Evaluation of extreme temperature events as simulated by CMIP6 models over Central Africa

Evaluation of extreme temperature events as simulated by CMIP6 models over Central 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 extreme temperature events as simulated by CMIP6 models over Central Africa Gabriel Fotso-Kamga, Thierry C. Fotso-Nguemo, Zéphirin D. Yepdo, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7155223/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 05 Mar, 2026 Read the published version in Theoretical and Applied Climatology → Version 1 posted 9 You are reading this latest preprint version Abstract Extreme temperature events pose significant risks to both natural environments and communities across Central Africa (CA). Gaining deeper insight into how these events vary is crucial to inform effective climate change mitigation and adaptation plans. The present study evaluates the suitability of global climate models from the Coupled Model Intercomparison Project Phase 6 (CMIP6), along with their multi-model ensemble mean (MME), in simulating the recent past spatial variations of extreme temperature events in CA. For this purpose, under the period 1985–2014, we assessed seven relevant indicators based on daily minimum and maximum temperatures, recommended by the Expert Team on Climate Change Detection and Indices (ETCCDI). We examined the spatial patterns of these extreme temperature events as simulated by sixteen CMIP6 models and their MME against CHIRTS and ERA5 observational and reanalysis datasets, focusing on percentile, absolute, and duration-based indices. The results showed that both individual models and the MME demonstrated reasonable skill in reproducing the spatial patterns of most extreme temperature indices, with the MME often showing better agreement with observations. However, the models faced challenges in accurately simulating the frequency of the percentile-based indices, notably TX90p and TN10p, across all sub-regions. Furthermore, the model performance varies depending on the specific index and the climatic sub-region. This study thereby elucidates the capabilities and shortcomings of CMIP6 models in representing extreme temperatures across Central Africa, providing valuable information for developing more reliable regional climate projections and assessing the future impacts of climate change on temperature extremes. Extreme temperature events Central Africa CMIP6 Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 1 Introduction Temperature extremes, encompassing both heat and cold, pose significant threats, including increased human mortality, stress on infrastructure, intensified wildfires, and disruptions to agriculture and transportation, all of which severely impact both natural and built environments ( Ryti et al. 2016 ; Ebi et al. 2021 ; Fotso-Nguemo et al. 2022 ). Variations in the frequency and magnitude of climatic parameters, or their corresponding indices, are generally employed to represent indicators of climate change. Prior investigations over the African domain have projected a heightened probability of extreme temperature and precipitation events occurring in conjunction with escalating anthropogenic greenhouse gas emissions ( Dosio et al. 2021 ; Fotso-Nguemo et al. 2023 ; Ngavom et al. 2024 ; Tamoffo et al. 2023 ). Despite extensive research on global extreme temperature trends and climate change ( Abatan et al. 2016 ; Aguilar et al. 2009 ; Donat et al. 2013 ; Dunn et al. 2020 ; Fotso-Kamga et al. 2023 ; Lewis and King 2015 ; Moron et al. 2016 ; Ozturk et al. 2021 ; van Der Walt et al. 2021), which has shown a worldwide increase in extreme heat and a decrease in extreme cold, there remains a significant gap in studies evaluating the reliability of data sources for accurately representating of mean climatological extremes, particularly in regions with sparse observational data, such as Central Africa (CA). Global climate models (GCMs) are valuable tools for simulating contemporary and future climatic conditions and generating the requisite long-term data for analyzing potential alterations in extreme events. However, comparative analyses with observational datasets frequently reveal systematic biases within these models, thereby indicating limitations in their capacity to reproduce observed/real conditions accurately and contributing to inherent uncertainties in their projections. These uncertainties inherent in climate model projections constitute a significant concern for decision- and policy-makers ( Zwiers et al. 2013 ). CMIP6 are the latest generation of climate models within the Coupled Model Intercomparison Project, featuring enhanced complexity and resolution compared to earlier phases like CMIP5 ( Eyring et al. 2016 ). These advancements aim to provide more accurate simulations of historical climate and more robust future projections. The review by Zwiers et al. ( 2013 ) has emphasized the importance of understanding observed historical extremes, as this is crucial for future progress in comprehending the implications of ongoing and future extreme climate events. The importance of using the most appropriate indices when assessing the impact of sustained extreme temperature events has been presented by many authors ( Perkins et al. 2012 ; Seneviratne et al. 2012 ; Zwiers et al. 2013 ). Assessing daily extreme temperature events in CA, using either observational or model simulations, remains significantly underdeveloped. We aim to fill this gap in the present study, by evaluating the capability of CMIP6 models to simulate a range of extreme temperature indices over CA from 1985 to 2014. By doing so, we thereby highlight both the models' strengths and limitations in capturing their spatial patterns. To some extent, this work complements the former study by Komelo et al. ( 2024 ), which evaluated CMIP6 model-simulated extreme precipitation and its spatial variability in CA, fostering climate model improvement. The remainder of the paper is structured as follows: the subsequent section ( Section 2 ) describes the study area, the datasets employed and the methodological approach. Section 3 presents and discusses the findings, and Section 4 wraps up and examines the findings. 2 Study area, data and methodology 2.1 Study area This study focuses on the CA domain, geographically defined between 15 °S and 15 °N latitude, and 5°–35° E longitude. This region encompasses most countries in Central Africa, including Cameroon, Chad, the Central African Republic, Equatorial Guinea, Gabon, Congo, and the Democratic Republic of Congo ( Komelo et al. 2024 ; Mbienda et al. 2022 ; Tamoffo et al. 2019 ). For quantitative analyses, we further divided the CA domain into five distinct climatic sub-regions, defined based on the Köppen-Geiger climate classification ( Peel et al. 2007 ). These sub-regions are : The Sudano-Sahelian (SS) sub-region (9.5°–14.5 °N, 8.0°–23.5 °E), which covers southern Chad and northern Cameroon, experiencing the semi-arid Sahelian climate, characterized by high temperatures, strong solar radiation, and a unimodal rainfall-like regime. The Northern Equatorial (NE) sub-region (2.0°–9.5°N, 8.0°–18.5 °E and 5.0°–9.5°N, 18.5°–32.0°E) covering southern Cameroon and CAR. This sub-region is characterized by a predominantly tropical wet and dry climate, featuring a long rainy season lasting approximately nine months, warm temperatures througtout the year with a notable annual range, and maximum temperatures that often precede the onset of the rainy season. The Equatorial East (EQE) sub-region (6.0°S–5.0°N, 18.5°–32.0°E) covers central parts of the Democratic Republic of Congo (DRC). This sub-region is characterized by a tropical rainforest-like climate, with consistently high temperatures and humidity, relatively small seasonal temperature variations, and a bimodal rainfall-like regime. The Equatorial West (EQW) sub-region (9.0°S–2.0°N, 8.0°–18.5°E and 9.0°–6.0°S, 18.5°–21.5°E) covers Equatorial Guinea, Gabon, Congo, southwestern DRC, and northern Angola. It features a tropical rainforest-type climate, similar to the EQE region, with consistently high temperatures and humidity and minimal seasonal variation. Finaly, Southern Equatorial (SE) sub-region (14.0°–6.0°S, 21.5°–32.0°E) covers southeastern DRC and northern Zambia. This subregion is characterized by subtropical climate, with warm temperatures year-round and a more distinct seasonal temperature cycle compared to the equatorial sub-regions. 2.2 Datasets Daily near-surface (2 m) temperature data from sixteen individual CMIP6 GCMs initiative and their multi-model ensemble (MME) ( Eyring et al. 2016 ) are utilized in this study. These globally available datasets span the period 1850–2014. Details of the individual models are provided in Table 1 . Table 1 Overview of the sixteen CMIP6 climate models used in this study. N° Model name Institute ID Resolution (Lon × Lat) References 1 BCC-ESM1 Beijing Climate Center (BCC) and China Meteorological Administration (CMA) 2.81° × 2.81° Wu et al. ( 2020 ) 2 CanESM5 Canadian Centre for Climate Modelling and Analysis, Victoria, Canada 2.81° × 2.81° Swart et al. ( 2019 ) 3 CNRM-CM6-1 Centre National de Recherches Météorologiques (CNRM); Centre Européen de Recherches et de Formation Avancée en Calcul Scientifique 1.41° × 1.41° Voldoire et al. ( 2019 ) 4 E3SM-2–0 Department of Energy's Energy Exascale Earth System Model, USA 1.00° × 1.00° Golaz et al. ( 2022 ) 5 EC-EARTH3 EC-EARTH consortium 0.70° × 0.70° Döscher et al. ( 2021 ) 6 GFDL-ESM4 Geophysical Fluid Dynamics Laboratory, Princeton, USA 1.25° × 1.00° Horowitz et al. ( 2018 ) 7 HadGEM3-GC31-LL Met Office Hadley Centre, Exeter, UK 1.88° × 1.25° Williams et al. ( 2021 ) 8 INM-CM5-0 Institute for Numerical Mathematics, Moscow, Russia 2.00° × 1.50° Volodin et al. ( 2019 ) 9 IPSL-CM6A-LR Institut Pierre Simon Laplace, Paris, France 2.50° × 1.27° Boucher et al. ( 2020 ) 10 MIROC6 The University of Tokyo, Chiba, Japan 1.41° × 1.41° Tatebe et al. ( 2019 ) 11 MPI-ESM1-2-HR Max Planck Institute for Meteorology, Hamburg, Germany 0.94° ×0.94 ° Müller et al. ( 2018 ) 12 MPI-ESM1-2-LR Max Planck Institute for Meteorology, Hamburg, Germany 1.90° × 1.90° Mauritsen et al. ( 2019 ) 13 NorESM2-LM Norwegian Meteorological Institute, Oslo, Norway 2.50° × 1.89° Seland et al. ( 2020 ) 14 NorESM2-MM Norwegian Meteorological Institute, Oslo, Norway 1.25° × 0.94° Seland et al. ( 2020 ) 15 SAM0-UNICON Seoul National University Atmosphere Model Version 0 with a Unified Convection Scheme 1.25° × 0.94° Park and Shin (2019) 16 UKESM1-0-LL Met Office Hadley Centre, Exeter, UK 1.88° × 1.25° Tang et al. ( 2019 ) To evaluate the CMIP6 model simulations, the study compares their output against gridded observational temperature products, which include : The Climate Hazards Center Infrared Temperature with Stations (CHIRTS) Version 1.0 dataset ( Funk et al. 2019 ), which serves as the primary reference for near-surface temperature, as it demonstrates good performance across Africa ( Parsons et al. 2022 ). CHIRTS provides daily minimum and maximum air temperature estimates on a 0.05° horizontal resolution, and over a quasi-global scale from 1983 to 2016. It is generated by blending satellite infrared temperatures and station data, enhanced with ERA5 for daily disaggregation. The fifth generation of the European Centre for Medium-Range Weather Forecasts Re-Analysis (ERA5) ( Hersbach et al. 2020 ) is also used to account for potential uncertainties between observational and reanalysis datasets. ERA5 combines global climate models with ground, ocean, and satellite observations to produce hourly estimates of various variables globally. We use the ERA5 2-m temperature records, available as hourly time scale from 1979 to the present, at a spatial resolution of 0.25° x 0.25°. 2.3 Methods We examined seven temperature indices defined by the ETCCDI (as detailed in Table 2 ), which were calculated from daily temperature data using the Climate Data Operators (CDO; https://code.mpimet.mpg.de/projects/cdo ). These particular temperature indices were chosen based on their relevance to socio-economic sectors such as health and agriculture, especially in Africa, as mentioned in previous research ( Aguilar et al. 2009 ; Fotso-Kamga et al. 2023 ; Mengistu et al. 2024 ; Ntoumos et al. 2020 ; Ozturk et al. 2021 ). Table 2 Selected temperature indices for this study. Label Index Name Index Category Description Unit T90 90th percentile of maximum temperature Percentile Temperature value below which 90% of the observed daily maximum temperatures (TX) are lower for a time period °C T10 10th percentile of minimum temperature Percentile Temperature value below which 10% of the observed daily minimum temperatures (TN) are lower for a time period °C TX90p Warm days Percentile Percentage of days for which TX > 90th percentile over a given period % TN10p Cool nights Percentile Percentage of days for which TN < 10th percentile over a given period % DTR Diurnal temperature range Absolute Mean difference between daily maximum temperature and daily minimum temperature. That is TX-TN °C WSDI Warm spell days index Duration Number of days per period when, in intervals of at least 6 consecutive days, the daily mean temperature is above 90th percentile for the period days CSDI Cold spell days index Duration Number of days per period when, in intervals of at least 6 consecutive days, the mean daily temperature is below the 10th percentile for the period. days To assess the capacity of CMIP6 simulations to reproduce daily temperature indices across Central Africa, a common 30-year period (1985–2014) was considered for both models and observations. To address the disparity in their native spatial resolutions, all datasets were interpolated to a uniform 1°×1° grid using conservative interpolation methods prior to analyses. We further quantified the agreement between the simulated and observed daily temperature indices using the following statistical metrics : The Taylor diagram ( Taylor 2001 ), which summarizes models' skill in capturing