainfall amount density. Majority of models overestimated rainfall patterns over Upper Nile. However, the 23 models have better skills over Upper Nile state compared to Trans Nzoia. The CAMS-CSM1-0, KACE-1-0-G, NorESM2-MM, HadGEM3-GC31-MM, CMCC-CM2-HR4 are most performed models over Upper Nile in South Sudan. Due to variation in models’ performance using 20 continuous, categorical and volumetric indices indexes, scatter plots and CDF. The all 23 models ranked based on perfect values of CC, BR2, Pbias, ME, NSE, MAE and RMSE were ranked from 1 to 10 regardless of the position of EnsMean in ranking over entire pixels in the IGAD region. Out of 23 models ranked, the best 10 performance models during MAM and JJAS season were selected from 13 models-maintained appearances in ranking. From this, the best 10 ranked models based on perfects scored values of of CC, BR2, Pbias, ME, NSE, MAE and RMSE are INM-CM5-0, CMCC-CM2-HR4, HadGEM3-GC31-MM, IPSL-CM6A-LR, TaiESM1, IITM-ESM, KIOST-ESM, GFDL-ESM4, CanESM5 and BCC-CSM2-MR during MAM (Figs. 10a). The best 10 perfect scored models during JJAS are EC-Earth3, INM-CM5-0, HadGEM3-GC31-MM, CMCC-CM2-HR4, KACE-1-0-G, IPSL-CM6A-LR, NorESM2-MM, GFDL-ESM4, MRI-ESM2-0 and BCC-CSM2-MR. Therefore, the INM-CM5-0, HadGEM3-GC31-MM, CMCC-CM2-HR4, IPSL-CM6A-LR, KACE-1-0-G, EC-Earth3, NorESM2-MM and GFDL-ESM4, TaiESM1 and KIOST-ESM are most ranked 10 models during MAM and JJAS season (Figs. 10b). Therefore, they were selected to compute EnsMean which will be used in computing future patterns of changes and variability in extreme events linked to food security over the IGAD region of Eastern Africa. 4. Summary And Conclusions This paper documents the performances of the 23 CMIP6 GCMs historical rainfall simulations over the IGAD region of Eastern Africa. The spatial patterns of rainfall total validated using 20 continuous, categorical and Volumatic statistical metrics, scatter and CDF plots with respect to CHIRPS-2.0 as reference datasets. The validation was carried out at regional, nation, sub-nation and local administrative levels during 1981–2014. The results showed the majority of CMIP6 GCMs individual models, EnsMean successfully reproduced the spatial patterns of MAM and JJAS total rainfall. The individual models and EnsMean captured the bi-model rainfall regime or peak over highlands of western Kenya and northern Uganda. Furthermore, the spatial patterns of continuous statistical metrics such CC, Pbias and RMSE showed most models scored highest skills over zones in highlands of western Ethiopia, most countries in South Sudan, and southern parts of Sudan. The lowest model’s skill is observed over northeastern and zones in Ethiopia, ASALs counties in Kenya and districts in Somalia. The CNRM-CM6-1HR considered the most model under-estimated total rainfall, failed to simulate annual cycle patterns, highest negative bias and RMSE values, and simulated lowest number of wet days over most parts of study domain. The colored code portrait, scatter and CDF plots of 23 CMIP6 GCMs historical simulations over five potential agricultural areas perceived as food basket of IGAD region suggested it is not sufficient to conclude the best performed models based on limited indexes or individual indexes. The 20 continuous, categorical and Volumatic shows variation within one individual category of index. Some models performed well under continuous and poorly under categorical and Volumatic and vice versa. The scatter plots and CDF explained the magnitude of discrepancies between individual models’ simulations and reference data. The majority of models considered in this study overestimated rainfall amounts over most parts of the IGAD region. Furthermore, except IPSL-CM6A-LR, INM-CM5-0 and MIROC6 during MAM and Ec-earth3, ACCESS-ESM1-5 during, other 20 CMIP6 examined over-estimated total rainfall over most parts of the region. The ranking skills of the 23 CMIP6 GCMs over El qadarif state in Sudan, Arsi district in Ethiopia, Upper Nile state in South Sudan, Trans Nzoia County in Kenya and Arua district in Uganda and entire pixels over IGAD region shows fluctuation in ranking the best 10 performance models within countries and sub-locations. Some models consistently appeared in top 10 best performed models, while others kept in and out of the list. For example, ranked best 10 models over entire spatial average of the IGAD region showed the INM-CM5-0, CMCC-CM2-HR4, HadGEM3-GC31-MM, IPSL-CM6A-LR, TaiESM1, IITM-ESM, KIOST-ESM, GFDL-ESM4, CanESM5 and BCC-CSM2-MR during MAM. Similarly, the EC-Earth3, INM-CM5-0, HadGEM3-GC31-MM, CMCC-CM2-HR4, KACE-1-0-G, IPSL-CM6A-LR, NorESM2-MM, GFDL-ESM4, MRI-ESM2-0 and BCC-CSM2-MR are most ranked 10 models during JJAS season. Therefore, INM-CM5-0, HadGEM3-GC31-MM, CMCC-CM2-HR4, IPSL-CM6A-LR, KACE-1-0-G, EC-Earth3, NorESM2-MM and GFDL-ESM4, TaiESM1 and KIOST-ESM are best 10 models consistently appeared in the top 10. These are the individual models used in computing the Multi-Models Ensemble (MME). The MME will be used in projecting the extremes rainfall events linked to food security under four Shared Socioeconomic Pathway (SSP) scenarios (SSP1-2.6, SSP2‐4.5, SSP3-7.0, and SSP5‐8.5) in ESGF database. Declarations Acknowledgements: This work is part of a PhD work at the Uni­versity of Nairobi, Kenya, Faculty of Science & Technology, Department of Earth & Climate Sciences. Also, author wish to acknowledge World Climate Research Programme (WCRP) Coupled Model Intercomparison Project (Phase 6) and USA, PCMDI/LLNL (California), France, IPSL, Germany, DKRZ and UK, CEDA for archiving and open access to the CMIP6 data Funding : Not applicable, however the authors wish to acknowledge the support by Intra-ACP Climate Services and Related Applications (ClimSA) to IGAD climate Prediction and Application Center (ICPAC) for the funds promised for the open access publication of this research. Declarations: I declare that this paper is my original work and has not been submitted elsewhere for publication. Where other people’s work, or my own work has been used, this has properly been acknowledged and referenced in accordance with the University of Nairobi’s requirements. Conflict of interest: All authors declare no competing interests Consent to participate : All authors consent to participate Consent for publication : All authors consent to publish this work Data Availability: The secondary datasets generated during analysis are available via request. The primary data are open access from cmip6 - Home | ESGF-CoG (llnl.gov) Code availability: We used Climate Data Operators (CDO), R-Packages and Climate Data Tool (CDT) codes Author contribution: All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Paulino Omoj. Omay. The first draft of the manuscript was written by Paulino Omoj. Omay supervised by Nzioka J. Muthama , Christopher Oludhe, Josiah M. Kinama . Final manuscript version reviewed by Guleid Artan and Zachary Atheru . All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. References Almazroui, M., Islam, M. N., Saeed, F., Saeed, S., Ismail, M., Ehsan, M. A., … Barlow, M. (2021). Projected Changes in Temperature and Precipitation Over the United States, Central America, and the Caribbean in CMIP6 GCMs. Earth Systems and Environment , 5 (1), 1–24. https://doi.org/10.1007/s41748-021-00199-5 Almazroui, M., Saeed, F., Saeed, S., Nazrul Islam, M., Ismail, M., Klutse, N. A. B., & Siddiqui, M. H. (2020). Projected Change in Temperature and Precipitation Over Africa from CMIP6. Earth Systems and Environment , 4 (3), 455–475. https://doi.org/10.1007/s41748-020-00161-x Anyah, R. O., & Qiu, W. (2012). Characteristic 20th and 21st century precipitation and temperature patterns and changes over the Greater Horn of Africa. International Journal of Climatology . https://doi.org/10.1002/joc.2270 Ayugi, B., Zhihong, J., Zhu, H., Ngoma, H., Babaousmail, H., Rizwan, K., & Dike, V. (2021). Comparison of CMIP6 and CMIP5 models in simulating mean and extreme precipitation over East Africa. International Journal of Climatology , 41 (15), 6474–6496. https://doi.org/10.1002/joc.7207 Babaousmail, H., Hou, R., Ayugi, B., Ojara, M., Ngoma, H., Karim, R., … Ongoma, V. (2021). Evaluation of the performance of cmip6 models in reproducing rainfall patterns over north africa. Atmosphere , 12 (4), 1–25. https://doi.org/10.3390/atmos12040475 Camberlin, P., & Okoola, R. E. (2003). The onset and cessation of the “long rains” in eastern Africa and their interannual variability. Theoretical and Applied Climatology . https://doi.org/10.1007/s00704-002-0721-5 Cattani, E., Merino, A., Guijarro, J. A., & Levizzani, V. (2018). East Africa Rainfall trends and variability 1983-2015 using three long-term satellite products. Remote Sensing . https://doi.org/10.3390/rs10060931 Clark, C. O., Webster, P. J., & Cole, J. E. (2003). Interdecadal variability of the relationship between the Indian Ocean zonal mode and East African coastal rainfall anomalies. Journal of Climate , 16 (3), 548–554. https://doi.org/10.1175/1520-0442(2003)0162.0.CO;2 Dinku, T., Funk, C., Peterson, P., Maidment, R., Tadesse, T., Gadain, H., & Ceccato, P. (2018). Validation of the CHIRPS satellite rainfall estimates over eastern Africa. Quarterly Journal of the Royal Meteorological Society . https://doi.org/10.1002/qj.3244 Dong, G., Zhang, H., Moise, A., Hanson, L., Liang, P., & Ye, H. (2016). CMIP5 model-simulated onset, duration and intensity of the Asian summer monsoon in current and future climate. Climate Dynamics , 46 (1–2), 355–382. https://doi.org/10.1007/s00382-015-2588-z Dosio, A., Jury, M. W., Almazroui, M., Ashfaq, M., Diallo, I., Engelbrecht, F. A., … Tamoffo, A. T. (2021). Projected future daily characteristics of African precipitation based on global (CMIP5, CMIP6) and regional (CORDEX, CORDEX-CORE) climate models. Climate Dynamics , Vol. 57, pp. 3135–3158. https://doi.org/10.1007/s00382-021-05859-w Endris, H. S., Omondi, P., Jain, S., Lennard, C., Hewitson, B., Chang’a, L., … Tazalika, L. (2013). Assessment of the performance of CORDEX regional climate models in simulating East African rainfall. Journal of Climate . https://doi.org/10.1175/JCLI-D-12-00708.1 Eyring, V., Bony, S., Meehl, G. A., Senior, C. A., Stevens, B., Stouffer, R. J., & Taylor, K. E. (2016). Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. Geoscientific Model Development . https://doi.org/10.5194/gmd-9-1937-2016 Funk, C., Peterson, P., Landsfeld, M., Pedreros, D., Verdin, J., Shukla, S., … Michaelsen, J. (2015). The climate hazards infrared precipitation with stations - A new environmental record for monitoring extremes. Scientific Data . https://doi.org/10.1038/sdata.2015.66 Gebrechorkos, Solomon H., Hülsmann, S., & Bernhofer, C. (2019). Regional climate projections for impact assessment studies in East Africa. Environmental Research Letters , 14 (4). https://doi.org/10.1088/1748-9326/ab055a Gebrechorkos, Solomon Hailu, Hülsmann, S., & Bernhofer, C. (2018). Evaluation of multiple climate data sources for managing environmental resources in East Africa. Hydrology and Earth System Sciences . https://doi.org/10.5194/hess-22-4547-2018 Gebresellase, S. H., Wu, Z., Xu, H., & Muhammad, W. I. (2022). Evaluation and selection of CMIP6 climate models in Upper Awash Basin (UBA), Ethiopia: Evaluation and selection of CMIP6 climate models in Upper Awash Basin (UBA), Ethiopia. Theoretical and Applied Climatology , 149 (3–4), 1521–1547. https://doi.org/10.1007/s00704-022-04056-x Ipcc. (2013). Working Group I Contribution to the IPCC Fifth Assessment Report, Climate Change 2013: The Physical Science Basis. Ipcc . Kimani, M. W., Hoedjes, J. C. B., & Su, Z. (2017). An assessment of satellite-derived rainfall products relative to ground observations over East Africa. Remote Sensing . https://doi.org/10.3390/rs9050430 Li, Y., Wang, C., & Su, F. (2021). Evaluation of CMIP6 Models over Two Third Pole Subregions with Contrasting Circulation Systems. Journal of Climate , 34 (22), 9133–9152. https://doi.org/10.1175/JCLI-D-21-0214.1 Ongoma, V., Chena, H., & Gaoa, C. (2018). Projected changes in mean rainfall and temperature over east Africa based on CMIP5 models. International Journal of Climatology , 38 (3), 1375–1392. https://doi.org/10.1002/joc.5252 Otieno, V. O., & Anyah, R. O. (2013). CMIP5 simulated climate conditions of the Greater Horn of Africa (GHA). Part II: Projected climate. Climate Dynamics . https://doi.org/10.1007/s00382-013-1694-z Peel, M. C., Finlayson, B. L., & McMahon, T. A. (2007). Updated world map of the Köppen-Geiger climate classification. Hydrology and Earth System Sciences , 11 (5), 1633–1644. https://doi.org/10.5194/hess-11-1633-2007 Scoccimarro, E., Gualdi, S., Bellucci, A., Zampieri, M., & Navarra, A. (2013). Heavy precipitation events in a warmer climate: Results from CMIP5 models. Journal of Climate . https://doi.org/10.1175/JCLI-D-12-00850.1 Seneviratne, S. I., Zhang, X., Adnan, M., Badi, W., Dereczynski, C., Di Luca, A., … Zhou, B. (2021). Weather and Climate Extreme Events in a Changing Climate. In Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change . https://doi.org/10.1017/9781009157896.013 Song, S., & Yan, X. (2022). Evaluation of events of extreme temperature change between neighboring days in CMIP6 models over China. Theoretical and Applied Climatology , 150 (1–2), 53–72. https://doi.org/10.1007/s00704-022-04142-0 Sow, M., Diakhaté, M., Dixon, R. D., Guichard, F., Dieng, D., & Gaye, A. T. (2020). Uncertainties in the annual cycle of rainfall characteristics over West Africa in CMIP5 models. Atmosphere , 11 (2). https://doi.org/10.3390/atmos11020216 Taylor, P. C., Fahey, D., & Doherty, S. (2017). DigitalCommons @ University of Nebraska - Lincoln Physical drivers of climate change . I (January 2017), 73–113. Vrac, M., Thao, S., & Yiou, P. (2022). Changes in temperature–precipitation correlations over Europe: are climate models reliable? Climate Dynamics , (2020). https://doi.org/10.1007/s00382-022-06436-5 Wang, C., Wilson, D., Haack, T., Clark, P., Lean, H., & Marshall, R. (2012). Effects of initial and boundary conditions of mesoscale models on simulated atmospheric refractivity. Journal of Applied Meteorology and Climatology , 51 (1), 115–132. https://doi.org/10.1175/JAMC-D-11-012.1 Williams, A. P., & Funk, C. (2011). A westward extension of the warm pool leads to a westward extension of the Walker circulation, drying eastern Africa. Climate Dynamics , 37 (11–12), 2417–2435. https://doi.org/10.1007/s00382-010-0984-y Yang, W., Seager, R., Cane, M. A., & Lyon, B. (2015). The rainfall annual cycle bias over East Africa in CMIP5 coupled climate models. Journal of Climate . https://doi.org/10.1175/JCLI-D-15-0323.1 Table Table 3 is available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files Table3.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-2747422","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":187201866,"identity":"112445a3-8a61-4563-a8aa-824539cea862","order_by":0,"name":"Paulino Omoj Omay","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA30lEQVRIiWNgGAWjYHACxgMJUMYDIMHDR4wemBZmA5AWNqK0QGk2CTBJSLk5+xmDAw932Mjptp8xq/yaYyfDxsD88NENPFose3IMDiSeSTM2O5Njdlt2WzLQYWzGxjl4tBgcAGlpO5y47QBQi+Q2ZqAWHjZpvFrOv4FqOf/GrFhyWz0RWm7AbLmRY8b4cdthwlosZzwrAGoB+uXGs2Jpxm3HediYCfjFnD9548OfbTZyZueTN378ua3anp+9+eFjvA5DMDkMmHlANDMe5Wha2B8w/iCgehSMglEwCkYmAADB/kwK5uMgBgAAAABJRU5ErkJggg==","orcid":"","institution":"University of Nairobi","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Paulino","middleName":"Omoj","lastName":"Omay","suffix":""},{"id":187201867,"identity":"3e28d64c-d935-450c-9752-3f92e63f834f","order_by":1,"name":"Nzioka J. Muthama","email":"","orcid":"","institution":"University of Nairobi","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Nzioka","middleName":"J.","lastName":"Muthama","suffix":""},{"id":187201868,"identity":"c03eadea-42c9-4a28-800d-02c76ab18bac","order_by":2,"name":"Christopher Oludhe","email":"","orcid":"","institution":"University of Nairobi","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Christopher","middleName":"","lastName":"Oludhe","suffix":""},{"id":187201869,"identity":"cb7308eb-aacc-464d-8e5c-be1d4df9d18d","order_by":3,"name":"Josiah M. Kinama","email":"","orcid":"","institution":"University of Nairobi","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Josiah","middleName":"M.","lastName":"Kinama","suffix":""},{"id":187201870,"identity":"837919ae-8f21-4722-9938-ecdbafb22b32","order_by":4,"name":"Guleid Artan","email":"","orcid":"","institution":"IGAD Climate Prediction and Applications Center","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Guleid","middleName":"","lastName":"Artan","suffix":""},{"id":187201871,"identity":"81875487-50ff-4515-bb43-5ed593e9e03c","order_by":5,"name":"Zachary Atheru","email":"","orcid":"","institution":"IGAD Climate Prediction and Applications Center","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zachary","middleName":"","lastName":"Atheru","suffix":""}],"badges":[],"createdAt":"2023-03-28 13:59:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-2747422/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-2747422/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":35046879,"identity":"2f557b59-f629-4d34-a200-79716e5603d3","added_by":"auto","created_at":"2023-03-30 15:13:37","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":568506,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eElevation map of the IGAD region of Eastern Africa, the sky-blue areas indicate the potential agricultural areas used to validate the Models. The purple points indicate local administrative areas used to compute and visualize the spatial patterns of correlation Coefficient (CC), Percent of bias (Pbias) and Root Mean Square Error (RMSE).\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-2747422/v1/8c615544b8a7909018c77297.png"},{"id":35047796,"identity":"a9636338-4f68-4e06-8903-828757a3cd0a","added_by":"auto","created_at":"2023-03-30 15:21:37","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":442243,"visible":true,"origin":"","legend":"\u003cp\u003eClimatology of rainfall in eastern Africa during MAM as simulated by 23 CMIP6 historical run, ensemble mean (Ensmean) and observation (CHIRPS v2.0) in simulating total rainfall climatology patterns relative to 1981–2014 reference period\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-2747422/v1/2553c118ca5e5c6bb9622f2a.png"},{"id":35047792,"identity":"f8e39c90-0fb8-4d00-8a41-81f12d0f968b","added_by":"auto","created_at":"2023-03-30 15:21:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":431254,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eClimatology of rainfall in eastern Africa during JJAS as simulated by 23 CMIP6 historical run, ensemble mean (Ensmean) and observation (CHIRPS v2.0) in simulating total rainfall climatology patterns relative to 1981–2014 reference period\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-2747422/v1/81f5ecacaaaa7aca34e806d1.png"},{"id":35046882,"identity":"aa943b31-5990-4c20-ab85-d5ca95208f93","added_by":"auto","created_at":"2023-03-30 15:13:37","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":324416,"visible":true,"origin":"","legend":"\u003cp\u003eAnnual rainfall cycle over East Africa for 1981–2014 based on observations (CHIRPS) and 23 CMIP 6 historical simulations averaged food and cash crops cultivation areas of over (a) Al qadaref, (b) Upper Nile, (c) Arsi Zone in Ethiopia, (d) Aura district in Uganda, (e) Trans Nzoia County in Kenya\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-2747422/v1/4dd3f93e3663a831ef73ed14.png"},{"id":35047794,"identity":"d29acac1-84bb-4cbd-9b32-15129be57bbe","added_by":"auto","created_at":"2023-03-30 15:21:37","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":838754,"visible":true,"origin":"","legend":"\u003cp\u003eSpatial patterns of correlation coefficient (CC) values (mm per season) of 23 models and EnsMean rainfall with respect to CHIRPS v.2.0 during MAM season over each local administrative unit in IGAD region for 1981–2014\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-2747422/v1/5cb0e7b9d000606828b38716.png"},{"id":35046886,"identity":"26aa1742-c562-4612-b9d5-a2cf1affd6e0","added_by":"auto","created_at":"2023-03-30 15:13:37","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1103515,"visible":true,"origin":"","legend":"\u003cp\u003eSpatial distribution of Percent of Bias (PBIAS) climatology of precipitation (mm_season−1) of 23 models and EnsMean rainfall with respect to CHIRPS v.2.0 during JJAS season over each local administrative unit in IGAD region for 1981–2014\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-2747422/v1/69dfa095b23265e9b3e1aa7f.png"},{"id":35048334,"identity":"b0a6daca-88a8-489e-a4bb-62d4de451168","added_by":"auto","created_at":"2023-03-30 15:29:37","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":865034,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eSpatial distribution of RMSE climatology of precipitation (mm_season−1) of 23 models and EnsMean rainfall with respect to CHIRPS v.2.0 during MAM season \u003c/em\u003eover each local administrative unit in IGAD region \u003cem\u003efor 1981–2014\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-2747422/v1/6655e295aad822f6ae02e316.png"},{"id":35046888,"identity":"15a879c0-0a09-4b18-8431-64b9ab7afc41","added_by":"auto","created_at":"2023-03-30 15:13:38","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":277582,"visible":true,"origin":"","legend":"\u003cp\u003eScatter plots comparing each 23 CMIP6 simulation against CHIRPS v2.0 area-average of all pixels in Arua district in Uganda during MAM and El Qadaref in Sudan during JJAS at dekadal time-scale reference to 1981-2014\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-2747422/v1/351f2d71613073caa06b57a0.png"},{"id":35047797,"identity":"09ab5a81-2d53-4da5-a685-f274ce30208d","added_by":"auto","created_at":"2023-03-30 15:21:37","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":232545,"visible":true,"origin":"","legend":"\u003cp\u003eCumulative Distribution Function comparing each 23 CMIP6 simulation against CHIRPS v2.0 area-average of all pixels in Trans Nzaia in Kenya and Upper Nile state in South Sudan at dekadal time-scale reference to1981-2014.