observed patterns and magnitude by providing three statistical metrics : i) the Root Mean Square Error (RMSE), corresponding to average magnitude of the errors between the reference data and other data sources. Defined as : $$\:RMSE=\sqrt{\frac{1}{N}{\sum\:}_{i=1}^{N}{\left({x}_{i}-{y}_{i}\right)}^{2}}$$ 1 ii) the Pattern Correlation Coefficient (PCC), represented by the azimuthal position. It is mathematically estimated as follows : $$\:PCC=\frac{1}{{\sigma\:}_{x}{\sigma\:}_{y}}\left[\frac{1}{N}{\sum\:}_{i=1}^{N}\left({x}_{i}-\overline{x}\right)\left({y}_{i}-\overline{y}\right)\right]$$ 2 iii) the Standard Deviation (SD): represented by the radial distance from origin of the diagram. It is the magnitude of the spread of the data around its mean, showing the amount of variation or dispersion of a set of values. It is expressed as : $$\:{\sigma\:}_{x}=\sqrt{\frac{1}{N}{\sum\:}_{i=1}^{N}{\left({x}_{i}-\overline{x}\right)}^{2}}$$ 3 and $$\:{\sigma\:}_{y}=\sqrt{\frac{1}{N}{\sum\:}_{i=1}^{N}{\left({y}_{i}-\overline{y}\right)}^{2}}$$ 4 The agreement between simulations and observation is quantified by their correlation and variability. Thus, A model is deemed to perform well if it satisfies the following criteria: (a) PCC ≥ 0.6, (b) SD within 1 ± 0.25, and (c) RMSE < 1. The portrait diagram analysis, which yields a compact overview of the similarities and differences between simulated and observed datasets for selected statistical parameters (including BIAS, RMSE, and PCC), thereby identifying potential areas of model agreement and bias. Small BIAS refers to less model errors. The BIAS is calculated as : $$\:BIAS=\frac{1}{N}{\sum\:}_{i=1}^{N}\left({x}_{i}-{y}_{i}\right)$$ 5 The Taylor Skill Score (TSS; Taylor 2001 ) is a numerical metric that integrates several statistical measures, including normalized SD and PCC, to assess how accurately models reproduce observed patterns and variability. PCC 0 is the maximum achievable correlation (taken as 1) for the TSS calculation. The TSS is calculated as : $$\:TSS=\frac{4{\left(1+PCC\right)}^{2}}{{\left(\frac{{\sigma\:}_{x}}{{\sigma\:}_{y}}+\frac{{\sigma\:}_{y}}{{\sigma\:}_{x}}\right)}^{2}{\left(1+{PCC}_{0}\right)}^{2}}$$ 6 A higher TSS signifies a model's greater accuracy in capturing the spatial patterns and magnitude of observed extreme temperature events. In these equations, x i corresponds to the model's value, while y i represents the observed reference value to the ith grid point. N indicates the number of data points. The terms \(\:\overline{x}\) and \(\:\overline{y}\) are the respective means of the model and reference values. To evaluate the ability of CMIP6 models to represent the spatial patterns of extreme temperature events over Central Africa, the spatial distribution of each of the seven relevant indices was first examined for both the models and the observational datasets. Subsequently, a Taylor diagram analysis was performed to provide a global statistical overview of model performance across the entire Central African domain. To gain deeper insights into the model performance at a local level, the analysis was further conducted for each of the five identified sub-regions using portrait diagrams and the Taylor Skill Score. 3 Results and Discussion 3.1 Spatial variability of percentile-based indices Figure 2 displays the spatial distribution of the 90th percentile of maximum temperature (T90) during the 1985–2014 period over CA, derived from the observations CHIRTS and ERA5, as well as the individual CMIP6 simulations and their MME. The T90 represents a threshold for what is considered a relatively warm temperature for a location during a specific period. The distribution of the T90 derived from the CHIRTS dataset ( Fig. 2 a ) shows values ranging from 24 to 46°C, with the highest records (≥ 40°C) located north of 9°N. The T90 presented by ERA5 ( Fig. 2 b ) has similar patterns to those of CHIRTS, though less intense by about 2°C in most regions of the domain (see supplementary material, Fig. S2a). Compared with CHIRTS, ERA5 showed good agreement in the representation of the T90, with RMSE/SD/PCC of about 0.30/0.90/0.96 (see supplementary material, Fig. S1 ; T90). Compared to CHIRTS, the MME and individual CMIP6 models generally capture the spatial pattern of the T90 ( Fig. 2 c-s ). However, regional differences exist. Compared to CHIRTS, CMIP6 models typically show RMSE values ranging from 0 to 0.5, SD from 0.75 to 1.25, and PCC from 0.85 to 0.96 (see Supplementary, Fig. S1 ; T90). MIROC6 exhibits slightly higher variability than other models. Across the study area, biases between the MME as well as the individual CMIP6 models and CHIRTS are mostly between 0 and ± 4°C (see Supplementary, Fig. S2b-r). Furthermore, outside the equatorial regions (EQE, EQW, and NE), MIROC6 overestimates the T90, with biases ranging from 4 to 10°C (see Supplementary, Fig. S2l). Figure 3 shows the spatial distribution of the 10th percentile of minimum temperature (T10) over the CA domain for 1985–2014, derived from CHIRTS, ERA5, individual CMIP6 simulations, and the MME. The T10 represents a threshold for relatively cold temperatures at a given location over a period. The T10 as seen by CHIRTS ( Fig. 3 a ) varies from 8°C to 24°C, with Zambia and northern Angola experiencing the warmest of these, while southern Sudan and the Central African Republic see the coolest. ERA5 ( Fig. 3 b ) depicts a similar pattern, though it exhibits less intense temperatures. Overall, ERA5 and CHIRTS match well, with strong agreement in RMSE/SD/PCC of 0.40/1.10/0.93 (see supplementary material, Fig. S1 ; T10). CMIP6 models and their multi-model ensemble (MME) accurately simulate the spatial pattern of the T10 when compared to CHIRTS and ERA5 observations ( Fig. 3 c-s ). Despite this, consistent underestimation of 1 to 6°C are observed across the study area when using CHIRTS as a reference (see Supplementary, Fig. S3b-r). These biases are evident in ERA5, the MME, and individual CMIP6 models. Among the models, MPI-ESM1-2-HR and MPI-ESM1-2-LR show the best performance with minimal biases, while INM-CM5-0 displays the largest. However, CMIP6 models confirm their good spatial agreement with CHIRTS, providing RMSE from 0 to 1, SD from 0.75 to 1.40, and PCC from 0.60 to 0.90 (see Supplementary, Fig. S1 ; T10). Figures 4 and 5 display the spatial distribution of the annual percentage of warm days (TX90p) and cool nights (TN10p), respectively, during the 1985–2014 period, obtained from the observations CHIRTS and ERA5, as well as the individual CMIP6 simulations and the MME. Specifically, TX90p is the percentage of days with daily maximum temperatures above the 90th percentile, and TN10p is the percentage of days with daily minimum temperatures below the 10th percentile. The spatial patterns of TX90p/TN10p shown by CHIRTS ( Figs. 4 a and 5 a ) are highly heterogeneous, though the magnitudes vary less (9.8–10.2%) over the study area. In terms of magnitude, ERA5, MME, and individual models reproduce similar patterns to CHIRTS, providing low biases over the domain. However, it is worth noting that poor correlation values (below 0.2) are noted among all the datasets (see Supplementary, Fig. S1 ; TX90p and TN10p). CHIRTS demonstrates spatially heterogeneous distributions of TX90p/TN10p ( Figs. 4 a and 5 a ), with magnitudes varying between 9.8% and 10.2%. ERA5, MME, and individual models reproduce similar spatial patterns to CHIRTS, with low biases of approximately ± 0.2% (Supplementary Figures S4 and S5). Despite this, correlation coefficients across all datasets are below 0.2 (Supplementary Figure S1 ). The low correlations observed in TX90p/TN10p between different data sources are not necessarily a sign of failure to mimic observations, but, rather, they suggest a high degree of spatio-temporal variability in these indices. As mentioned by Zwiers et al. ( 2013 ), climate research often focuses on indicators of event frequency and intensity, such as "moderate extremes" defined by percentile thresholds (e.g., 90th/10th), which, while within seasonal observation ranges, are valuable for trend analysis, a point relevant to the observed variability. 3.2 Spatial variability of absolute and duration-based indices Figure 6 presents the spatial distribution of the diurnal temperature range (DTR) across the CA domain for 1985–2014, using data from CHIRTS, ERA5, individual CMIP6 models, and their MME. The DTR reflects how much the temperature changes from the hottest part of the day to the coldest part of the night. According to CHIRTS ( Fig. 6 a ), the DTR starts at about 5°C in coastal areas and gradually increases to around 9°C in regions closer to the Equator. Moving away from the Equator, the DTR shows a noticeable increase, peaking at about 15°C in the northern and southern extremities of the CA domain. ERA5 ( Fig. 6 b ) shows a highly consistent DTR pattern with CHIRTS, demonstrating strong agreement with RMSE of 0.12, SD of 1.05, and PCC of 0.99 (see supplementary material, Fig. S1 ; DTR). Figures 6 c-s show that the MME and the individual CMIP6 models successfully simulate the spatial pattern of the DTR observed in CHIRTS and ERA5. While biases across individual models range from 1 to 6°C, the MME, along with BCC-ESM1, UKESM1-LL, EC-Earth3, E3SM-2-0, and SAM0-UNICON, exhibit lower bias, generally ≤ 2°C (see Supplementary, Fig. S6). Indeed, these models display better alignment with CHIRTS, reflected in RMSE/SD/PCC values between 0.25 and 0.75 / 0.75 and 1.25 / 0.70 and 0.95 (see Supplementary, Fig. S1 ; DTR). Preceding Donat et al. ( 2014 ), studies indicated that DTR across northern and western areas of Africa is linked to both the El Niño-Southern Oscillation (ENSO) and the North Atlantic Oscillation (NAO), with the latter showing a stronger association, particularly evident in the western Arab region bordering the Atlantic. This observation suggests the possibility of analogous correlations between DTR and these climate oscillations in CA, which warrant investigation using CMIP6 model simulations. Figures 7 and 8 display the spatial distributions of the annual Warm Spell Days Index (WSDI) and Cold Spell Days Index (CSDI) across CA from 1985 to 2014, using CHIRTS and ERA5 observations with CMIP6 model simulations and their MME. WSDI indicates the length of prolonged warm spells, and CSDI that of cold spells. For the WSDI ( Fig. 7 ), ERA5 and CHIRTS display consistent spatial patterns, with values increasing from less than 10 days in equatorial regions (NE, EQW, EQE) to approximately 25 days in the northern and southern areas (SS and SE). Strong agreement between ERA5 and CHIRTS is supported by RMSE, SD, and PCC values within the ranges of 0.25–0.75, 0.25–0.50, and 0.85–0.95, respectively (see Supplementary, Fig. S1 ; WSDI and CSDI). The MME and most CMIP6 models reproduce the overall WSDI patterns but tend to overestimate warm spell durations, especially in the Democratic Republic of Congo (DRC), where biases reach 20 days (see Supplementary, Fig. S7). However, NorESM2-MM and SAM0-UNICON outperform the other models, with RMSE/SD/PCC closer to that of ERA5. Regarding the Cold Spell Days Index (CSDI) in Fig. 8 , CMIP6 models and the MME tend to overestimate compared to CHIRTS. This overestimation aligns with the simulated WSDI patterns, though the PCC is notably weaker, below 0.6 across most models (see Supplementary, Fig. S1 ; CSDI). The observed WSDI/CSDI patterns in the region may be related to the El Niño–Southern Oscillation (ENSO), potentially following warm/cold ENSO events, as demonstrated by Moron et al. ( 2016 ) in the tropical North of Africa. 3.3 Statistical Characterization of Extreme Temperature Indices within Climatic Sub-regions We examined the ability of CMIP6 models in reproducing observed extreme temperature indices across the diverse climatic sub-regions of CA. This involved a detailed statistical evaluation using bias, RMSE, PCC, and TSS, providing insights into model accuracy at a localized scale. Using CHIRTS observations as the reference data , Fig. 9 displays portrait diagrams of seven extreme temperature indices calculated for the five sub-regions. As shown in the first column of Fig. 9 , the majority of biases are relatively minor, staying within the range of -20–20% across all the sub-regions. Specifically, the T90 and T10 indices tend to show negative biases across these sub-regions, whereas the WSDI and CSDI indices more often exhibit positive biases. Notably, WSDI and CSDI exhibit the most significant biases, with some CMIP6 models and their MMEs overestimating by more than 60% in the EQW, EQE, and SE sub-regions. It's worth noting that the MME typically reduces bias compared to individual models for the most temperature indices and across all sub-regions. The notable exception is the WSDI index in the EQW, EQE, and SE sub-regions, where substantial positive biases persist even in the MME. Regarding the RMSE (see Fig. 9 , second column), CMIP6 models and their MMEs generally show good agreement with observations (RMSE < 1) across all five sub-regions for most temperature indices, except for CSDI and WSDI. Consistent with bias findings, simulations tend to provide higher RMSE values (greater than 1) for both CSDI and WSDI indices across the EQE, EQW, and SE sub-regions. The most significant errors are displayed for WSDI (RMSE > 2 in EQE and EQW). However, for WSDI, it's worth noting that BCC-ESM1, NorESM2-MM, and NorESM2-LM models perform better, achieving RMSE values comparable to the ERA5 dataset. For the PCC (see Fig. 9 , third column), most CMIP6 models show good correlation (PCC > 0.6) with CHIRTS observations across several sub-regions for specific indices: T10, T90, and DTR in SS and NE; T10, T90, and WSDI in EQE; T10 and T90 in EQW; and T10, WSDI, and CSDI in NE. Conversely, all CMIP6 models exhibit lower correlation (PCC < 0.6) for TN10p and TX90p. Similarly, ERA5 reanalysis also shows low correlation for these two indices across different areas, indicating potential discrepancies between observation sources ( Komelo et al. 2024 ). However, consistent with bias and RMSE results, the MME generally demonstrates a better PCC for all indices compared to individual models. Overall, the representation of the majority of extreme temperature indices by the MME agrees with CHIRTS observations over the different climatic sub-regions, except for TN10p and TX90p where low PCC values are recorded. Moreover, when compared to the observations, errors or differences in both mean temperature and extreme temperature indices are generally a problem common to climate models ( Zebaze et al. 2019 ; Mengistu et al. 2024 ), and our results here are consistent with those studies over the African domains ( Ntoumos et al. 2020 ; Fotso-Kamga et al. 2023 ). The study by Mengistu et al. ( 2024 ) on extreme temperature events in major South African cities using CMIP6 models has suggested adjusted datasets that show fairly good results for different statistical metrics. In addition, our findings show that the accurate representation of the MME is often the result of the overestimation from some models being balanced out by the underestimation from others in the ensemble mean, rather than due to optimal parameterization. As suggested by Sonkoué et al. ( 2019 ), a preliminary selection of good models before making an MME is crucial to provide consistent and accurate results. On the other hand , Figs. 10 and 11 display the TSS metric results for the seven extreme temperature indices across the five climatic subregions, with CHIRTS observations used as reference data. For the T90 ( Fig. 10 a ), a large number of individual models demonstrate satisfactory performance (TSS ≥ 0.6) over SS and NE sub-regions. While individual model competence is somewhat lower in EQE and EQW, SE records the fewest satisfactory individual models (EC-Earth3, GFDL-ESM4, and NorESM2-MM). The MME records satisfactory performance over SS, NE, EQE, and EQW. For the T10 ( Fig. 10 b ), the majority of the individual models achieve satisfactory TSS scores (≥ 0.6) across the five sub-regions. Some simulations show strong individual model performance (CanESM5, CNRM-CM6-1, HadGEM3-GC31-LL, and IPSL-CM6A-LR) over all the sub-regions. For this index, the MME records satisfactory performance over all five sub-regions. For the DTR ( Fig. 10 c ), the model performance is more varied. A moderate number of individual models are satisfactory in SS and EQW. NE has a relatively high number of well-performing individual models (with TSS ≥ 0.6). In contrast, few individual models are satisfactory in EQE (EC-Earth3, HadGEM3-GC31-LL, and UKESM1-0-L) and SE (E3SM-2-0, EC-Earth3, MIROC6, and NorESM2-MM). Here, the MME shows satisfactory performance over SS, NE, and EQW. The WSDI ( Fig. 11 a ) is generally challenging for most individual models. Satisfactory performance (TSS ≥ 0.6) is noted for NorESM2-LM and CanESM5 in some areas, but almost all models perform well only over SE. Few models are satisfactory in other sub-regions, with SS having the fewest (NorESM2-LM and NorESM2-MM). The MME performs satisfactorily for WSDI over the NE, EQW, and SE. Regarding the CSDI ( Fig. 11 b ), most individual models do not accurately capture the observations over sub-regions. While a few individual models are satisfactory with TSS ≥ 0.6 in SS (BCC-ESM1 and UKESM1-0-L), NE (BCC-ESM1, EC-Earth3, HadGEM3-GC31-LL, and MIROC6), EQE (E3SM-2-0, EC-Earth3, GFDL-ESM4, MIROC6, and SAM0-UNICON), and EQW (HadGEM3-GC31-LL), SE has no satisfactory model. The MME shows satisfactory performance over NE and EQW. For TX90p and TN10p ( Fig. 11 c, d ), both the individual CMIP6 models and the MME consistently yield low TSS values (< 0.4) across all five sub-regions. This poor performance is in agreement with the PCC results shown in Fig. 9 . Overall, the number of models achieving satisfactory performance (TSS ≥ 0.6) varies significantly depending on the extreme temperature index and the sub-region. This suggests that different models have varying strengths and weaknesses in simulating different aspects of extreme temperature variability across different climatic zones. However, the MME generally shows better performance compared to individual CMIP6 models. Across the five sub-regions, both individual CMIP6 models and the MME tend to have a larger number of well-performing models over the SS and NE sub-regions compared to EQE, EQW, and SE. This could reflect differences in the complexity of the climate variability and processes in these regions. In fact, a unimodal precipitation distribution characterizes the SS, NE, and SE sub-regions of CA, while a bimodal distribution is observed in the EQE and EQW sub-regions, according to the findings of Fotso-Kamga et al. (2019). This could indicate the importance of seasons in the occurrence of extreme temperature events. For instance, throughout seasons, the occurrence of both warm and cold temperature extremes is influenced by factors determining the overall seasonal climate, such as sunlight, elevation, and proximity to the ocean, as well as natural climate processes like ENSO ( Miralles et al. 2014 ; Lewis and King 2015 ). The complex interplay of these factors across different timescales ( Sillmann et al. 2017 ) contributes to the challenge faced by climate models in accurately reproducing the variability of extreme temperature indices. Notably, the present study reveals that CMIP6 models do not consistently achieve a satisfactory TSS (≥ 0.6) for the TX90p and TN10p indices across all five sub-regions. This strongly suggests a significant challenge for these models in accurately representing the frequency of extreme warm and cold temperature events over the CA domain, potentially originating from issues with the simulation of the temperature distribution's tails, often registered in moderate extremes ( Zwiers et al. 2013 ). Besides, even ERA5 reanalysis shows low consistency with the reference CHIRTS for both TX90p and TN10p indices over the five sub-regions, raising the problem of discrepancies among different sources of observations ( Camberlin et al. 2019 ), which often contributes to varying performance of models depending on the considered chosen dataset as reference (Fotso-Kamga et al. 2019; Taguela et al. 2020 ). 4 Summary and Conclusion This study evaluated the representation of extreme temperature events as simulated by CMIP6 models over Central Africa (CA), with a focus on their spatial variability across five climatic sub-regions. We analyzed several percentile-based, absolute, and duration-based extreme temperature indices, comparing the simulations from individual CMIP6 models and their Multi-Model Ensemble (MME) against CHIRTS observations, with consideration of ERA5 reanalysis as well. Our analysis reveals that CMIP6 models generally show agreement with CHIRTS observations for the majority of extreme temperature indices across the CA sub-regions. Furthermore, the MME often demonstrates better agreement than individual models, with its accuracy frequently arising from the compensation of overestimations and underestimations across the ensemble. However, CMIP6 models, including the MME, do not consistently capture CHIRTS observations for the TX90p and TN10p indices across all sub-regions. The number of models achieving satisfactory performance varied considerably depending on the specific extreme temperature index and the sub-region under consideration. Notably, both individual models and the MME tended to perform better over the Sudano-Sahelian (SS) and Northern Equatorial (NE) sub-regions compared to the Equatorial East (EQE), Equatorial West (EQW), and Southern Equatorial (SE) sub-regions. This disparity in performance may be linked to the complexity of climate variability and processes in these regions, such as the unimodal precipitation distribution in SS and NE versus the bimodal distribution in EQE and EQW, as suggested by previous research. In conclusion, while the CMIP6 models show reasonable ability to represent many extreme temperature indices over Central Africa, challenges remain, particularly for the frequency of the percentile-based indices like TX90p and TN10p, and in specific sub-regions. These findings underscore the need for continued evaluation and potential improvements in climate models to enhance their accuracy in simulating the full spectrum of extreme temperature events across this diverse region. Improving the simulation of these extremes is crucial for better understanding potential climate change impacts and informing regional adaptation strategies. Declarations Conflicts of Interest The authors confirm that there are no conflicts of interest associated with this work. Author Contribution G.F-K., T.C.F-N and A.D. conceptualized the study. G.F-K, T.C.F-N., A.D., Z.D.Y. and D.A.V. defined the methodology. G.F-K, T.C.F-N., S.Z., Z.N. and A.T.T. conducted the analysis. G.F-K. and T.C.F-N. wrote the main manuscript text. All authors reviewed the manuscript. Acknowledgement We gratefully thank the climate modeling groups detailed in Table 1 for their essential work in producing and making their model output openly accessible through the Earth System Grid Federation (ESGF) platforms. Furthermore, we would like to express our gratitude to the LMI-NEXUS initiative for their support during the realization of this work. References Abatan AA, Abiodun BJ, Lawal KA, Gutowski WJ Jr (2016) Trends in extreme temperature over Nigeria from percentile-based threshold indices. Int J Climatol 36:2527–2540. https://doi.org/10.1002/joc.4539 Aguilar E, Aziz Barry A, Brunet M, Ekang L, Fernandes A, Massoukina M, Zhang X (2009) Changes in temperature and precipitation extremes in western central Africa, Guinea Conakry, and Zimbabwe, 1955–2006. J Geophys Res Atmos 114:D2. https://doi.org/10.1029/2008JD011075 Allen M, Dube O, Solecki W, Aragón-Durand F, Cramer W, Humphreys S, Kainuma M, Kala J, Mahowald N, Mulugetta Y et al (2018) Chap. 1: Framing and Context. In: Masson-Delmotte V, Zhai P, Pörtner H-O, Roberts D, Skea J, Shukla PR, Pirani A, Moufouma-Okia W (eds) IPCC Special Report: Global Warming of 1.5°C. Cambridge University Press Boucher O, Servonnat J, Albright AL, Aumont O, Balkanski Y, Bastrikov V, Vuichard N (2020) Presentation and evaluation of the IPSL-CM6A‐LR climate model. J Adv Model Earth Syst 12:e2019MS002010. https://doi.org/10.1029/2019MS002010 Camberlin P, Barraud G, Bigot S, Dewitte O, Makanzu Imwangana F, Maki Mateso JC, Samba G (2019) Evaluation of remotely sensed rainfall products over Central Africa. Q J R Meteorol Soc 145:2115–2138. https://doi.org/10.1002/qj.3540 Donat MG, Alexander LV, Yang H, Durre I, Vose R, Dunn RJ, Kitching S (2013) Updated analyses of temperature and precipitation extreme indices since the beginning of the twentieth century: The HadEX2 dataset. J Geophys Res Atmos 118:2098–2118. https://doi.org/10.1002/jgrd.50150 Donat MG, Peterson TC, Brunet M, King AD, Almazroui M, Kolli RK, Rhaz E, K (2014) Changes in extreme temperature and precipitation in the Arab region: long-term trends and variability related to ENSO and NAO. Int J Climatol 34. https://doi.org/10.1002/joc.3703 Dosio A, Jury MW, Almazroui M, Ashfaq M, Diallo I, Engelbrecht FA, Tamoffo AT (2021) Projected future daily characteristics of African precipitation based on global (CMIP5, CMIP6) and regional (CORDEX, CORDEX-CORE) climate models. Clim Dyn 57:3135–3158. https://doi.org/10.1007/s00382-021-05741-z Döscher R, Acosta M, Alessandri A, Anthoni P, Arneth A, Arsouze T, Zhang Q (2021) The EC-earth3 Earth system model for the climate model intercomparison project 6. Geosci Model Dev Discuss 2021:1–90. https://doi.org/10.5194/gmd-2021-26 Dunn RJ, Alexander LV, Donat MG, Zhang X, Bador M, Herold N, Yussof BH, M. N. A (2020) Development of an updated global land in situ-based data set of temperature and precipitation extremes: HadEX3. J Geophys Res Atmos 125. https://doi.org/10.1029/2019JD032263 . e2019JD032263 Ebi KL, Capon A, Berry P, Broderick C, de Dear R, Havenith G, Honda Y, Kovats RS, Ma W, Malik A et al (2021) Hot weather and heat extremes: Health risks. Lancet 398:698–708. https://doi.org/10.1016/S0140-6736(21)01053-0 Eyring V, Bony S, Meehl GA, Senior CA, Stevens B, Stouffer RJ, Taylor KE (2016) Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. Geosci Model Dev 9:1937–1958. https://doi.org/10.5194/gmd-9-1937-2016 Fotso-Kamga G, Fotso‐Nguemo TC, Diallo I, Yepdo ZD, Pokam WM, Vondou DA, Lenouo A (2020) An evaluation of COSMO‐CLM regional climate model in simulating precipitation over Central Africa. Int J Climatol 40:2891–2912. https://doi.org/10.1002/joc.6366 Fotso-Kamga G, Fotso‐Nguemo TC, Diallo I, Nyanchi GT, Yepdo ZD, Chouto S, Lenouo A (2023) Projected changes in extreme precipitation and temperature events over Central Africa from COSMO‐CLM simulations under the global warming level of 1.5° C and above. Int J Climatol 43:6330–6351. https://doi.org/10.1002/joc.8183 Fotso-Nguemo TC, Vondou DA, Diallo I, Diedhiou A, Weber T, Tanessong RS et al (2022) Potential impact of 1.5, 2 and 3 C global warming levels on heat and discomfort indices changes over Central Africa. Sci Total Environ 804:1–11. https://doi.org/10.1016/j.scitotenv.2021.150117 Fotso-Nguemo TC, Weber T, Diedhiou A, Chouto S, Vondou DA, Rechid D, Jacob D (2023) Projected impact of increased global warming on heat stress and exposed population over Africa. Earths Future 11. https://doi.org/10.1029/2022EF003268 . e2022EF003268 Funk C, Peterson P, Peterson S, Shukla S, Davenport F, Michaelsen J, Mata N (2019) A high-resolution 1983–2016 T max climate data record based on infrared temperatures and stations by the Climate Hazard Center. J Clim 32:5639–5658. https://doi.org/10.1175/JCLI-D-18-0402.1 Golaz JC, Van Roekel LP, Zheng X, Roberts AF, Wolfe JD, Lin W, Bader DC (2022) The DOE E3SM model version 2: Overview of the physical model and initial model evaluation. J Adv Model Earth Syst 14. https://doi.org/10.1029/2022MS003156 . e2022MS003156 Hersbach H, Bell B, Berrisford P, Hirahara S, Horányi A, Muñoz-Sabater J, Thépaut JN (2020) The ERA5 global reanalysis. Q J R Meteorol Soc 146:1999–2049. https://doi.org/10.1002/qj.3803 Horowitz LW, John JG, Blanton C, McHugh C, Radhakrishnan A, Rand K, Zeng Y (2018) NOAA-GFDL GFDL-ESM4 model output prepared for CMIP6 DAMIP hist-aer. https://doi.org/10.22033/ESGF/CMIP6.13327 . World Data Center for Climate (WDCC) at DKRZ Komelo CK, Fotso-Nguemo TC, Ngavom Z, Dessacka AK, Taguela TN, Yepdo ZD, Tchawoua C (2024) Evaluation of extreme precipitation events as simulated by CMIP6 models over Central Africa: Spatial patterns. Theor Appl Climatol 155:9579–9599. https://doi.org/10.1007/s00704-024-04987-9 Lewis SC, King AD (2015) Dramatically increased rate of observed hot record breaking in recent Australian temperatures. Geophys Res Lett 42:7776–7784. https://doi.org/10.1002/2015GL065555 Lu C, Sun Y, Zhang X (2018) Multimodel detection and attribution of changes in warm and cold spell durations. Environ Res Lett 13:074013. https://doi.org/10.1088/1748-9326/aac539 Mauritsen T, Bader J, Becker T, Behrens J, Bittner M, Brokopf R, Roeckner E (2019) Developments in the MPI-M Earth System Model version 1.2 (MPI‐ESM1. 