\u003c/p\u003e","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-2747422/v1/082ba7fb68695d0c054522b9.png"},{"id":35047798,"identity":"a17b4c94-9293-4ac5-aa75-f214754a485b","added_by":"auto","created_at":"2023-03-30 15:21:38","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":1033703,"visible":true,"origin":"","legend":"\u003cp\u003eSummary of ranking the best 10 performance out of 23 CMIP 6 models with respect to CHIRPS v.2.0 based on 7 continuous indexes values over Entire IGAD region during (a) MAM and (b) JJAS seasons\u003c/p\u003e","description":"","filename":"10.png","url":"https://assets-eu.researchsquare.com/files/rs-2747422/v1/fcac2fde0de27cc4f8ab90a2.png"},{"id":35172114,"identity":"6c44d9ab-1ae0-4aff-9d2b-20f20d362391","added_by":"auto","created_at":"2023-04-03 01:59:36","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6072672,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-2747422/v1/969e66d7-f771-4025-80f4-efc82e3696bd.pdf"},{"id":35048336,"identity":"d3428efc-45ab-47b2-b986-988d75c88710","added_by":"auto","created_at":"2023-03-30 15:29:37","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":494779,"visible":true,"origin":"","legend":"","description":"","filename":"Table3.docx","url":"https://assets-eu.researchsquare.com/files/rs-2747422/v1/38aa569f3f0e1ddfe2dedee2.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Evaluation of CMIP6 Historical Simulations over IGAD region of Eastern Africa","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eAnthropogenic activities have been considered as the main driver of change in climate over many parts of the world (Taylor et al.,2017). The Intergovernmental Panel on Climate Change (IPCC) six Assessment Report(AR) showed that the frequency and intensity of extreme events increased in recent years over many regions around the world due to changes in physical characteristics of drivers within climate systems(Seneviratne et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Climate models are one of tools used to simulate past climate conditions (Otieno and Anyah, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2013\u003c/span\u003eb), present and projection of future under different scenarios (Solomon et al.,2019). The results from these different models vary from region, country, meteorological variables or even if applied on specific location due to variation in models initial and boundary conditions (Wang et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Hence, it is critical to validate the skills of these models before use in impact studies, application in a specific sector or region(Gebresellase et al.,2022).\u003c/p\u003e \u003cp\u003eSeveral researchers around the world(IPCC, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2013\u003c/span\u003e),Africa(Almazroui et al.,2020), Asia(Dong et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), Europe(Vrac et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), America and the Caribbean(Almazroui et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) used models simulation and scenarios to assess the past and future climate signals. Also, these products used extensively at Africa continents sub-regions of West Africa(Sow et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), North Africa(Babaousmail et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), South Africa(Scoccimarro et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) East Africa(Yang et al.,2015). These studies showed variation in skills of different Coupled Model Intercomparison Project Phase3 (CMIP3), CMIP4, CMIP 5 used in AR 3,4 and 5 and recent CMIP6 Global climate models (GCM) simulations(Eyring et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) used in IPCC AR6. Even products downscaled at regional level such the Coordinated Regional Downscaling Experiment (CORDEX) simulations over east Africa(Endris et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) showed variation in performance of models within region, countries and sub-nation levels. All these studies explained the importance of validating the accuracy of General circulation models (GCMs) and Regional Climate Models (RCMs) historical simulations before being applied in assessments of climate change signals at present time and projected climate change.\u003c/p\u003e \u003cp\u003eIPCC AR6 noted a significant improvement in CMIP6 outputs compared to previous versions of CMIPs, whether, improvement in higher spatial resolution, parameterization schemes, biogeochemical cycles and physical processes(Li et al.,2021). Despite the progress made on modeling, representations of climate systems, physical parameterizations, high spatial resolution, dynamical downscaling and other forms, the CMIPs and CORDEX simulations still shows bias (Ayugi et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), variation in ability of models to reproduce observed climate variability and changes (Ongoma et al.,2018). In addition, if a model shows a good skill over specific region and location based on CMIP3 and CMIP5 simulations it does not necessarily have similar skills in reproducing the observed patterns using simulations from CMIP6 (Dosio et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe risks and stresses in the future depend on the quality of current information simulated by models in the present time. In addition, the effectiveness of adaptation and mitigation actions and sustainable planning for the future rely on the ability of GCMs historical outputs to simulate the current patterns of rainfall patterns and extremes events. The analysis on a seasonal basis such as March, April, May (MAM) and June, July, August, September (JJAS) as the main rainy season over IGAD member states are rarely studied using CMIPs simulations. Therefore, the overall goal of this paper is to evaluate the skills of 23 CMIP6 GCMs historical precipitation outputs (rainfall for purpose of this study) in simulating the MAM and JJAS climatology of total rainfall patterns, then to select best 10 performance models to compute Multi Models Ensembles (MME) used in projecting wet and dry days characteristics over IGAD region of East Africa. The rest of the paper is organized as follows: details of data and techniques used are explained in Section \u003cspan refid=\"Sec2\" class=\"InternalRef\"\u003e2\u003c/span\u003e while results are presented in Section \u003cspan refid=\"Sec7\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Lastly, the summary and discussions are presented in Section \u003cspan refid=\"Sec11\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e"},{"header":"2. Data And Methodology","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Study area descriptions\u003c/h2\u003e \u003cp\u003eThe study focuses on Intergovernmental Authority on Development (IGAD) member states of Sudan, Eritrea, Djibouti, South Sudan, Ethiopia, Kenya, Somalia and Uganda (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The geographical coordinates of the region are latitude 21.4\u0026deg;-51.2\u0026deg;E and Longitude 5\u003csup\u003e\u0026deg;\u003c/sup\u003e-23.2\u003csup\u003e\u0026deg;\u003c/sup\u003eN. The region is characterized by complex topography. The region's elevation varies from an area below sea level over Sudan to highest points of Mount Kenya at 5,199 m as the second highest mountain in Africa after Mount Kilimanjaro (5,895 m) in Tanzania, Rift Valley extended from Kenya to Ethiopia.\u003c/p\u003e \u003cp\u003eThe IGAD region climate is affected severely by these high elevation landmarks, seasonal movement of intertropical convergence zone (ITCZ) north and southward which is the one of factors determined the variation in four different rainfall seasons such as December, January, February (DJF), March, April, May (MAM), June, July, August, September (JJAS) and October, November, December (OND). Also, many studies show the climate of region influenced by El Nino/Southern Oscillation (Ogallo, 1988, Indeje \u003cem\u003eet al.\u003c/em\u003e, 2000, Anyah and Semazzi, 2006,Anyah and Qiu, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) as well as variability of sea-surface temperature over the Indian Ocean(Williams and Funk, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). The impacts of different ENSO phases (El Ni\u0026ntilde;o and La Ni\u0026ntilde;a or neutral) have different impacts over different parts of the region (Clark et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2003\u003c/span\u003e, Anyah and Qiu, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). The variation in climatic zones whether warm deserts or humid highland climates are mainly driven by orography, geography and micro- synoptic systems (Peel, Finlayson, and McMahon, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). These local effects offer an opportunity or could be a challenge to the CMIP6 simulations to reproduce observed climate conditions over the region.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003e2.2 Data\u003c/h3\u003e\n\u003cp\u003eThis study uses 23 daily historical simulations of CMIP6 models that participated in the new Sixth Phase of CMIP6 (Eyring et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).The daily precipitation for historical (1981\u0026ndash;2014) simulations obtained from the CMIP6 website (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://esgf-node.llnl.gov/search/cmip6/\u003c/span\u003e\u003cspan address=\"https://esgf-node.llnl.gov/search/cmip6/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The published 53 CMIP6 list and information regarding models ID, institution description, different agencies, countries and nominal resolutions found in from CMIP6 institution values (wcrp-cmip.github.io). These 53 WCRP-CMIP CMIP6 models had gone through scrutiny of availability of datasets in CIP6 database and resolution. The selected 23 CMIP6 simulations, institutions, model names, and resolution information are listed in Table \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. In this study, first member realization outputs (r1i1p1f1) are uutilized. To overcome the challenges related to insufficient insitu data sets to be gridded for weather and climate studies over the IGAD region and eastern Africa in general(Camberlin and Okoola, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2003\u003c/span\u003e, Su et al., 2008, Dinku et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2018\u003c/span\u003ea), the High-resolution Satellite Rainfall Estimates (SRE) products selected for this study are Climate Hazards Group (CHG) Infrared Precipitation with in-situ station (CHIRPS) daily datasets from the University of California at Santa Barbara (UCSB). The CHIRPS is used as reference data due to better performance in IGAD region compared to other SRE data sets ( Kimani et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2017\u003c/span\u003e, Cattani et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Dinku et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Gebrechorkos et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2018\u003c/span\u003e, Ayugi