2) and its response to increasing CO2. J Adv Model Earth Syst 11:998–1038. https://doi.org/10.1029/2018MS001400 Mbienda AJK, Guenang GM, Kaissassou S, Sandjon AT, Tanessong RS, Lontio SLM, Djomou ZY (2022) Boundary layer schemes in the regional climate model RegCM4. 6 over Central Africa. Clim Dyn 58:691–709. https://doi.org/10.1007/s00382-021-05971-4 Mengistu AG, Woyessa YE, Tesfuhuney WA, Steyn AS, Lee SS (2024) Assessing the impact of climate change on future extreme temperature events in major South African cities. Theor Appl Climatol 155:1807–1819. https://doi.org/10.1007/s00704-023-04770-0 Miralles DG, Teuling AJ, Van Heerwaarden CC (2014) & Vilà-Guerau de Arellano, J. Mega-heatwave temperatures due to combined soil desiccation and atmospheric heat accumulation. Nature Geosci 7:345–349. https://doi.org/10.1038/ngeo2147 Moron V, Oueslati B, Pohl B, Rome S, Janicot S (2016) Trends of mean temperatures and warm extremes in northern tropical Africa (1961–2014) from observed and PPCA-reconstructed time series. J Geophys Res Atmos 121:5298–5319. https://doi.org/10.1002/2015JD024103 Müller WA, Jungclaus JH, Mauritsen T, Baehr J, Bittner M, Budich R, Marotzke J (2018) A higher-resolution version of the max planck institute earth system model (MPI‐ESM1. 2‐HR). J Adv Model Earth Syst 10:1383–1413. https://doi.org/10.1029/2017MS001211 Ngavom Z, Fotso-Nguemo TC, Vondou DA, Fotso-Kamga G, Zebaze S, Yepdo ZD, Diedhiou A (2024) Projected changes in population exposure to extreme precipitation events over Central Africa under the global warming levels of 1.5° C and 2° C: insights from CMIP6 simulations. Model Earth Syst Environ 10:5753–5769. https://doi.org/10.1007/s40808-024-01968-3 Ntoumos A, Hadjinicolaou P, Zittis G, Lelieveld J (2020) Updated assessment of temperature extremes over the Middle East–North Africa (MENA) region from observational and CMIP5 data. Atmosphere 11:813. https://doi.org/10.3390/atmos11080813 Ozturk T, Saygili-Araci FS, Kurnaz ML (2021) Projected changes in extreme temperature and precipitation indices over CORDEX-MENA domain. Atmosphere 12:622. https://doi.org/10.3390/atmos12050622 Park S, Shin J, Kim S, Oh E, Kim Y (2019) Global climate simulated by the Seoul National University atmosphere model version 0 with a unified convection scheme (SAM0-UNICON). J Clim 32:2917–2949. https://doi.org/10.1175/JCLI-D-18-0157.1 Parsons D, Stern D, Ndanguza D, Sylla MB (2022) Evaluation of satellite-based air temperature estimates at eight diverse sites in Africa. Climate 10:98. https://doi.org/10.3390/cli10070098 Peel MC, Finlayson BL, McMahon TA (2007) Updated world map of the Köppen-Geiger climate classification. Hydrol Earth Syst Sci 11:1633–1644. https://doi.org/10.5194/hess-11-1633-2007 Perkins SE, Alexander LV, Nairn JR (2012) Increasing frequency, intensity and duration of observed global heatwaves and warm spells. Geophys Res Lett 39. https://doi.org/10.1029/2012GL053624 Ryti NR, Guo Y, Jaakkola JJ (2016) Global association of cold spells and adverse health effects: a systematic review and meta-analysis. Environ Health Perspect 124:12–22. https://doi.org/10.1289/ehp.1408794 Seland Ø, Bentsen M, Olivié D, Toniazzo T, Gjermundsen A, Graff LS, Schulz M (2020) Overview of the Norwegian Earth System Model (NorESM2) and key climate response of CMIP6 DECK, historical, and scenario simulations. Geosci Model Dev 13:6165–6200. https://doi.org/10.5194/gmd-13-6165-2020 Seneviratne SI, Nicholls N, Easterling D, Goodess CM, Kanae S, Kossin J, Luo Y, Marengo J, McInnes K, Rahimi M, Reichstein M, Sorteberg A, Vera C, Zhang X (2012) Changes in climate extremes and their impacts on the natural physical environment. In: Field C. B., Barros V., Stocker T. F., Qin D., Dokken D., Ebi K. L., Mastrandrea M. D., Mach K. J., Plattner G.-K., Allen S. K., Tignor M., Midgley P. M. (Eds.), Intergovernmental Panel on Climate Change special report on managing the risks of extreme events and disasters to advance climate change adaptation, pp. 109–230. Cambridge University Press Sillmann J, Thorarinsdottir T, Keenlyside N, Schaller N, Alexander LV, Hegerl G, Zwiers FW (2017) Understanding, modeling and predicting weather and climate extremes: Challenges and opportunities. Weather Clim Extremes 18:65–74. https://doi.org/10.1016/j.wace.2017.03.003 Sonkoué D, Monkam D, Fotso-Nguemo TC, Yepdo ZD, Vondou DA (2019) Evaluation and projected changes in daily rainfall characteristics over Central Africa based on a multi-model ensemble mean of CMIP5 simulations. Theor Appl Climatol 137:2167–2186. https://doi.org/10.1007/s00704-018-2708-8 Swart NC, Cole JN, Kharin VV, Lazare M, Scinocca JF, Gillett NP, Sigmond M (2019) CCCma CanESM5 model output prepared for CMIP6 FAFMIP. https://doi.org/10.22033/ESGF/CMIP6.13670 . World Data Center for Climate (WDCC) at DKRZ Taguela TN, Vondou DA, Moufouma-Okia W, Fotso‐Nguemo TC, Pokam WM, Tanessong RS, Tchotchou D, L. A (2020) CORDEX multi‐RCM hindcast over Central Africa: Evaluation within observational uncertainty. J Geophys Res Atmos 125. https://doi.org/10.1029/2019JD031607 . e2019JD031607 Tamoffo AT, Vondou DA, Pokam WM, Haensler A, Yepdo ZD, Fotso-Nguemo TC, Nouayou R (2019) Daily characteristics of Central African rainfall in the REMO model. Theor Appl Climatol 137:2351–2368. https://doi.org/10.1007/s00704-018-2720-z Tamoffo AT, Weber T, Akinsanola AA, Vondou DA (2023) Projected changes in extreme rainfall and temperature events and possible implications for Cameroon's socio-economic sectors. Meteorol Appl 30:e2119. https://doi.org/10.1002/met.2119 Tang Y, Rumbold S, Ellis R, Kelley D, Mulcahy J, Sellar A, Jones C (2019) MOHC UKESM1. 0-LL model output prepared for CMIP6 CMIP historical. https://doi.org/10.22033/ESGF/CMIP6.13400 . World Data Center for Climate (WDCC) at DKRZ Tatebe H, Ogura T, Nitta T, Komuro Y, Ogochi K, Takemura T, Kimoto M (2019) Description and basic evaluation of simulated mean state, internal variability, and climate sensitivity in MIROC6. Geosci Model Dev 12:2727–2765. https://doi.org/10.5194/gmd-12-2727-2019 Taylor KE (2001) Summarizing multiple aspects of model performance in a single diagram. J Geophys Res Atmos 106:7183–7192. https://doi.org/10.1029/2000JD900719 van Der Walt AJ, Fitchett JM (2021) Trend analysis of cold extremes in South Africa: 1960–2016. Int J Climatol 41:2060–2081. https://doi.org/10.1002/joc.6828 Voldoire A, Saint-Martin D, Sénési S, Decharme B, Alias A, Chevallier M, Waldman R (2019) Evaluation of CMIP6 deck experiments with CNRM‐CM6‐1. J Adv Model Earth Syst 11:2177–2213. https://doi.org/10.1029/2018MS001683 Volodin E, Mortikov E, Gritsun A, Lykossov V, Galin V, Diansky N, Emelina S (2019) IPCC DDC: INM INM-CM5-H model output prepared for CMIP6 HighResMIP control-1950. https://doi.org/10.22033/ESGF/CMIP6.13444 . World Data Center for Climate (WDCC) at DKRZ Williams C, Lunt D, Singarayer J, Guarino MV (2021) CMIP6 PMIP NERC HadGEM3-GC31-LL lig127k. https://doi.org/10.22033/ESGF/CMIP6.13601 . World Data Center for Climate (WDCC) at DKRZ Wu T, Zhang F, Zhang J, Jie W, Zhang Y, Wu F, Hu A (2020) Beijing Climate Center Earth System Model version 1 (BCC-ESM1): model description and evaluation of aerosol simulations. Geosci Model Dev 13:977–1005. https://doi.org/10.5194/gmd-13-977-2020 Zebaze S, Jain S, Salunke P, Shafiq S, Mishra SK (2019) Assessment of CMIP5 multimodel mean for the historical climate of Africa. Atmos Sci Lett 20:e926. https://doi.org/10.1002/asl.926 Zwiers FW, Alexander LV, Hegerl GC, Knutson TR, Kossin JP, Naveau P, Zhang X (2013) Climate extremes: challenges in estimating and understanding recent changes in the frequency and intensity of extreme climate and weather events. Climate science for serving society: research, modeling and prediction priorities. Springer, pp 339–389. https://doi.org/10.1007/978-94-007-6692-1_13 Additional Declarations No competing interests reported. Supplementary Files ESM.pdf Cite Share Download PDF Status: Published Journal Publication published 05 Mar, 2026 Read the published version in Theoretical and Applied Climatology → Version 1 posted Editorial decision: Revision requested 19 Sep, 2025 Reviews received at journal 12 Sep, 2025 Reviewers agreed at journal 18 Aug, 2025 Reviews received at journal 14 Aug, 2025 Reviewers agreed at journal 12 Aug, 2025 Reviewers invited by journal 12 Aug, 2025 Editor assigned by journal 20 Jul, 2025 Submission checks completed at journal 20 Jul, 2025 First submitted to journal 18 Jul, 2025 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-7155223","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":502269629,"identity":"428c26f2-ae64-4c3c-a56e-45716b65ea29","order_by":0,"name":"Gabriel Fotso-Kamga","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8ElEQVRIiWNgGAWjYBACCQglByIYHzAwHIDShLUYM/AwMDAbQLSAaCK1sEkQpUWy/XTyix8MBnL2/IePVfPU3JHjZ2Bm+4BPizRP7jbLHgYDYx6JtLTbPMeeGUs2MDPPwKdFjiF3mwEPw5/EHgkes9s8bIcTNxzgP4zXYXL8b7cZ/mEwSOzhP2NWzPMPpIWZGa8WaYnczY95QFoYcsyYeduI0CI54+02ZhmQX26kJUvO7TtsLNlMQIvE+dzNH98AQ4y9//DBD2++HZbjZ2/Gr4UBFB2M/yAsJh4QSVADUAk8Fhh/EFY9CkbBKBgFIxAAAFBgQ298indLAAAAAElFTkSuQmCC","orcid":"","institution":"University of Yaounde 1","correspondingAuthor":true,"prefix":"","firstName":"Gabriel","middleName":"","lastName":"Fotso-Kamga","suffix":""},{"id":502269630,"identity":"10f799e7-4c9a-4dff-bb13-2ac11bbc609a","order_by":1,"name":"Thierry C. Fotso-Nguemo","email":"","orcid":"","institution":"University of Yaounde 1","correspondingAuthor":false,"prefix":"","firstName":"Thierry","middleName":"C.","lastName":"Fotso-Nguemo","suffix":""},{"id":502269631,"identity":"a77ca1b7-8a73-428b-9c9d-a00275a9b23a","order_by":2,"name":"Zéphirin D. Yepdo","email":"","orcid":"","institution":"University of Yaounde 1","correspondingAuthor":false,"prefix":"","firstName":"Zéphirin","middleName":"D.","lastName":"Yepdo","suffix":""},{"id":502269632,"identity":"3ae92363-6d76-4d3a-b7ae-45cabe255d82","order_by":3,"name":"Alain T. Tamoffo","email":"","orcid":"","institution":"University of Yaounde 1","correspondingAuthor":false,"prefix":"","firstName":"Alain","middleName":"T.","lastName":"Tamoffo","suffix":""},{"id":502269633,"identity":"a95d9635-7f44-429b-897d-9e66ab6fdb38","order_by":4,"name":"Zakariahou Ngavom","email":"","orcid":"","institution":"University of Yaounde 1","correspondingAuthor":false,"prefix":"","firstName":"Zakariahou","middleName":"","lastName":"Ngavom","suffix":""},{"id":502269634,"identity":"cdc31c9f-20b8-4895-a177-bc372e30cb90","order_by":5,"name":"Sinclair Zebaze","email":"","orcid":"","institution":"University of Yaounde 1","correspondingAuthor":false,"prefix":"","firstName":"Sinclair","middleName":"","lastName":"Zebaze","suffix":""},{"id":502269635,"identity":"5067fd39-e18b-436b-88db-c8f58af40efd","order_by":6,"name":"Cyrille K. Kayimbo","email":"","orcid":"","institution":"University of Dschang","correspondingAuthor":false,"prefix":"","firstName":"Cyrille","middleName":"K.","lastName":"Kayimbo","suffix":""},{"id":502269636,"identity":"d6d200a2-8079-4c2b-8b11-946c4ee0e42c","order_by":7,"name":"Derbetini A. Vondou","email":"","orcid":"","institution":"University of Yaounde 1","correspondingAuthor":false,"prefix":"","firstName":"Derbetini","middleName":"A.","lastName":"Vondou","suffix":""},{"id":502269638,"identity":"635d6134-8162-4ed6-9d03-72dca3d32cf3","order_by":8,"name":"Arona Diedhiou","email":"","orcid":"","institution":"University of Grenoble Alpes, IRD, CNRS, Grenoble INP, IGE","correspondingAuthor":false,"prefix":"","firstName":"Arona","middleName":"","lastName":"Diedhiou","suffix":""}],"badges":[],"createdAt":"2025-07-18 07:53:27","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7155223/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7155223/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00704-026-06065-6","type":"published","date":"2026-03-05T15:58:40+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":89488163,"identity":"62bd0df6-726f-4b20-a2e3-39658de560a3","added_by":"auto","created_at":"2025-08-20 13:14:54","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1018355,"visible":true,"origin":"","legend":"\u003cp\u003eThe study domain (15° S–15° N, 5° –35° E) with the five climatic sub-regions namely: Sudano-Sahelian (SS), Northern Equatorial (NE), Equatorial East (EQE), Equatorial West (EQW), and Southern Equatorial (SE).\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/358bb5515fd15020fa511ad3.png"},{"id":89488162,"identity":"877cf4a8-6aff-4d55-89d0-94262f196708","added_by":"auto","created_at":"2025-08-20 13:14:54","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":995675,"visible":true,"origin":"","legend":"\u003cp\u003eMean (1985–2014) spatial distribution of the 90th percentile of maximum temperature (T90; in °C), derived from a) CHIRTS, b) ERA5, c) MME and d-s) individual CMIP6 simulation members.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/5d208d6b0533a93231f5ba03.png"},{"id":89488490,"identity":"4aeeb55a-0dc1-4942-bd74-77d2d3ec3d43","added_by":"auto","created_at":"2025-08-20 13:22:54","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1144033,"visible":true,"origin":"","legend":"\u003cp\u003eMean (1985–2014) spatial distribution of the 10th percentile of minimum temperature (T10; in °C), derived from a) CHIRTS, b) ERA5, c) MME and d-s) individual CMIP6 simulation members.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/2511b2c47ad67db5a5e87f8a.png"},{"id":89488486,"identity":"c5f22a88-c4b6-449a-85b8-2322ffbcb312","added_by":"auto","created_at":"2025-08-20 13:22:54","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2322621,"visible":true,"origin":"","legend":"\u003cp\u003eMean (1985–2014) spatial distribution of the warm days (TX90p; in %), derived from a) CHIRTS, b) ERA5, c) MME and d-s) individual CMIP6 simulation members.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/5ca4e9ad542754a1bc5d5e0a.png"},{"id":89489452,"identity":"ac27dd06-7da8-4bd1-8c10-498187bbe26f","added_by":"auto","created_at":"2025-08-20 13:30:54","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":2129355,"visible":true,"origin":"","legend":"\u003cp\u003eMean (1985–2014) spatial distribution of the cool nights (TN10p; in %), derived from a) CHIRTS, b) ERA5, c) MME and d-s) individual CMIP6 simulation members.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/cb80452ad93e68b1f47689d5.png"},{"id":89488166,"identity":"b6427cf1-a731-4984-8266-a0cb54a76021","added_by":"auto","created_at":"2025-08-20 13:14:54","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":950769,"visible":true,"origin":"","legend":"\u003cp\u003eMean (1985–2014) spatial distribution of the diurnal temperature range (DTR; in °C), derived from a) CHIRTS, b) ERA5, c) MME and d-s) individual CMIP6 simulation members.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/ff6e6cc43caa5582dc5df762.png"},{"id":89488169,"identity":"aae09ed9-e231-4f6c-9a14-7fb750eb8859","added_by":"auto","created_at":"2025-08-20 13:14:54","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1105328,"visible":true,"origin":"","legend":"\u003cp\u003eMean (1985–2014) spatial distribution of the warm spell days index (WSDI; in day), derived from a) CHIRTS, b) ERA5, c) MME and d-s) individual CMIP6 simulation members.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/86053c0773a24bc9216e67f5.png"},{"id":89488180,"identity":"e48a4393-799f-40ce-8c34-37b36590618e","added_by":"auto","created_at":"2025-08-20 13:14:54","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":1151777,"visible":true,"origin":"","legend":"\u003cp\u003eMean (1985–2014) spatial distribution of the cold spell days index (CSDI; in day), derived from a) CHIRTS, b) ERA5, c) MME and d-s) individual CMIP6 simulation members.\u003c/p\u003e","description":"","filename":"image8.png","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/0fccbe486222950dc97c0d9f.png"},{"id":89488491,"identity":"80f9d962-886e-4438-a27e-a4bf03d81ba7","added_by":"auto","created_at":"2025-08-20 13:22:54","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":1063157,"visible":true,"origin":"","legend":"\u003cp\u003ePortrait diagrams showing the annual values of the percentage bias (BIAS; first column), normalized root mean square error (RMSE; second column) and pattern correlation coefficient (PCC;third column) between CHIRTS observations and all the datasets; over the analyses sub-regions SS (first row), NE (second row), EQE (third row), EQW (fourth row) and SE (fifth row).\u003c/p\u003e","description":"","filename":"image9.png","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/c8fd351f1e9fdec30d2df364.png"},{"id":89488177,"identity":"d67d60da-764c-41ba-b65c-d59905e1bbc9","added_by":"auto","created_at":"2025-08-20 13:14:54","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":179212,"visible":true,"origin":"","legend":"\u003cp\u003eTaylor skill score (TSS) displaying the annual values of a) 90th percentile of maximum temperature (T90), b) 10th percentile of minimum temperature (T10) and c) diurnal temperature range (DTR); over the analyses subregions: SS (red), NE (yellow), EQE (green), EQW (bleu) and SE (purple).\u003c/p\u003e","description":"","filename":"image10.png","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/c82f69b3dc8c6287fdc38a61.png"},{"id":89488175,"identity":"27449eef-9ee5-47ac-840a-4291e3e6e7e9","added_by":"auto","created_at":"2025-08-20 13:14:54","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":192040,"visible":true,"origin":"","legend":"\u003cp\u003eSame as for Fig. 10 but for a) warm spell days index (WSDI), b) cold spell days index (CSDI), c) warm days (TX90p) and d) cool days (TN10p).\u003c/p\u003e","description":"","filename":"image11.png","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/3660a0ba19988278b7438ece.png"},{"id":104250688,"identity":"5ace508c-637c-4e91-a204-a18c19dc9686","added_by":"auto","created_at":"2026-03-09 16:05:33","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":14782821,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/35eff732-1e41-42b4-a366-f1e627028648.pdf"},{"id":89489451,"identity":"e0ef315a-9596-4fe4-b8b8-fa8e8fbc0efa","added_by":"auto","created_at":"2025-08-20 13:30:54","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":3838700,"visible":true,"origin":"","legend":"","description":"","filename":"ESM.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7155223/v1/1f4a0ea53bee8c9b6fc68334.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Evaluation of extreme temperature events as simulated by CMIP6 models over Central Africa","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eTemperature extremes, encompassing both heat and cold, pose significant threats, including increased human mortality, stress on infrastructure, intensified wildfires, and disruptions to agriculture and transportation, all of which severely impact both natural and built environments (\u003c/span\u003eRyti et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Ebi et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Fotso-Nguemo et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eVariations in the frequency and magnitude of climatic parameters, or their corresponding indices, are generally employed to represent indicators of climate change. Prior investigations over the African domain have projected a heightened probability of extreme temperature and precipitation events occurring in conjunction with escalating anthropogenic greenhouse gas emissions (\u003c/span\u003eDosio et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Fotso-Nguemo et al. \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Ngavom et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Tamoffo et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eDespite extensive research on global extreme temperature trends and climate change (\u003c/span\u003eAbatan et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Aguilar et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Donat et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Dunn et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Fotso-Kamga et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Lewis and King \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Moron et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Ozturk et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003evan Der Walt et al. 2021), which has shown a worldwide increase in extreme heat and a decrease in extreme cold, there remains a significant gap in studies evaluating the reliability of data sources for accurately representating of mean climatological extremes, particularly in regions with sparse observational data, such as Central Africa (CA).\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eGlobal climate models (GCMs) are valuable tools for simulating contemporary and future climatic conditions and generating the requisite long-term data for analyzing potential alterations in extreme events. However, comparative analyses with observational datasets frequently reveal systematic biases within these models, thereby indicating limitations in their capacity to reproduce observed/real conditions accurately and contributing to inherent uncertainties in their projections. These uncertainties inherent in climate model projections constitute a significant concern for decision- and policy-makers (\u003c/span\u003eZwiers et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eCMIP6 are the latest generation of climate models within the Coupled Model Intercomparison Project, featuring enhanced complexity and resolution compared to earlier phases like CMIP5 (\u003c/span\u003eEyring et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThese advancements aim to provide more accurate simulations of historical climate and more robust future projections. The review by\u003c/span\u003e Zwiers et al. (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ehas emphasized the importance of understanding observed historical extremes, as this is crucial for future progress in comprehending the implications of ongoing and future extreme climate events. The importance of using the most appropriate indices when assessing the impact of sustained extreme temperature events has been presented by many authors (\u003c/span\u003ePerkins et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Seneviratne et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Zwiers et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2013\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eAssessing daily extreme temperature events in CA, using either observational or model simulations, remains significantly underdeveloped. We aim to fill this gap in the present study, by evaluating the capability of CMIP6 models to simulate a range of extreme temperature indices over CA from 1985 to 2014. By doing so, we thereby highlight both the models' strengths and limitations in capturing their spatial patterns. To some extent, this work complements the former study by\u003c/span\u003e Komelo et al. (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ewhich evaluated CMIP6 model-simulated extreme precipitation and its spatial variability in CA, fostering climate model improvement.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe remainder of the paper is structured as follows: the subsequent section (\u003c/span\u003eSection \u003cspan refid=\"Sec2\" class=\"InternalRef\"\u003e2\u003c/span\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e) describes the study area, the datasets employed and the methodological approach.\u003c/span\u003e Section \u003cspan refid=\"Sec6\" class=\"InternalRef\"\u003e3\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003epresents and discusses the findings, and\u003c/span\u003e Section \u003cspan refid=\"Sec10\" class=\"InternalRef\"\u003e4\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ewraps up and examines the findings.\u003c/span\u003e\u003c/p\u003e"},{"header":"2 Study area, data and methodology","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e2.1 Study area\u003c/span\u003e\u003c/h2\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThis study focuses on the CA domain, geographically defined between 15 \u0026deg;S and 15 \u0026deg;N latitude, and 5\u0026deg;\u0026ndash;35\u0026deg; E longitude. This region encompasses most countries in Central Africa, including Cameroon, Chad, the Central African Republic, Equatorial Guinea, Gabon, Congo, and the Democratic Republic of Congo (\u003c/span\u003eKomelo et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Mbienda et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Tamoffo et al. \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eFor quantitative analyses, we further divided the CA domain into five distinct climatic sub-regions, defined based on the K\u0026ouml;ppen-Geiger climate classification (\u003c/span\u003ePeel et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThese sub-regions are\u003c/span\u003e:\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe Sudano-Sahelian (SS) sub-region (9.5\u0026deg;\u0026ndash;14.5 \u0026deg;N, 8.0\u0026deg;\u0026ndash;23.5 \u0026deg;E), which covers southern Chad and northern Cameroon, experiencing the semi-arid Sahelian climate, characterized by high temperatures, strong solar radiation, and a unimodal rainfall-like regime.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe Northern Equatorial (NE) sub-region (2.0\u0026deg;\u0026ndash;9.5\u0026deg;N, 8.0\u0026deg;\u0026ndash;18.5 \u0026deg;E and 5.0\u0026deg;\u0026ndash;9.5\u0026deg;N, 18.5\u0026deg;\u0026ndash;32.0\u0026deg;E) covering southern Cameroon and CAR. This sub-region is characterized by a predominantly tropical wet and dry climate, featuring a long rainy season lasting approximately nine months, warm temperatures througtout the year with a notable annual range, and maximum temperatures that often precede the onset of the rainy season.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe Equatorial East (EQE) sub-region (6.0\u0026deg;S\u0026ndash;5.0\u0026deg;N, 18.5\u0026deg;\u0026ndash;32.0\u0026deg;E) covers central parts of the Democratic Republic of Congo (DRC). This sub-region is characterized by a tropical rainforest-like climate, with consistently high temperatures and humidity, relatively small seasonal temperature variations, and a bimodal rainfall-like regime.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe Equatorial West (EQW) sub-region (9.0\u0026deg;S\u0026ndash;2.0\u0026deg;N, 8.0\u0026deg;\u0026ndash;18.5\u0026deg;E and 9.0\u0026deg;\u0026ndash;6.0\u0026deg;S, 18.5\u0026deg;\u0026ndash;21.5\u0026deg;E) covers Equatorial Guinea, Gabon, Congo, southwestern DRC, and northern Angola. It features a tropical rainforest-type climate, similar to the EQE region, with consistently high temperatures and humidity and minimal seasonal variation.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eFinaly, Southern Equatorial (SE) sub-region (14.0\u0026deg;\u0026ndash;6.0\u0026deg;S, 21.5\u0026deg;\u0026ndash;32.0\u0026deg;E) covers southeastern DRC and northern Zambia. This subregion is characterized by subtropical climate, with warm temperatures year-round and a more distinct seasonal temperature cycle compared to the equatorial sub-regions.