et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The CHIRPS product is developed at 0.05\u0026deg; spatial resolution at daily, pentadal, dekadal, and monthly temporal resolution and available from 1981 to near present (Funk et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). The data has been used by many researchers in previous studies in the region (Dinku et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2018\u003c/span\u003e, Kimani et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2017\u003c/span\u003e, Ocen et al., 2021, Ayugi et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The selected 23 models and CHIRPS datasets were rescaled using from original resolutions to ten-kilometer (0.1 deg) using bilinear interpolation method used by researcher Song and Yan, (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) to overcome challenges of differences in resolutions of CMIP6 models and CHRIPS satellite rainfall estimates over IGAD region of Eastern Africa. The Ensemble mean (EnsMean) computed to reduce systematic errors and biases. To compare models\u0026rsquo; simulations and CHIRPS, the temporal range datasets of 1981\u0026ndash;2014, and the mean value of all pixels over each five potential agricultural areas of Al qadarif state in Sudan, Arsi district in Ethiopia, Upper Nile state in South Sudan, Trans Nzoia County in Kenya and Arua district in Uganda and entire IGAD region were used in validation. These potential agricultural subregions were adapted based on geographical location and agriculture capacity, food security, magnitude of cash and food crops produced in these regions (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). In addition, these five regions are an important agricultural area of the IGAD region supplying other parts of the region with food consumed locally or for export.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eList of 23 CMIP6 models in this study and their institutions, model names and spatial resolutions\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026times;\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCMIP6 Model Name\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eInstitution\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCountry\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSpatial resolution\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eACCESS-ESM1-5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCSIRO-BOM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAustralia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e1.9\u0026deg; \u0026times; 1.2\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAWI-CM-1-1-MR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAWI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eUSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e0.935\u0026deg;\u0026times;0.9375\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBCC-CSM2-MR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBBC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eChina\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e1.1\u0026deg; \u0026times; 1.1\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCAMS-CSM1-0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCAMS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eChina\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e1.1\u0026deg; \u0026times; 1.1\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCanESM5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCCCma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCanada\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e2.8\u0026deg; \u0026times; 2.8\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCMCC-CM2-HR4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCMCC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eItaly\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e0.942\u0026deg;\u0026times;1.25\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCNRM-CM6-1-HR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCNRM-CERFACS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFrance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e0.5\u0026deg; \u0026times; 0.5\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eE3SM-1-0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eE3SM-Project\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eUSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e0:94\u0026times;1:25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEC-Earth3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eEC-Earth Consortium\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eEurope\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e0.7\u0026deg; \u0026times; 0.7\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGFDL-ESM4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNOAA-GFDL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eUSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e1.3\u0026deg; \u0026times; 1\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGISS-E2-2-G\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNASA- GISS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eUSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e1.25\u0026deg;\u0026times;1.25\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHadGEM3-GC31-MM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMOHC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eUK\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e0.942\u0026deg;\u0026times;1.25\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIITM-ESM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIITM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIndia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e1.9 \u0026times; 1.9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eINM-CM5-0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eINM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRussia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e2\u0026deg; \u0026times; 1.5\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIPSL-CM6A-LR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIPSL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFrance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e2.5\u0026deg; \u0026times; 1.3\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eKACE-1-0-G\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNIMS-KMA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSouth Korea\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e1.875\u0026deg;\u0026times;1.25\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eKIOST-ESM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eKIOST\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eKorea\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e1.875◦ \u0026times; 1.875◦\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMIROC6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eJAMSTEC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eJapan\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e1.4\u0026deg; \u0026times; 1.4\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMPI-ESM1-2-HR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMPI-M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGermany\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e0.9\u0026deg; \u0026times; 0.9\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMRI-ESM2-0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMRI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eJapan\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e1.125\u0026deg;\u0026times;1.125\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNorESM2-MM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNCC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNorway\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e0.94\u0026deg; \u0026times; 1.25\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTaiESM1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCcliCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTaiwan\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e1.25◦ \u0026times; 0.94◦\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUKESM1-0-LL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMOHC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eUK\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c5\"\u003e \u003cp\u003e1.9\u0026deg; \u0026times; 1.3\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e\n\u003ch3\u003e2.3 Methodology\u003c/h3\u003e\n\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.3.1 Statistical deterministic metrics\u003c/h2\u003e \u003cp\u003eThe climatological mean is used to assess how CMIP6 simulations capture the annual cycles and inter-annual variability patterns. The performance of CMIP6 simulations were analyzed using 20 continuous statistical measures, categorical and volumetric indices computed over all local administrative areas in the IGAD region. The continuous indices are Correlation Coefficient (CC), Percent Bias (Pbias) ratio, Mean Error (ME), Root Mean Squared Error (RMSE), Nash-Sutcliffe Efficiency (NSE). The Categorical indices are Probability of Detection (POD), Probability of False Detection (POFD), False Alarm Ratio (FAR), Critical Success Index (CSI), Heidke Skill Score (HSS). The volumetric indices are Mean Quantile Bias (MQB), Mean Quantile Error (MQE), Volumetric Hit Index (VHI), Quantile Probability of Detection (QPOD), Volumetric False Alarm Ratio (VFAR), Quantile False Alarm Ratio (QFAR), Volumetric Miss Index (VMI), Volumetric Critical Success Index (VCSI). The CC, PBIAS and RMSE continuous indices selected to visualize the spatial patterns of model\u0026rsquo;s performance relative to CHIRPS v.2.0 reference datasets. The CC used to measure the relationship strength between each 23 individual models and CHIRPS v.2.0 reference datasets. The perfect relationship value of CC is 1.0. The PBIAS measures the model\u0026rsquo;s simulation values if it is bigger or smaller than observed. The perfect score is 0.0, therefore low-magnitude values indicate accurate model simulation. The positive values show model overestimation, whereas negative values show underestimation. The RMSE measures the differences or errors between models\u0026rsquo; simulations and CHIRPS. The optimal value of RMSE is 0.0. The color code portrait, scatter plots, Cumulative Distribution Function (CDF) used to compare accuracy and consistency of 23 CMIP6 models with respect to CHIRPS v.2.0. The 20 continuous, categorical and Volumetric statistical metrics values and ranking the performance of each 23 CMIP6 models over Al qadarif state in Sudan, Arsi zone in Ethiopia, Arua districts in Uganda, Trans Nzioa county in Kenya and entire average area of IGAD region during MAM season and during JJAS season. The comprehensive rating, ranking and selecting best 10 performance CMIP6 to guide parts 2 of this analysis, which is projected patterns of future wet days and dry spells patterns, the CC, PR2, Pbias, ME, MAE, RMSE and NSE was computed over entire averaged areas of IGAD region, then the values scored sorted from highest to lowest performance model and color code used for each model and best 10 models consistently appeared in ranking during MAM and JJAS season selected. For mathematical expression and purpose of this paper, if \u003cem\u003eO\u003c/em\u003e is representing CHIRPS rainfall measurements; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({O}^{-}\\)\u003c/span\u003e\u003c/span\u003e representing average of the measurements; \u003cem\u003eS\u003c/em\u003e representing CMIP6 simulations; \u003cem\u003en\u003c/em\u003e representing the number of data samples and, If the A, B, C and D represent Hits, false alarm, misses and correct negative respectively, the statistical skill scores are computed based on a likelihood in formulas in Eq.