\u003c/span\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e2.2 Datasets\u003c/span\u003e\u003c/h2\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eDaily near-surface (2 m) temperature data from sixteen individual CMIP6 GCMs initiative and their multi-model ensemble (MME) (\u003c/span\u003eEyring et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eare utilized in this study. These globally available datasets span the period 1850\u0026ndash;2014. Details of the individual models are provided in\u003c/span\u003e Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\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\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eOverview of the sixteen CMIP6 climate models used in this study.\u003c/span\u003e\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=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eN\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eModel name\u003c/span\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eInstitute ID\u003c/span\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eResolution (Lon \u0026times; Lat)\u003c/span\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eReferences\u003c/span\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e1\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eBCC-ESM1\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eBeijing Climate Center (BCC) and China Meteorological Administration (CMA)\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e2.81\u0026deg; \u0026times; 2.81\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eWu et al. (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e2\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eCanESM5\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eCanadian Centre for Climate Modelling and Analysis, Victoria, Canada\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e2.81\u0026deg; \u0026times; 2.81\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSwart et al. (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e3\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eCNRM-CM6-1\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eCentre National de Recherches M\u0026eacute;t\u0026eacute;orologiques (CNRM); Centre Europ\u0026eacute;en de Recherches et de Formation Avanc\u0026eacute;e en Calcul Scientifique\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e1.41\u0026deg; \u0026times; 1.41\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eVoldoire et al. (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e4\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eE3SM-2\u0026ndash;0\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eDepartment of Energy's Energy Exascale Earth System Model, USA\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e1.00\u0026deg; \u0026times; 1.00\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eGolaz et al. (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e5\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eEC-EARTH3\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eEC-EARTH consortium\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e0.70\u0026deg; \u0026times; 0.70\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eD\u0026ouml;scher et al. (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e6\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eGFDL-ESM4\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eGeophysical Fluid Dynamics Laboratory, Princeton, USA\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e1.25\u0026deg; \u0026times; 1.00\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eHorowitz et al. (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2018\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e7\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eHadGEM3-GC31-LL\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eMet Office Hadley Centre, Exeter, UK\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e1.88\u0026deg; \u0026times; 1.25\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eWilliams et al. (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e8\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eINM-CM5-0\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eInstitute for Numerical Mathematics, Moscow, Russia\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e2.00\u0026deg; \u0026times; 1.50\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eVolodin et al. (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e9\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eIPSL-CM6A-LR\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eInstitut Pierre Simon Laplace, Paris, France\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e2.50\u0026deg; \u0026times; 1.27\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eBoucher et al. (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e10\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eMIROC6\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe University of Tokyo, Chiba, Japan\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e1.41\u0026deg; \u0026times; 1.41\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eTatebe et al. (\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e11\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eMPI-ESM1-2-HR\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eMax Planck Institute for Meteorology, Hamburg, Germany\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e0.94\u0026deg; \u0026times;0.94 \u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eM\u0026uuml;ller et al. (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2018\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e12\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eMPI-ESM1-2-LR\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eMax Planck Institute for Meteorology, Hamburg, Germany\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e1.90\u0026deg; \u0026times; 1.90\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMauritsen et al. (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e13\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eNorESM2-LM\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eNorwegian Meteorological Institute, Oslo, Norway\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e2.50\u0026deg; \u0026times; 1.89\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSeland et al. (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e14\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eNorESM2-MM\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eNorwegian Meteorological Institute, Oslo, Norway\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e1.25\u0026deg; \u0026times; 0.94\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSeland et al. (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e15\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eSAM0-UNICON\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eSeoul National University Atmosphere Model Version 0 with a Unified Convection Scheme\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e1.25\u0026deg; \u0026times; 0.94\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ePark and Shin (2019)\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e16\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eUKESM1-0-LL\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eMet Office Hadley Centre, Exeter, UK\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e1.88\u0026deg; \u0026times; 1.25\u0026deg;\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eTang et al. (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eTo evaluate the CMIP6 model simulations, the study compares their output against gridded observational temperature products, which include\u003c/span\u003e:\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe Climate Hazards Center Infrared Temperature with Stations (CHIRTS) Version 1.0 dataset (\u003c/span\u003eFunk et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ewhich serves as the primary reference for near-surface temperature, as it demonstrates good performance across Africa (\u003c/span\u003eParsons et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eCHIRTS provides daily minimum and maximum air temperature estimates on a 0.05\u0026deg; horizontal resolution, and over a quasi-global scale from 1983 to 2016. It is generated by blending satellite infrared temperatures and station data, enhanced with ERA5 for daily disaggregation.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe fifth generation of the European Centre for Medium-Range Weather Forecasts Re-Analysis (ERA5) (\u003c/span\u003eHersbach et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eis also used to account for potential uncertainties between observational and reanalysis datasets. ERA5 combines global climate models with ground, ocean, and satellite observations to produce hourly estimates of various variables globally. We use the ERA5 2-m temperature records, available as hourly time scale from 1979 to the present, at a spatial resolution of 0.25\u0026deg; x 0.25\u0026deg;.\u003c/span\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e2.3 Methods\u003c/span\u003e\u003c/h2\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eWe examined seven temperature indices defined by the ETCCDI (as detailed in\u003c/span\u003e Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e), which were calculated from daily temperature data using the Climate Data Operators (CDO;\u003c/span\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://code.mpimet.mpg.de/projects/cdo\u003c/span\u003e\u003cspan address=\"https://code.mpimet.mpg.de/projects/cdo\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThese particular temperature indices were chosen based on their relevance to socio-economic sectors such as health and agriculture, especially in Africa, as mentioned in previous research (\u003c/span\u003eAguilar et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Fotso-Kamga et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Mengistu et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Ntoumos et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Ozturk et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\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\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eSelected temperature indices for this study.\u003c/span\u003e\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=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eLabel\u003c/span\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eIndex Name\u003c/span\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eIndex Category\u003c/span\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eDescription\u003c/span\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eUnit\u003c/span\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eT90\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e90th percentile of maximum temperature\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ePercentile\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eTemperature value below which 90% of the observed daily maximum temperatures (TX) are lower for a time period\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e\u0026deg;C\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eT10\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e10th percentile of minimum temperature\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ePercentile\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eTemperature value below which 10% of the observed daily minimum temperatures (TN) are lower for a time period\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e\u0026deg;C\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eTX90p\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eWarm days\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ePercentile\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ePercentage of days for which TX\u0026thinsp;\u0026gt;\u0026thinsp;90th percentile over a given period\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e%\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eTN10p\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eCool nights\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ePercentile\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ePercentage of days for which TN\u0026thinsp;\u0026lt;\u0026thinsp;10th percentile over a given period\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e%\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eDTR\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eDiurnal\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003etemperature\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003erange\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eAbsolute\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eMean difference between daily maximum temperature and daily minimum temperature. That is TX-TN\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e\u0026deg;C\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eWSDI\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eWarm spell days index\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eDuration\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eNumber of days per period when, in intervals of at least 6 consecutive days, the daily mean temperature is above 90th percentile for the period\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003edays\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eCSDI\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eCold spell days index\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eDuration\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eNumber of days per period when, in intervals of at least 6 consecutive days, the mean daily temperature is below the 10th percentile for the period.\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003edays\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eTo assess the capacity of CMIP6 simulations to reproduce daily temperature indices across Central Africa, a common 30-year period (1985\u0026ndash;2014) was considered for both models and observations. To address the disparity in their native spatial resolutions, all datasets were interpolated to a uniform 1\u0026deg;\u0026times;1\u0026deg; grid using conservative interpolation methods prior to analyses. We further quantified the agreement between the simulated and observed daily temperature indices using the following statistical metrics\u003c/span\u003e:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe Taylor diagram (\u003c/span\u003eTaylor \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2001\u003c/span\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e), which summarizes models' skill in capturing observed patterns and magnitude by providing three statistical metrics\u003c/span\u003e:\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ei) the Root Mean Square Error (RMSE), corresponding to average magnitude of the errors between the reference data and other data sources. Defined as\u003c/span\u003e:\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:RMSE=\\sqrt{\\frac{1}{N}{\\sum\\:}_{i=1}^{N}{\\left({x}_{i}-{y}_{i}\\right)}^{2}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eii) the Pattern Correlation Coefficient (PCC), represented by the azimuthal position. It is mathematically estimated as follows\u003c/span\u003e:\u003cdiv id=\"Equ2\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ2\" name=\"EquationSource\"\u003e\n$$\\:PCC=\\frac{1}{{\\sigma\\:}_{x}{\\sigma\\:}_{y}}\\left[\\frac{1}{N}{\\sum\\:}_{i=1}^{N}\\left({x}_{i}-\\overline{x}\\right)\\left({y}_{i}-\\overline{y}\\right)\\right]$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e2\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eiii) the Standard Deviation (SD): represented by the radial distance from origin of the diagram. It is the magnitude of the spread of the data around its mean, showing the amount of variation or dispersion of a set of values. It is expressed as\u003c/span\u003e:\u003cdiv id=\"Equ3\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ3\" name=\"EquationSource\"\u003e\n$$\\:{\\sigma\\:}_{x}=\\sqrt{\\frac{1}{N}{\\sum\\:}_{i=1}^{N}{\\left({x}_{i}-\\overline{x}\\right)}^{2}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e3\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eand\u003c/span\u003e\u003cdiv id=\"Equ4\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ4\" name=\"EquationSource\"\u003e\n$$\\:{\\sigma\\:}_{y}=\\sqrt{\\frac{1}{N}{\\sum\\:}_{i=1}^{N}{\\left({y}_{i}-\\overline{y}\\right)}^{2}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e4\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe agreement between simulations and observation is quantified by their correlation and variability. Thus, A model is deemed to perform well if it satisfies the following criteria: (a) PCC\u0026thinsp;\u0026ge;\u0026thinsp;0.6, (b) SD within 1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.25, and (c) RMSE\u0026thinsp;\u0026lt;\u0026thinsp;1.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe portrait diagram analysis, which yields a compact overview of the similarities and differences between simulated and observed datasets for selected statistical parameters (including BIAS, RMSE, and PCC), thereby identifying potential areas of model agreement and bias. Small BIAS refers to less model errors. The BIAS is calculated as\u003c/span\u003e:\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003cdiv id=\"Equ5\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ5\" name=\"EquationSource\"\u003e\n$$\\:BIAS=\\frac{1}{N}{\\sum\\:}_{i=1}^{N}\\left({x}_{i}-{y}_{i}\\right)$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e5\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe Taylor Skill Score (TSS;\u003c/span\u003e Taylor \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2001\u003c/span\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e) is a numerical metric that integrates several statistical measures, including normalized SD and PCC, to assess how accurately models reproduce observed patterns and variability. PCC\u003c/span\u003e\u003csub\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e0\u003c/span\u003e\u003c/sub\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eis the maximum achievable correlation (taken as 1) for the TSS calculation. The TSS is calculated as\u003c/span\u003e:\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003cdiv id=\"Equ6\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ6\" name=\"EquationSource\"\u003e\n$$\\:TSS=\\frac{4{\\left(1+PCC\\right)}^{2}}{{\\left(\\frac{{\\sigma\\:}_{x}}{{\\sigma\\:}_{y}}+\\frac{{\\sigma\\:}_{y}}{{\\sigma\\:}_{x}}\\right)}^{2}{\\left(1+{PCC}_{0}\\right)}^{2}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e6\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eA higher TSS signifies a model's greater accuracy in capturing the spatial patterns and magnitude of observed extreme temperature events.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eIn these equations, x\u003c/span\u003e\u003csub\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ei\u003c/span\u003e\u003c/sub\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ecorresponds to the model's value, while y\u003c/span\u003e\u003csub\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ei\u003c/span\u003e\u003c/sub\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003erepresents the observed reference value to the ith grid point. N indicates the number of data points. The terms\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\overline{x}\\)\u003c/span\u003e\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eand\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\overline{y}\\)\u003c/span\u003e\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eare the respective means of the model and reference values.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eTo evaluate the ability of CMIP6 models to represent the spatial patterns of extreme temperature events over Central Africa, the spatial distribution of each of the seven relevant indices was first examined for both the models and the observational datasets. Subsequently, a Taylor diagram analysis was performed to provide a global statistical overview of model performance across the entire Central African domain. To gain deeper insights into the model performance at a local level, the analysis was further conducted for each of the five identified sub-regions using portrait diagrams and the Taylor Skill Score.\u003c/span\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"3 Results and Discussion","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e3.1 Spatial variability of percentile-based indices\u003c/span\u003e\u003c/h2\u003e\u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003edisplays the spatial distribution of the 90th percentile of maximum temperature (T90) during the 1985\u0026ndash;2014 period over CA, derived from the observations CHIRTS and ERA5, as well as the individual CMIP6 simulations and their MME. The T90 represents a threshold for what is considered a relatively warm temperature for a location during a specific period. The distribution of the T90 derived from the CHIRTS dataset (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e) shows values ranging from 24 to 46\u0026deg;C, with the highest records (\u0026ge;\u0026thinsp;40\u0026deg;C) located north of 9\u0026deg;N. The T90 presented by ERA5 (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e) has similar patterns to those of CHIRTS, though less intense by about 2\u0026deg;C in most regions of the domain (see supplementary material, Fig. S2a). Compared with CHIRTS, ERA5 showed good agreement in the representation of the T90, with RMSE/SD/PCC of about 0.30/0.90/0.96 (see supplementary material, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e; T90). Compared to CHIRTS, the MME and individual CMIP6 models generally capture the spatial pattern of the T90 (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec-s\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e). However, regional differences exist. Compared to CHIRTS, CMIP6 models typically show RMSE values ranging from 0 to 0.5, SD from 0.75 to 1.25, and PCC from 0.85 to 0.96 (see Supplementary, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e; T90). MIROC6 exhibits slightly higher variability than other models. Across the study area, biases between the MME as well as the individual CMIP6 models and CHIRTS are mostly between 0 and \u0026plusmn;\u0026thinsp;4\u0026deg;C (see Supplementary, Fig. S2b-r). Furthermore, outside the equatorial regions (EQE, EQW, and NE), MIROC6 overestimates the T90, with biases ranging from 4 to 10\u0026deg;C (see Supplementary, Fig. S2l).\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eshows the spatial distribution of the 10th percentile of minimum temperature (T10) over the CA domain for 1985\u0026ndash;2014, derived from CHIRTS, ERA5, individual CMIP6 simulations, and the MME. The T10 represents a threshold for relatively cold temperatures at a given location over a period. The T10 as seen by CHIRTS (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e) varies from 8\u0026deg;C to 24\u0026deg;C, with Zambia and northern Angola experiencing the warmest of these, while southern Sudan and the Central African Republic see the coolest. ERA5 (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e) depicts a similar pattern, though it exhibits less intense temperatures. Overall, ERA5 and CHIRTS match well, with strong agreement in RMSE/SD/PCC of 0.40/1.10/0.93 (see supplementary material, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e; T10). CMIP6 models and their multi-model ensemble (MME) accurately simulate the spatial pattern of the T10 when compared to CHIRTS and ERA5 observations (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec-s\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e). Despite this, consistent underestimation of 1 to 6\u0026deg;C are observed across the study area when using CHIRTS as a reference (see Supplementary, Fig. S3b-r). These biases are evident in ERA5, the MME, and individual CMIP6 models. Among the models, MPI-ESM1-2-HR and MPI-ESM1-2-LR show the best performance with minimal biases, while INM-CM5-0 displays the largest. However, CMIP6 models confirm their good spatial agreement with CHIRTS, providing RMSE from 0 to 1, SD from 0.75 to 1.40, and PCC from 0.60 to 0.90 (see Supplementary, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e; T10).\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFigures\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003edisplay the spatial distribution of the annual percentage of warm days (TX90p) and cool nights (TN10p), respectively, during the 1985\u0026ndash;2014 period, obtained from the observations CHIRTS and ERA5, as well as the individual CMIP6 simulations and the MME. Specifically, TX90p is the percentage of days with daily maximum temperatures above the 90th percentile, and TN10p is the percentage of days with daily minimum temperatures below the 10th percentile. The spatial patterns of TX90p/TN10p shown by CHIRTS (\u003c/span\u003eFigs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e) are highly heterogeneous, though the magnitudes vary less (9.8\u0026ndash;10.2%) over the study area. In terms of magnitude, ERA5, MME, and individual models reproduce similar patterns to CHIRTS, providing low biases over the domain. However, it is worth noting that poor correlation values (below 0.2) are noted among all the datasets (see Supplementary, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e; TX90p and TN10p). CHIRTS demonstrates spatially heterogeneous distributions of TX90p/TN10p (\u003c/span\u003eFigs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e), with magnitudes varying between 9.8% and 10.2%. ERA5, MME, and individual models reproduce similar spatial patterns to CHIRTS, with low biases of approximately\u0026thinsp;\u0026plusmn;\u0026thinsp;0.2% (Supplementary Figures S4 and S5). Despite this, correlation coefficients across all datasets are below 0.2 (Supplementary Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). The low correlations observed in TX90p/TN10p between different data sources are not necessarily a sign of failure to mimic observations, but, rather, they suggest a high degree of spatio-temporal variability in these indices. As mentioned by\u003c/span\u003e Zwiers et al. (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eclimate research often focuses on indicators of event frequency and intensity, such as \"moderate extremes\" defined by percentile thresholds (e.g., 90th/10th), which, while within seasonal observation ranges, are valuable for trend analysis, a point relevant to the observed variability.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e3.2 Spatial variability of absolute and duration-based indices\u003c/span\u003e\u003c/h2\u003e\u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003epresents the spatial distribution of the diurnal temperature range (DTR) across the CA domain for 1985\u0026ndash;2014, using data from CHIRTS, ERA5, individual CMIP6 models, and their MME. The DTR reflects how much the temperature changes from the hottest part of the day to the coldest part of the night. According to CHIRTS (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e), the DTR starts at about 5\u0026deg;C in coastal areas and gradually increases to around 9\u0026deg;C in regions closer to the Equator. Moving away from the Equator, the DTR shows a noticeable increase, peaking at about 15\u0026deg;C in the northern and southern extremities of the CA domain. ERA5 (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e) shows a highly consistent DTR pattern with CHIRTS, demonstrating strong agreement with RMSE of 0.12, SD of 1.05, and PCC of 0.99 (see supplementary material, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e; DTR).