\u0026nbsp;(5) to (8). Then continuous statistical measures (CC, Pbias, ME, MAE, RMSE, NSE, IOA), categorical (POD, POFD, FAR, CSI, and HSS) and volumetric indices (MQB, MQE, VHI, QPOD, VFAR, QFAR, VMI, VCSI) computed using formulas in Eq.\u0026nbsp;(1) to 20) respectively as described in Table \u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e below.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescriptions of 20 continuous statistical, categorical and volumetric indices used in validation of CMIP6 simulations\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eName\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFormulas\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePerfect Score\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"5\" rowspan=\"6\"\u003e \u003cp\u003eContinuous indices\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCOR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({CC}_{GS}=\\frac{\\frac{1}{n}\\sum _{i=1}^{n}({O}_{i}-\\stackrel{-}{O}\\left)\\right({S}_{i}- \\stackrel{-}{S} )}{\\sqrt{\\frac{1}{n}\\sum _{i=1}^{n}{\\left\\{\\right({O}_{i}-\\stackrel{-}{O})}^{2}.\\frac{1}{n}\\sum _{i=1}^{n}{({S}_{i}-\\stackrel{-}{S} )}^{2}\\}}}\\dots \\dots \\dots \\dots \\dots .\\left(1\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePBIAS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\text{P}\\text{B}\\text{I}\\text{A}\\text{S}\\left(\\text{\\%}\\right)= \\frac{\\sum _{i=1}^{n}( {S}_{i}- {O}_{i})}{\\sum _{i=1}^{n}{O}_{i}}100\\%\\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\left(2\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eME\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(ME= \\frac{1}{N} \\sum \\left(S-O\\right) \\dots \\dots \\dots \\dots \\dots .\\left(3\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMAE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(MAE= \\frac{1}{N}\\sum _{i=1}^{n}।\\left({S}_{i}-{O}_{i}\\right)।\\dots \\dots \\dots .\\left(4\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRMSE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(RMSE= \\sqrt{\\sum _{i=1}^{n}\\frac{{({s}_{i}-{s}^{-})}^{2}}{n}} \\dots \\dots \\dots ..\\left(5\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNSE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\text{N}\\text{S}\\text{E}=1- \\frac{\\sum _{i=1}^{n}{\\left({S}_{i} - {O}_{i} \\right)}^{2}}{\\sum _{i=1}^{n}{\\left( {O}_{i} - {O}^{-} \\right)}^{2}}\\dots \\dots \\dots \\dots .\\left(6\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIOA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(IOA= 1-\\frac{\\sum _{i=1}^{n}{\\left({O}_{i}-{S}_{i}\\right)}^{2}}{\\sum _{i=1}^{n}{(।S}_{i}-{O}^{-}।+{{।O}_{i}-{O}^{-}।)}^{2} } , 0\\le d\\le 1 \\dots \\dots .\\left(7\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003eCategorical indices\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePOD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(POD=\\frac{A}{A+C} \\dots \\dots \\dots \\dots \\dots ..\\left(8\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePOFD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(POFD=\\frac{B}{B+D} \\dots \\dots \\dots \\dots \\dots ..\\left(9\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFAR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(FAR= \\frac{B}{A+B} \\dots \\dots \\dots \\dots \\left(10\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCSI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\text{C}\\text{S}\\text{I}= \\frac{A}{A+B+C}\\dots \\dots \\dots .\\left(11\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHSS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(HSS= \\frac{2.(A.D-B.C)}{\\left(A+C).\\left(C+D\\right)+\\left(A+B\\right).(B+D\\right)}\\dots \\dots \\dots \\dots .\\left(12\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"8\" rowspan=\"9\"\u003e \u003cp\u003evolumetric indices\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMQB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(MQB= \\sum _{i=1}^{n}\\frac{\\left({P}_{S} । {P}_{S}\\ge t)-({P}_{O} । {P}_{O}\\ge t\\right)}{n} \\dots \\dots \\dots \\dots \\dots \\dots ..\\left(13\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMQE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(MQE= \\sum _{i=1}^{n}\\frac{\\left({P}_{S} । {P}_{S}\\ge t)-({P}_{O} । {P}_{O}\\ge t\\right)}{n} \\dots \\dots \\dots \\dots \\dots \\dots ..\\left(14\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVHI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(VHI= \\frac{\\sum _{i=1}^{n}\\left({S}_{i}\\left|({S}_{i=1}\u0026gt;t \\\u0026amp; {O}_{i} \u0026gt;t)\\right.\\right)}{\\sum _{i=1}^{n}({S}_{i} \\left|({S}_{i}\\right.\u0026gt;t))+\\sum _{i=1}^{n}({O}_{i}\u0026lfloor;({S}_{i}\\le t \\\u0026amp; {O}_{i}\u0026gt;t))}\\dots \\dots \\dots \\left(15\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQPOD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(QPOD= \\frac{ \\sum _{i=1}^{n}।\\left({P}_{S} । {P}_{S}\\right)\\ge t{P}_{O}\\ge t }{ \\sum _{i=1}^{n}। \\left({P}_{S}। {P}_{S} \\ge t{P}_{O} \\ge t \\right)+ \\sum _{i=1}^{n}। ({P}_{O}। {P}_{S}\u0026lt;t{P}_{O}\\ge t) }\\dots \\dots \\dots \\dots \\left(16\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVFAR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(VFAR= \\frac{\\sum _{i=1}^{n}\\left({S}_{i}\\left|({S}_{i=1}\u0026gt;t \\\u0026amp; {O}_{i} \u0026gt;t)\\right.\\right)}{\\sum _{i=1}^{n}({S}_{i} \\left|({S}_{i}\\right.\u0026gt;t \\\u0026amp; {O}_{i} \u0026gt;t))+\\sum _{i=1}^{n}({S}_{i}\u0026lfloor;({S}_{i}\u0026gt;t \\\u0026amp; {O}_{i}\u0026gt;t))+\\sum _{i=1}^{n}\\left({S}_{i}\\left|({S}_{i=1}\u0026gt;t \\\u0026amp; {O}_{i}\\le t)\\right.\\right) }\\left(17\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQFAR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(QFAR= \\frac{\\sum _{i=1}^{n}\\sum _{i=1}^{n}।\\left({P}_{S} । {P}_{S}\\right)\\ge t{P}_{O}\\ge t }{\\sum _{i=1}^{n} \\sum _{i=1}^{n}। \\left({P}_{S}। {P}_{S} \\ge t{P}_{O} \\ge t \\right)+ \\sum _{i=1}^{n}। ({P}_{O}। {P}_{S}\\ge t{P}_{O}\u0026lt;t) }\\left(18\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVMI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(VMI= \\frac{\\sum _{i=1}^{n}\\left({O}_{i}। \\right({S}_{i} \\le t\\\u0026amp;{O}_{i}\u0026gt;t\\left)\\right)}{\\sum _{i=1}^{n}\\left({S}_{i}। \\left({S}_{i}\u0026gt;t\\\u0026amp;{O}_{i}\u0026gt;t\\right)\\right)+\\sum _{i=1}^{n}\\left({O}_{i}। \\left({S}_{i}\\le t\\\u0026amp;{O}_{i}\u0026gt;t\\right)\\right)}\\dots \\dots ..\\left(19\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVCSI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(VCSI= \\frac{\\sum _{i=1}^{n}\\left({S}_{i}\\left|({S}_{i=1}\u0026gt;t \\\u0026amp; {O}_{i} \u0026gt;t)\\right.\\right)}{\\sum _{i=1}^{n}({S}_{i} \\left|({S}_{i}\\right.\u0026gt;t \\\u0026amp; {O}_{i} \u0026gt;t))+\\sum _{i=1}^{n}({S}_{i}\u0026lfloor;({S}_{i}\u0026gt;t \\\u0026amp; {O}_{i}\u0026gt;t))}\\dots \\dots \\dots \\dots ..\\left(20\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"3. Result And Discussions","content":"\u003cdiv class=\"Section2\" id=\"Sec8\"\u003e\n \u003ch2\u003e3.1 Seasonal rainfall climatology\u003c/h2\u003e\n \u003cp\u003eThe spatial climatology patterns of total rainfall of 23 CMIP6 historical simulations, ensemble mean (Ensmean) and observation (CHIRPS v2.0) of the 1981\u0026ndash;2014 average is presented in Figs. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e and \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e. Most models successfully reproduce the spatial patterns of total rainfall over highlands of western Ethiopia, western South Sudan, dry conditions over extreme northern parts of Sudan, arid and semi-arid climate over southern, northeastern Kenya, south-eastern Ethiopia, and most parts of Somalia during MAM (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e)and JJAS(Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). The majority of Models reproduced highest total rainfall amount (200\u0026ndash;600 mm) and 800-1200mm during MAM and JJAS respectively. Total rainfall increased from MAM to JJAS which is associated with the northward movement of ITCZ as main driver of simulated and projected rainfall over the region(Souverijns et al., 2016). Despite the important of MAM for parts of region close to Equator and JJAS seasonal rainfall over northern sector of IGAD region, models are able to capture the highest amount of rainfall over highlands of western Ethiopia, while northern parts of Sudan, Northeastern Kenya, southeastern Ethiopia, central and northern Somalia received lowest amount during MAM and JJAS. These results are in agreement with results from CORDEX Regional Climate Models carried out by Endris et al., (\u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003eb) over East African. Additionally, Majority of models under-estimated MAM and JJAS rainfall over highlands of western Kenya and well reproduced the patterns over costal. The results also reveal that CNRM-CM6-1-HR, GISS-E2-2-G, KIOST-ESM and CAMS-CSM1-0 tend to under-estimated rainfall, contrary to ACCESS-ESM1-5 over-estimated rainfall over South Sudan, central and highland of western Ethiopia and most parts of Uganda during JJAS. The Ensmean well represented MAM and JJAS rainfall compared to individual models, however, the arid and semi-arid regions in Kenya, Ethiopia and Somalia over-estimated the total rainfall over most parts of IGAD region.\u003c/p\u003e\n \u003cp\u003eThe total rainfall annual cycle patterns of 23 CMIP6 GCMs, ensemble mean and CHRIPS v2.0 presented in Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e. The datasets extracted over five potential food and cash crops cultivation areas of Al qadaref state in Sudan, Upper Nile state in South Sudan, Arsi Zone in Ethiopia, Aura district in Uganda and Trans Nzoia County in Kenya. The results show the models well reproduced the annual rainfall cycle, unimodal rainfall regimes over Al qadaref state in Sudan (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ea) and upper Nile state (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eb), peak in April and May, then July to October over Arsi zone in Ethiopia (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ec) and biomodal over Arua districts (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ed) and Trans Nzoia County (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ee). Annual cycle, explained the importance of MAM and OND seasons for equatorial eastern Africa countries, JJAS for the northern sector of GHA. It is clear that, most models under-estimate peak of rainfall in August over Al qadaref, June-August over Arua, February-May over Arsi, February-September over Trans Nzoia, while over-estimate April-September rainfall over Upper Nile, October-December over Arzi, Arua and Trans Nzoia. On a seasonal timescale, the MAM and JJAS over the Highlands of western Kenya, JJAS over Central Ethiopia and northeastern Sudan under-estimated rainfall, while northern Uganda during MAM and OND, central Ethiopia during OND, northeastern South Sudan during JJAS over-estimated rainfall. The MAM under-estimated cycle pattern in Kenya, JJAS in Sudan, over-estimate during OND season over Kenya and Uganda are consistent with CMIP5 results by Ongoma et al., (\u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e), also, in agreement with most CMIP3 simulations(Anyah \u0026amp; Qiu, \u003cspan class=\"CitationRef\"\u003e2012\u003c/span\u003e). The ensemble means of the models reproduce an annual cycle to a large extent compared to most individual models. ACCESS-ESM1-5 is only model over-estimates interannual variability from January to December over Al qadarif state, while CNRM-CM6-1-HR failed to simulate annual cycle over Al qadarif state, Uper Nile state, Arsi zone, and Arua district. The results also show the model\u0026apos;s patterns confirmed influence of north\u0026ndash;south passage of the ITCZ as reported in study by Clark et al., (\u003cspan class=\"CitationRef\"\u003e2003\u003c/span\u003e) and possible influence of ENSO phenomena among others climate drivers during OND rainfall(Ongoma et al., 2019). Seems, the effect of EA\u0026ndash;Indian Ocean, \u0026ndash;Asian monsoon, atmosphere\u0026ndash;ocean\u0026ndash;monsoon interactions are the main driver of bias in MAM and OND rains as reported by Yang et al., (\u003cspan class=\"CitationRef\"\u003e2015\u003c/span\u003e) in his study of Annual Cycle Bias over East Africa in CMIP5. Generally, the performance of the CMIP6 models is characterized by low skills in reproducing MAM rainfall patterns compared to peaks in OND season.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003e3.2 Spatial Patterns Of Statistical Deterministic Metrics\u003c/h3\u003e\n\u003cp\u003eThe spatial patterns of CC, PBIAS and RMSE statistical deterministic metrics during MAM and JJAS seasons over each local administrative unit in the IGAD region presented in Figs. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e\u0026ndash;7. The patterns of correlation between 23 CMIP6 models and EnsMean simulations relative to CHIRPS v2.0 reference datasets for the period from 1981 to 2014. The results revealed that all individual models during MAM (Figs. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003ea-w), and JJAS seasons (Figs. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003ea-w) and EnsMean (Figs. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003ex) observed the positive correlation over the entire IGAD region. Most parts of the region recorded weak CC values between 0-0.2, while the highest values not exceeding 0.8. Majority of zones in western Ethiopia recorded the highest CC, followed by South Sudan counties and southern parts of Sudan. The ACCESS-ESM1-5, CanESM5, CMCC-CM2-HR4, IPSL-CM6A-LR, MRI-ESM2-0 and NorESM2-MM are the most performed CMIP6 simulations over South Sudan with CC values exceeding 0.4. The majority of models recorded higher correlation over southern and central parts of Somalia compared to northern parts of the country. The majority of sub-counties and Parishes in northern Uganda recorded higher CC compared to central and southern parts. The GFDL-ESM4 performed better than other models over most parts of Uganda (Figs. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003ej). Majority of consistencies in Kenya, zones in central and northern Ethiopia recorded lowest correlation. All 23 models recorded less than 0.1 CC values over all districts in Djibouti, while the districts in northwestern Eritrea recorded better CC compared to other parts of the country. All models recorded lower CC over northern parts of Sudan. Compared to individual models, the EnsMean patterns showed the improved CC values over all parts of the region, with western Ethiopia, southern parts of Sudan and counties in South Sudan showing a remarkable correlation ranging between 0.5\u0026ndash;0.8(Figs. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003ex).\u003c/p\u003e\n\u003cp\u003eThe Pbias patterns during JJAS showed all 23 models over-estimated rainfall over all wards in southern and northeastern Kenya, southeastern Ethiopia, most districts in Somalia and Djibouti. Again, the ACCESS-ESM1-5 and MIROC6 the two models of over-estimated rainfall over the IGAD region. Also, with exception of ACCESS-ESM1-5 and MIROC6, all zones in highlands of western Ethiopia, all districts in Sudan under-estimated rainfall (Figs. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e). The ACCESS-ESM1-5 and MIROC6 are most model over-estimated (Figs. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003ea, r) and CNRM-CM6-1-HR is the most under-estimated rainfall over most parts of the region (Figs. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eg). The amount of rainfall over-estimated(under-estimated) for both MAM and JJAS exceeded 80mm per season over these areas. The patterns of under-estimation (over-estimation) are closely with seasons and dry or wet climate of the region. For instance, MAM and JJAS which are not main rainy seasons in Sudan and most parts of Kenya observed significant over-estimation of rainfall. The EnsMean simulated lowest Pbias compared to individual models over South Sudan compared to other countries. Generally, the Pbias has reduced over Sudan, Ethiopia and South Sudan, while northeastern Kenya, southeastern Ethiopia and most parts of Somalia over-estimation of rainfall amount which exceeded 70mm per/dekad (Figs. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003ex).\u003c/p\u003e\n\u003cdiv class=\"Section2\" id=\"Sec10\"\u003e\n \u003ch2\u003e3.3 Regional and sub-national statistical metrics\u003c/h2\u003e\n \u003cp\u003eTable \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e illustrates colored code portraits of 23 CMIP6 GCMs historical simulation with respect to CHIRPS-2.0 over Trans Nzioa county in Kenya during MAM and Arsi zone in Ethiopia during JJAS season. Assessment done using 20 different continuous, categorical and Volumatic indexes. The results show it is premature to conclude the best performed models based on limited indexes. In other words, it is not inclusive to selected the best performed models based on continuous indexes such as correlation, bias and RMSE, or categorical indexes (FAR, CSI, HSS) and Volumatic indexes such as VHI, VFAR, VMI, VCSI indexes. The individual models show variation in skills within 20 indexes considered in analysis. Some models performed very well based on continuous indexes and poorly under categorical and Volumatic indexes and vice versa. For example, MIROC6 performed very well using Volumatic indexes and poorly using continuous over Trans Nzoia. Similarly, the EC-Earth3 and CNRM-CM6-1-HR performed poorly under continuous indexes and better under both categorical and Volumatic indexes. The EnsMean scored best skill based on continuous indexes and poorly under Volumatic. The INM-CM5-0, HadGEM3-GC31-MM, CMCC-CM2-HR4, AWI-CM-1-1-MR, TaiESM1, NorESM2-MM, IITM-ESM, IPSL-CM6A-LR, GFDL-ESM4 and EC-Earth3 are the best 10 performed models over Trans Nzoia in Kenya during MAM season. Ranking individual models show INM-CM5-0 scored the best skills based on continuous indexes (Pbias, ME, MAE, RMSE, NSE and IOA), POD and CSI categorical indexes. Again, CMCC-CM2-HR4 scored better based on volumatic indexes (MQB, VHI, VFAR, VMI and VCSI). The colored code portraits of 23 CMIP6 over Arsi zone in Ethiopia during JJAS season shows failed assessment using categorical indexes because all values are the same for all models. Within one categorical or continuous index, there is variation in the way models perform. For instance, the EC-Earth3 performed the best using correlation and poorly under PBIAS, ME, MAE, RMSE. The ACCESS-ESM1-5, EC-Earth3, IPSL-CM6A-LR, MIROC6 and MRI-ESM2-0 performed well under categorical and Volumatic and poorly under continuous indexes. The BCC-CSM2-MR, INM-CM5-0, CAMS-CSM1-0, HadGEM3-GC31-MM, GFDL-ESM4, KACE-1-0-G, UKESM1-0-LL, NorESM2-MM, AWI-CM-1-1-MR and MPI-ESM1-2-HR are the best 10 performed models over Arsi in Ethiopia during. Again, the EnsMean scored best skill based on continuous indexes and poorly under Volumatic during JJAS over Arsi. It is worthy to mention that the validation using continuous indexes are much better and representative compared to categorical and Volumatic indexes.\u003c/p\u003e\n \n \u003cp\u003eThe scatter plots in Figs. \u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003e compare the 23 CMIP6 simulation, EnsMean and CHIRPS products at dekadal time-scales. The sample of results used are for Arua district in Uganda and El Qadaref in Sudan during MAM during JJAS respectively. There is wide scatter for all 23 CMIP6 products and EnsMean over Arua compared to El Qadaref. The values toward models\u0026rsquo; simulations (toward CHIRPS) explained the overestimated(underestimation) of the rainfall patterns. The majority of models overestimated rainfall, especially rainfall values over 50mm.The CNRM-CM6-1-HR and EC-Earth3 are the models that underestimated the rainfall over Arua and El Qadaref. The BCC-CSM2-MR, ACCESS-ESM1-5, CMCC-CM2-HR4, MIROC6, TaiESM1, MRI-ESM2-0 and GFDL-ESM4 are the most models with wider scattered rainfall values of overestimated rainfall over Arua during MAM season. The majority of models are much better performed over El Qadaref state compared to Arua district. The ACCESS-ESM1-5, AWI-CM-1-1-MR and MPI-ESM1-2-HR are the most common models with wide scatter over El Qadaref. The HadGEM3-GC31-MM shows the least scatter and best performed models (rainfall values are in agreement with model simulation). On the other hand, the EnsMean showed substantial difference scatter from individual models which exhibits systematic underestimation of rainfall amount. These due to high variability and number of Models overestimating/underestimating the rainfall amount, especially rainfall values exceeded 50mm. This scatter may be attributed to uncertainty in standard calendar, original resolution and variation associated with CHIRPS satellite estimates.\u003c/p\u003e\n \u003cp\u003eThe Cumulative Distribution Function comparing each 23 CMIP6 simulation against CHIRPS during MAM and JJAS seasons presented in Figs. \u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003e. The MAM season is represented by Trans Nzaia in Kenya and the JJAS season by Upper Nile in South Sudan. The results show that CDF the model\u0026rsquo;s performance varies from each model and rainfall amounts categories (0\u0026ndash;50 and 100-150mm). Majority of models over Trans Nzoia show a tight agreement with observation at lower rainfall compared to high amounts. The CMCC-CM2-HR4, BCC-CSM2-MR, IPSL-CM6A-LR, IITM-ESM, NorESM2-MM, MRI-ESM2-0 are the best performing models (shows a tight simulation). The AWI-CM-1-1-MR, CNRM-CM6-1-HR, EC-Earth3, MIROC6, MPI-ESM1-2-HR, UKESM1-0-LL and GISS-E2-2-G are most models that show a loose pattern. The AWI-CM-1-1-MR, CAMS-CSM1-0, CNRM-CM6-1-HR, EC-Earth3, MPI-ESM1-2-HR and UKESM1-0-LL underestimated the rainfall patterns. The performance of all models improved during JJAS in Upper Nile compared to Trans Nzoia, with ACCESS-ESM1-5 and BCC-CSM2-MR as the most model with loose simulation patterns for all rainfall amount density. Majority of models overestimated rainfall patterns over Upper Nile. However, the 23 models have better skills over Upper Nile state compared to Trans Nzoia. The CAMS-CSM1-0, KACE-1-0-G, NorESM2-MM, HadGEM3-GC31-MM, CMCC-CM2-HR4 are most performed models over Upper Nile in South Sudan.\u003c/p\u003e\n \u003cp\u003eDue to variation in models\u0026rsquo; performance using 20 continuous, categorical and volumetric indices indexes, scatter plots and CDF. The all 23 models ranked based on perfect values of CC, BR2, Pbias, ME, NSE, MAE and RMSE were ranked from 1 to 10 regardless of the position of EnsMean in ranking over entire pixels in the IGAD region. Out of 23 models ranked, the best 10 performance models during MAM and JJAS season were selected from 13 models-maintained appearances in ranking. From this, the best 10 ranked models based on perfects scored values of of CC, BR2, Pbias, ME, NSE, MAE and RMSE are INM-CM5-0, CMCC-CM2-HR4, HadGEM3-GC31-MM, IPSL-CM6A-LR, TaiESM1, IITM-ESM, KIOST-ESM, GFDL-ESM4, CanESM5 and BCC-CSM2-MR during MAM (Figs. 10a). The best 10 perfect scored models during JJAS are EC-Earth3, INM-CM5-0, HadGEM3-GC31-MM, CMCC-CM2-HR4, KACE-1-0-G, IPSL-CM6A-LR, NorESM2-MM, GFDL-ESM4, MRI-ESM2-0 and BCC-CSM2-MR. Therefore, the INM-CM5-0, HadGEM3-GC31-MM, CMCC-CM2-HR4, IPSL-CM6A-LR, KACE-1-0-G, EC-Earth3, NorESM2-MM and GFDL-ESM4, TaiESM1 and KIOST-ESM are most ranked 10 models during MAM and JJAS season (Figs. 10b). Therefore, they were selected to compute EnsMean which will be used in computing future patterns of changes and variability in extreme events linked to food security over the IGAD region of Eastern Africa.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"4. Summary And Conclusions","content":"\u003cp\u003eThis paper documents the performances of the 23 CMIP6 GCMs historical rainfall simulations over the IGAD region of Eastern Africa. The spatial patterns of rainfall total validated using 20 continuous, categorical and Volumatic statistical metrics, scatter and CDF plots with respect to CHIRPS-2.0 as reference datasets. The validation was carried out at regional, nation, sub-nation and local administrative levels during 1981\u0026ndash;2014. The results showed the majority of CMIP6 GCMs individual models, EnsMean successfully reproduced the spatial patterns of MAM and JJAS total rainfall. The individual models and EnsMean captured the bi-model rainfall regime or peak over highlands of western Kenya and northern Uganda. Furthermore, the spatial patterns of continuous statistical metrics such CC, Pbias and RMSE showed most models scored highest skills over zones in highlands of western Ethiopia, most countries in South Sudan, and southern parts of Sudan. The lowest model\u0026rsquo;s skill is observed over northeastern and zones in Ethiopia, ASALs counties in Kenya and districts in Somalia. The CNRM-CM6-1HR considered the most model under-estimated total rainfall, failed to simulate annual cycle patterns, highest negative bias and RMSE values, and simulated lowest number of wet days over most parts of study domain. The colored code portrait, scatter and CDF plots of 23 CMIP6 GCMs historical simulations over five potential agricultural areas perceived as food basket of IGAD region suggested it is not sufficient to conclude the best performed models based on limited indexes or individual indexes. The 20 continuous, categorical and Volumatic shows variation within one individual category of index. Some models performed well under continuous and poorly under categorical and Volumatic and vice versa. The scatter plots and CDF explained the magnitude of discrepancies between individual models\u0026rsquo; simulations and reference data. The majority of models considered in this study overestimated rainfall amounts over most parts of the IGAD region. Furthermore, except IPSL-CM6A-LR, INM-CM5-0 and MIROC6 during MAM and Ec-earth3, ACCESS-ESM1-5 during, other 20 CMIP6 examined over-estimated total rainfall over most parts of the region. The ranking skills of the 23 CMIP6 GCMs over El qadarif state in Sudan, Arsi district in Ethiopia, Upper Nile state in South Sudan, Trans Nzoia County in Kenya and Arua district in Uganda and entire pixels over IGAD region shows fluctuation in ranking the best 10 performance models within countries and sub-locations. Some models consistently appeared in top 10 best performed models, while others kept in and out of the list. For example, ranked best 10 models over entire spatial average of the IGAD region showed the INM-CM5-0, CMCC-CM2-HR4, HadGEM3-GC31-MM, IPSL-CM6A-LR, TaiESM1, IITM-ESM, KIOST-ESM, GFDL-ESM4, CanESM5 and BCC-CSM2-MR during MAM. Similarly, the EC-Earth3, INM-CM5-0, HadGEM3-GC31-MM, CMCC-CM2-HR4, KACE-1-0-G, IPSL-CM6A-LR, NorESM2-MM, GFDL-ESM4, MRI-ESM2-0 and BCC-CSM2-MR are most ranked 10 models during JJAS season. Therefore, INM-CM5-0, HadGEM3-GC31-MM, CMCC-CM2-HR4, IPSL-CM6A-LR, KACE-1-0-G, EC-Earth3, NorESM2-MM and GFDL-ESM4, TaiESM1 and KIOST-ESM are best 10 models consistently appeared in the top 10. These are the individual models used in computing the Multi-Models Ensemble (MME). The MME will be used in projecting the extremes rainfall events linked to food security under four Shared Socioeconomic Pathway (SSP) scenarios (SSP1-2.6, SSP2‐4.5, SSP3-7.0, and SSP5‐8.5) in ESGF database.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u003c/strong\u003e This work is part of a PhD work at the Uni\u0026shy;versity of Nairobi, Kenya, Faculty of Science \u0026amp; Technology, Department of Earth \u0026amp; Climate Sciences. Also, author wish to acknowledge World Climate Research Programme (WCRP) Coupled Model Intercomparison Project (Phase 6) and USA, PCMDI/LLNL (California), France, IPSL, Germany, DKRZ and UK, CEDA for archiving and open access to the CMIP6 data \u0026nbsp;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e Not applicable, however the authors wish to acknowledge the support by Intra-ACP Climate Services and Related Applications (ClimSA) to IGAD climate Prediction and Application Center (ICPAC) for the funds promised for the open access publication of this research.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclarations:\u003c/strong\u003e I declare that this paper is my original work and has not been submitted elsewhere for publication. Where other people\u0026rsquo;s work, or my own work has been used, this has properly been acknowledged and referenced in accordance with the University of Nairobi\u0026rsquo;s requirements.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest:\u003c/strong\u003e All authors declare no competing interests\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e All authors consent to participate\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e All authors consent to publish this work\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability:\u003c/strong\u003e The secondary datasets generated during analysis are available via request. The primary data are open access from cmip6 - Home | ESGF-CoG (llnl.gov)\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode availability:\u003c/strong\u003e We used Climate Data Operators (CDO), R-Packages and Climate Data Tool (CDT) codes \u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contribution:\u0026nbsp;\u003c/strong\u003eAll authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by\u003cem\u003e\u0026nbsp;Paulino Omoj. Omay.\u0026nbsp;\u003c/em\u003eThe first draft of the manuscript was written by\u003cem\u003e\u0026nbsp;Paulino Omoj. Omay\u0026nbsp;\u003c/em\u003esupervised by\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003cem\u003eNzioka J. Muthama\u003csup\u003e\u0026nbsp;\u003c/sup\u003e, \u0026nbsp;Christopher Oludhe,\u003csup\u003e\u0026nbsp;\u003c/sup\u003eJosiah M. Kinama .\u0026nbsp;\u003c/em\u003eFinal\u003cem\u003e\u0026nbsp;\u003c/em\u003emanuscript\u0026nbsp;version reviewed by\u003cem\u003e\u0026nbsp;Guleid Artan \u0026nbsp;and Zachary Atheru .\u0026nbsp;\u003c/em\u003eAll authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAlmazroui, M., Islam, M. N., Saeed, F., Saeed, S., Ismail, M., Ehsan, M. A., \u0026hellip; Barlow, M. (2021). Projected Changes in Temperature and Precipitation Over the United States, Central America, and the Caribbean in CMIP6 GCMs. \u003cem\u003eEarth Systems and Environment\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e(1), 1\u0026ndash;24. https://doi.org/10.1007/s41748-021-00199-5\u003c/li\u003e\n \u003cli\u003eAlmazroui, M., Saeed, F., Saeed, S., Nazrul Islam, M., Ismail, M., Klutse, N. A. B., \u0026amp; Siddiqui, M. H. (2020). Projected Change in Temperature and Precipitation Over Africa from CMIP6. \u003cem\u003eEarth Systems and Environment\u003c/em\u003e, \u003cem\u003e4\u003c/em\u003e(3), 455\u0026ndash;475. https://doi.org/10.1007/s41748-020-00161-x\u003c/li\u003e\n \u003cli\u003eAnyah, R. O., \u0026amp; Qiu, W. (2012). Characteristic 20th and 21st century precipitation and temperature patterns and changes over the Greater Horn of Africa. \u003cem\u003eInternational Journal of Climatology\u003c/em\u003e. https://doi.org/10.1002/joc.2270\u003c/li\u003e\n \u003cli\u003eAyugi, B., Zhihong, J., Zhu, H., Ngoma, H., Babaousmail, H., Rizwan, K., \u0026amp; Dike, V. (2021). Comparison of CMIP6 and CMIP5 models in simulating mean and extreme precipitation over East Africa. \u003cem\u003eInternational Journal of Climatology\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(15), 6474\u0026ndash;6496. https://doi.org/10.1002/joc.7207\u003c/li\u003e\n \u003cli\u003eBabaousmail, H., Hou, R., Ayugi, B., Ojara, M., Ngoma, H., Karim, R., \u0026hellip; Ongoma, V. (2021). Evaluation of the performance of cmip6 models in reproducing rainfall patterns over north africa. \u003cem\u003eAtmosphere\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(4), 1\u0026ndash;25. https://doi.org/10.3390/atmos12040475\u003c/li\u003e\n \u003cli\u003eCamberlin, P., \u0026amp; Okoola, R. E. (2003). The onset and cessation of the \u0026ldquo;long rains\u0026rdquo; in eastern Africa and their interannual variability. \u003cem\u003eTheoretical and Applied Climatology\u003c/em\u003e. https://doi.org/10.1007/s00704-002-0721-5\u003c/li\u003e\n \u003cli\u003eCattani, E., Merino, A., Guijarro, J. A., \u0026amp; Levizzani, V. (2018). East Africa Rainfall trends and variability 1983-2015 using three long-term satellite products. \u003cem\u003eRemote Sensing\u003c/em\u003e. https://doi.org/10.3390/rs10060931\u003c/li\u003e\n \u003cli\u003eClark, C. O., Webster, P. J., \u0026amp; Cole, J. E. (2003). Interdecadal variability of the relationship between the Indian Ocean zonal mode and East African coastal rainfall anomalies. \u003cem\u003eJournal of Climate\u003c/em\u003e, \u003cem\u003e16\u003c/em\u003e(3), 548\u0026ndash;554. https://doi.org/10.1175/1520-0442(2003)016\u0026lt;0548:IVOTRB\u0026gt;2.0.CO;2\u003c/li\u003e\n \u003cli\u003eDinku, T., Funk, C., Peterson, P., Maidment, R., Tadesse, T., Gadain, H., \u0026amp; Ceccato, P. (2018). Validation of the CHIRPS satellite rainfall estimates over eastern Africa. \u003cem\u003eQuarterly Journal of the Royal Meteorological Society\u003c/em\u003e. https://doi.org/10.1002/qj.3244\u003c/li\u003e\n \u003cli\u003eDong, G., Zhang, H., Moise, A., Hanson, L., Liang, P., \u0026amp; Ye, H. (2016). CMIP5 model-simulated onset, duration and intensity of the Asian summer monsoon in current and future climate. \u003cem\u003eClimate Dynamics\u003c/em\u003e, \u003cem\u003e46\u003c/em\u003e(1\u0026ndash;2), 355\u0026ndash;382. https://doi.org/10.1007/s00382-015-2588-z\u003c/li\u003e\n \u003cli\u003eDosio, A., Jury, M. W., Almazroui, M., Ashfaq, M., Diallo, I., Engelbrecht, F. A., \u0026hellip; Tamoffo, A. T. (2021). Projected future daily characteristics of African precipitation based on global (CMIP5, CMIP6) and regional (CORDEX, CORDEX-CORE) climate models. \u003cem\u003eClimate Dynamics\u003c/em\u003e, Vol. 57, pp. 3135\u0026ndash;3158. https://doi.org/10.1007/s00382-021-05859-w\u003c/li\u003e\n \u003cli\u003eEndris, H. S., Omondi, P., Jain, S., Lennard, C., Hewitson, B., Chang\u0026rsquo;a, L., \u0026hellip; Tazalika, L. (2013). Assessment of the performance of CORDEX regional climate models in simulating East African rainfall. \u003cem\u003eJournal of Climate\u003c/em\u003e. https://doi.org/10.1175/JCLI-D-12-00708.1\u003c/li\u003e\n \u003cli\u003eEyring, V., Bony, S., Meehl, G. A., Senior, C. A., Stevens, B., Stouffer, R. J., \u0026amp; Taylor, K. E. (2016). Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. \u003cem\u003eGeoscientific Model Development\u003c/em\u003e. https://doi.org/10.5194/gmd-9-1937-2016\u003c/li\u003e\n \u003cli\u003eFunk, C., Peterson, P., Landsfeld, M., Pedreros, D., Verdin, J., Shukla, S., \u0026hellip; Michaelsen, J. (2015). The climate hazards infrared precipitation with stations - A new environmental record for monitoring extremes. \u003cem\u003eScientific Data\u003c/em\u003e. https://doi.org/10.1038/sdata.2015.66\u003c/li\u003e\n \u003cli\u003eGebrechorkos, Solomon H., H\u0026uuml;lsmann, S., \u0026amp; Bernhofer, C. (2019). Regional climate projections for impact assessment studies in East Africa. \u003cem\u003eEnvironmental Research Letters\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(4). https://doi.org/10.1088/1748-9326/ab055a\u003c/li\u003e\n \u003cli\u003eGebrechorkos, Solomon Hailu, H\u0026uuml;lsmann, S., \u0026amp; Bernhofer, C. (2018). Evaluation of multiple climate data sources for managing environmental resources in East Africa. \u003cem\u003eHydrology and Earth System Sciences\u003c/em\u003e. https://doi.org/10.5194/hess-22-4547-2018\u003c/li\u003e\n \u003cli\u003eGebresellase, S. H., Wu, Z., Xu, H., \u0026amp; Muhammad, W. I. (2022). Evaluation and selection of CMIP6 climate models in Upper Awash Basin (UBA), Ethiopia: Evaluation and selection of CMIP6 climate models in Upper Awash Basin (UBA), Ethiopia. \u003cem\u003eTheoretical and Applied Climatology\u003c/em\u003e, \u003cem\u003e149\u003c/em\u003e(3\u0026ndash;4), 1521\u0026ndash;1547. https://doi.org/10.1007/s00704-022-04056-x\u003c/li\u003e\n \u003cli\u003eIpcc. (2013). Working Group I Contribution to the IPCC Fifth Assessment Report, Climate Change 2013: The Physical Science Basis. \u003cem\u003eIpcc\u003c/em\u003e.\u003c/li\u003e\n \u003cli\u003eKimani, M. W., Hoedjes, J. C. B., \u0026amp; Su, Z. (2017). An assessment of satellite-derived rainfall products relative to ground observations over East Africa. \u003cem\u003eRemote Sensing\u003c/em\u003e. https://doi.org/10.3390/rs9050430\u003c/li\u003e\n \u003cli\u003eLi, Y., Wang, C., \u0026amp; Su, F. (2021). Evaluation of CMIP6 Models over Two Third Pole Subregions with Contrasting Circulation Systems. \u003cem\u003eJournal of Climate\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(22), 9133\u0026ndash;9152. https://doi.org/10.1175/JCLI-D-21-0214.1\u003c/li\u003e\n \u003cli\u003eOngoma, V., Chena, H., \u0026amp; Gaoa, C. (2018). Projected changes in mean rainfall and temperature over east Africa based on CMIP5 models. \u003cem\u003eInternational Journal of Climatology\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e(3), 1375\u0026ndash;1392. https://doi.org/10.1002/joc.5252\u003c/li\u003e\n \u003cli\u003eOtieno, V. O., \u0026amp; Anyah, R. O. (2013). CMIP5 simulated climate conditions of the Greater Horn of Africa (GHA). Part II: Projected climate. \u003cem\u003eClimate Dynamics\u003c/em\u003e. https://doi.org/10.1007/s00382-013-1694-z\u003c/li\u003e\n \u003cli\u003ePeel, M. C., Finlayson, B. L., \u0026amp; McMahon, T. A. (2007). Updated world map of the K\u0026ouml;ppen-Geiger climate classification. \u003cem\u003eHydrology and Earth System Sciences\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(5), 1633\u0026ndash;1644. https://doi.org/10.5194/hess-11-1633-2007\u003c/li\u003e\n \u003cli\u003eScoccimarro, E., Gualdi, S., Bellucci, A., Zampieri, M., \u0026amp; Navarra, A. (2013). Heavy precipitation events in a warmer climate: Results from CMIP5 models. \u003cem\u003eJournal of Climate\u003c/em\u003e. https://doi.org/10.1175/JCLI-D-12-00850.1\u003c/li\u003e\n \u003cli\u003eSeneviratne, S. I., Zhang, X., Adnan, M., Badi, W., Dereczynski, C., Di Luca, A., \u0026hellip; Zhou, B. (2021). Weather and Climate Extreme Events in a Changing Climate. In \u003cem\u003eClimate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change\u003c/em\u003e. https://doi.org/10.1017/9781009157896.013\u003c/li\u003e\n \u003cli\u003eSong, S., \u0026amp; Yan, X. (2022). Evaluation of events of extreme temperature change between neighboring days in CMIP6 models over China. \u003cem\u003eTheoretical and Applied Climatology\u003c/em\u003e, \u003cem\u003e150\u003c/em\u003e(1\u0026ndash;2), 53\u0026ndash;72. https://doi.org/10.1007/s00704-022-04142-0\u003c/li\u003e\n \u003cli\u003eSow, M., Diakhat\u0026eacute;, M., Dixon, R. D., Guichard, F., Dieng, D., \u0026amp; Gaye, A. T. (2020). Uncertainties in the annual cycle of rainfall characteristics over West Africa in CMIP5 models. \u003cem\u003eAtmosphere\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(2). https://doi.org/10.3390/atmos11020216\u003c/li\u003e\n \u003cli\u003eTaylor, P. C., Fahey, D., \u0026amp; Doherty, S. (2017). \u003cem\u003eDigitalCommons @ University of Nebraska - Lincoln Physical drivers of climate change\u003c/em\u003e. \u003cem\u003eI\u003c/em\u003e(January 2017), 73\u0026ndash;113.\u003c/li\u003e\n \u003cli\u003eVrac, M., Thao, S., \u0026amp; Yiou, P. (2022). Changes in temperature\u0026ndash;precipitation correlations over Europe: are climate models reliable? \u003cem\u003eClimate Dynamics\u003c/em\u003e, (2020). https://doi.org/10.1007/s00382-022-06436-5\u003c/li\u003e\n \u003cli\u003eWang, C., Wilson, D., Haack, T., Clark, P., Lean, H., \u0026amp; Marshall, R. (2012). Effects of initial and boundary conditions of mesoscale models on simulated atmospheric refractivity. \u003cem\u003eJournal of Applied Meteorology and Climatology\u003c/em\u003e, \u003cem\u003e51\u003c/em\u003e(1), 115\u0026ndash;132. https://doi.org/10.1175/JAMC-D-11-012.1\u003c/li\u003e\n \u003cli\u003eWilliams, A. P., \u0026amp; Funk, C. (2011). A westward extension of the warm pool leads to a westward extension of the Walker circulation, drying eastern Africa. \u003cem\u003eClimate Dynamics\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(11\u0026ndash;12), 2417\u0026ndash;2435. https://doi.org/10.1007/s00382-010-0984-y\u003c/li\u003e\n \u003cli\u003eYang, W., Seager, R., Cane, M. A., \u0026amp; Lyon, B. (2015). The rainfall annual cycle bias over East Africa in CMIP5 coupled climate models. \u003cem\u003eJournal of Climate\u003c/em\u003e. https://doi.org/10.1175/JCLI-D-15-0323.1\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table","content":"\u003cp\u003eTable 3 is available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-2747422/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-2747422/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAccuracy of model\u0026rsquo;s simulations are critical for climate change and its socio-economic impact. In this study, we evaluated 23 Global climate models participating in the Coupled Model Intercomparison Project phase 6 (CMIP6). The main objective was to identify top 10 best performance models in capturing patterns of rainfall for the 1981\u0026ndash;2014 period over the Intergovernmental Authority on Development (IGAD) region of Eastern Africa. The total rainfall, annual cycle, continuous, categorical and Volumatic statistical metrics, scatter plots, Cumulative Distribution Function (CDF) and colored code portrait were used to assess the patterns of total rainfall. Results indicate that most CMIP6 models generally capture the characteristics of the observed climatology pattern of total rainfall, bimodal and unimodal rainfall regimes. The majority of models over Arid and Semi-Arid Lands (ASALs) in Kenya, Somalia, Ethiopia and Sudan scored lowest skills, highest bias and over-estimated rainfall. In addition, 21 out of 23 CMIP6 over-estimated rainfall over most parts of the region. The ACCESS-ESM1-5 and MIROC6 are the most over-estimated models opposed to CNRM-CM6-1HR as the most model under-estimated rainfall, highest bias and RMSE values. The regional and sub-national analysis showed, it is inconclusive to select best performed models based on individual metric. Out of 23 models, the INM-CM5-0, HadGEM3-GC31-MM, CMCC-CM2-HR4, IPSL-CM6A-LR, KACE-1-0-G, EC-Earth3, NorESM2-MM, GFDL-ESM4, TaiESM1 and KIOST-ESM are the best 10 performance models over IGAD region. These findings highlight the importance of selecting best performance models for mapping present and future hotspots and extreme rainfall events over the IGAD region of Eastern Africa.\u003c/p\u003e","manuscriptTitle":"Evaluation of CMIP6 Historical Simulations over IGAD region of Eastern Africa","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-03-30 15:13:32","doi":"10.21203/rs.3.rs-2747422/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ad739667-68b9-449a-8386-fac6ab7e6221","owner":[],"postedDate":"March 30th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2023-04-04T07:44:23+00:00","versionOfRecord":[],"versionCreatedAt":"2023-03-30 15:13:32","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-2747422","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-2747422","identity":"rs-2747422","version":["v1"]},"buildId":"cBFmMYwuxLRRLfASyISRj","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}