\u003c/span\u003e Figures\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec-s \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eshow that the MME and the individual CMIP6 models successfully simulate the spatial pattern of the DTR observed in CHIRTS and ERA5. While biases across individual models range from 1 to 6\u0026deg;C, the MME, along with BCC-ESM1, UKESM1-LL, EC-Earth3, E3SM-2-0, and SAM0-UNICON, exhibit lower bias, generally\u0026thinsp;\u0026le;\u0026thinsp;2\u0026deg;C (see Supplementary, Fig. S6). Indeed, these models display better alignment with CHIRTS, reflected in RMSE/SD/PCC values between 0.25 and 0.75 / 0.75 and 1.25 / 0.70 and 0.95 (see Supplementary, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e; DTR). Preceding\u003c/span\u003e Donat et al. (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003estudies indicated that DTR across northern and western areas of Africa is linked to both the El Ni\u0026ntilde;o-Southern Oscillation (ENSO) and the North Atlantic Oscillation (NAO), with the latter showing a stronger association, particularly evident in the western Arab region bordering the Atlantic. This observation suggests the possibility of analogous correlations between DTR and these climate oscillations in CA, which warrant investigation using CMIP6 model simulations.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFigures\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e and \u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003edisplay the spatial distributions of the annual Warm Spell Days Index (WSDI) and Cold Spell Days Index (CSDI) across CA from 1985 to 2014, using CHIRTS and ERA5 observations with CMIP6 model simulations and their MME. WSDI indicates the length of prolonged warm spells, and CSDI that of cold spells. For the WSDI (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e), ERA5 and CHIRTS display consistent spatial patterns, with values increasing from less than 10 days in equatorial regions (NE, EQW, EQE) to approximately 25 days in the northern and southern areas (SS and SE). Strong agreement between ERA5 and CHIRTS is supported by RMSE, SD, and PCC values within the ranges of 0.25\u0026ndash;0.75, 0.25\u0026ndash;0.50, and 0.85\u0026ndash;0.95, respectively (see Supplementary, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e; WSDI and CSDI). The MME and most CMIP6 models reproduce the overall WSDI patterns but tend to overestimate warm spell durations, especially in the Democratic Republic of Congo (DRC), where biases reach 20 days (see Supplementary, Fig. S7). However, NorESM2-MM and SAM0-UNICON outperform the other models, with RMSE/SD/PCC closer to that of ERA5. Regarding the Cold Spell Days Index (CSDI) in\u003c/span\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e, \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eCMIP6 models and the MME tend to overestimate compared to CHIRTS. This overestimation aligns with the simulated WSDI patterns, though the PCC is notably weaker, below 0.6 across most models (see Supplementary, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e; CSDI). The observed WSDI/CSDI patterns in the region may be related to the El Ni\u0026ntilde;o\u0026ndash;Southern Oscillation (ENSO), potentially following warm/cold ENSO events, as demonstrated by\u003c/span\u003e Moron et al. (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ein the tropical North of Africa.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e3.3 Statistical Characterization of Extreme Temperature Indices within Climatic Sub-regions\u003c/span\u003e\u003c/h2\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eWe examined the ability of CMIP6 models in reproducing observed extreme temperature indices across the diverse climatic sub-regions of CA. This involved a detailed statistical evaluation using bias, RMSE, PCC, and TSS, providing insights into model accuracy at a localized scale.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eUsing CHIRTS observations as the reference data\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003edisplays portrait diagrams of seven extreme temperature indices calculated for the five sub-regions.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eAs shown in the first column of\u003c/span\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e, \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ethe majority of biases are relatively minor, staying within the range of -20\u0026ndash;20% across all the sub-regions. Specifically, the T90 and T10 indices tend to show negative biases across these sub-regions, whereas the WSDI and CSDI indices more often exhibit positive biases. Notably, WSDI and CSDI exhibit the most significant biases, with some CMIP6 models and their MMEs overestimating by more than 60% in the EQW, EQE, and SE sub-regions. It's worth noting that the MME typically reduces bias compared to individual models for the most temperature indices and across all sub-regions. The notable exception is the WSDI index in the EQW, EQE, and SE sub-regions, where substantial positive biases persist even in the MME. Regarding the RMSE (see\u003c/span\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e, \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003esecond column), CMIP6 models and their MMEs generally show good agreement with observations (RMSE\u0026thinsp;\u0026lt;\u0026thinsp;1) across all five sub-regions for most temperature indices, except for CSDI and WSDI. Consistent with bias findings, simulations tend to provide higher RMSE values (greater than 1) for both CSDI and WSDI indices across the EQE, EQW, and SE sub-regions. The most significant errors are displayed for WSDI (RMSE\u0026thinsp;\u0026gt;\u0026thinsp;2 in EQE and EQW). However, for WSDI, it's worth noting that BCC-ESM1, NorESM2-MM, and NorESM2-LM models perform better, achieving RMSE values comparable to the ERA5 dataset. For the PCC (see\u003c/span\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e, \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ethird column), most CMIP6 models show good correlation (PCC\u0026thinsp;\u0026gt;\u0026thinsp;0.6) with CHIRTS observations across several sub-regions for specific indices: T10, T90, and DTR in SS and NE; T10, T90, and WSDI in EQE; T10 and T90 in EQW; and T10, WSDI, and CSDI in NE. Conversely, all CMIP6 models exhibit lower correlation (PCC\u0026thinsp;\u0026lt;\u0026thinsp;0.6) for TN10p and TX90p. Similarly, ERA5 reanalysis also shows low correlation for these two indices across different areas, indicating potential discrepancies between observation sources (\u003c/span\u003eKomelo et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eHowever, consistent with bias and RMSE results, the MME generally demonstrates a better PCC for all indices compared to individual models.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eOverall, the representation of the majority of extreme temperature indices by the MME agrees with CHIRTS observations over the different climatic sub-regions, except for TN10p and TX90p where low PCC values are recorded. Moreover, when compared to the observations, errors or differences in both mean temperature and extreme temperature indices are generally a problem common to climate models (\u003c/span\u003eZebaze et al. \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Mengistu et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eand our results here are consistent with those studies over the African domains (\u003c/span\u003eNtoumos et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Fotso-Kamga et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe study by\u003c/span\u003e Mengistu et al. (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eon extreme temperature events in major South African cities using CMIP6 models has suggested adjusted datasets that show fairly good results for different statistical metrics. In addition, our findings show that the accurate representation of the MME is often the result of the overestimation from some models being balanced out by the underestimation from others in the ensemble mean, rather than due to optimal parameterization. As suggested by\u003c/span\u003e Sonkou\u0026eacute; et al. (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), a \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003epreliminary selection of good models before making an MME is crucial to provide consistent and accurate results.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eOn the other hand\u003c/span\u003e, Figs.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003e and \u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003edisplay the TSS metric results for the seven extreme temperature indices across the five climatic subregions, with CHIRTS observations used as reference data.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eFor the T90 (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003ea\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e), a large number of individual models demonstrate satisfactory performance (TSS\u0026thinsp;\u0026ge;\u0026thinsp;0.6) over SS and NE sub-regions. While individual model competence is somewhat lower in EQE and EQW, SE records the fewest satisfactory individual models (EC-Earth3, GFDL-ESM4, and NorESM2-MM). The MME records satisfactory performance over SS, NE, EQE, and EQW. For the T10 (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eb\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e), the majority of the individual models achieve satisfactory TSS scores (\u0026ge;\u0026thinsp;0.6) across the five sub-regions. Some simulations show strong individual model performance (CanESM5, CNRM-CM6-1, HadGEM3-GC31-LL, and IPSL-CM6A-LR) over all the sub-regions. For this index, the MME records satisfactory performance over all five sub-regions. For the DTR (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003ec\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e), the model performance is more varied. A moderate number of individual models are satisfactory in SS and EQW. NE has a relatively high number of well-performing individual models (with TSS\u0026thinsp;\u0026ge;\u0026thinsp;0.6). In contrast, few individual models are satisfactory in EQE (EC-Earth3, HadGEM3-GC31-LL, and UKESM1-0-L) and SE (E3SM-2-0, EC-Earth3, MIROC6, and NorESM2-MM). Here, the MME shows satisfactory performance over SS, NE, and EQW. The WSDI (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003ea\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e) is generally challenging for most individual models. Satisfactory performance (TSS\u0026thinsp;\u0026ge;\u0026thinsp;0.6) is noted for NorESM2-LM and CanESM5 in some areas, but almost all models perform well only over SE. Few models are satisfactory in other sub-regions, with SS having the fewest (NorESM2-LM and NorESM2-MM). The MME performs satisfactorily for WSDI over the NE, EQW, and SE. Regarding the CSDI (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003eb\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e), most individual models do not accurately capture the observations over sub-regions. While a few individual models are satisfactory with TSS\u0026thinsp;\u0026ge;\u0026thinsp;0.6 in SS (BCC-ESM1 and UKESM1-0-L), NE (BCC-ESM1, EC-Earth3, HadGEM3-GC31-LL, and MIROC6), EQE (E3SM-2-0, EC-Earth3, GFDL-ESM4, MIROC6, and SAM0-UNICON), and EQW (HadGEM3-GC31-LL), SE has no satisfactory model. The MME shows satisfactory performance over NE and EQW. For TX90p and TN10p (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003ec, d\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e), both the individual CMIP6 models and the MME consistently yield low TSS values (\u0026lt;\u0026thinsp;0.4) across all five sub-regions. This poor performance is in agreement with the PCC results shown in\u003c/span\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eOverall, the number of models achieving satisfactory performance (TSS\u0026thinsp;\u0026ge;\u0026thinsp;0.6) varies significantly depending on the extreme temperature index and the sub-region. This suggests that different models have varying strengths and weaknesses in simulating different aspects of extreme temperature variability across different climatic zones. However, the MME generally shows better performance compared to individual CMIP6 models. Across the five sub-regions, both individual CMIP6 models and the MME tend to have a larger number of well-performing models over the SS and NE sub-regions compared to EQE, EQW, and SE. This could reflect differences in the complexity of the climate variability and processes in these regions. In fact, a unimodal precipitation distribution characterizes the SS, NE, and SE sub-regions of CA, while a bimodal distribution is observed in the EQE and EQW sub-regions, according to the findings of Fotso-Kamga et al. (2019). This could indicate the importance of seasons in the occurrence of extreme temperature events. For instance, throughout seasons, the occurrence of both warm and cold temperature extremes is influenced by factors determining the overall seasonal climate, such as sunlight, elevation, and proximity to the ocean, as well as natural climate processes like ENSO (\u003c/span\u003eMiralles et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Lewis and King \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2015\u003c/span\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e). The complex interplay of these factors across different timescales (\u003c/span\u003eSillmann et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003econtributes to the challenge faced by climate models in accurately reproducing the variability of extreme temperature indices. Notably, the present study reveals that CMIP6 models do not consistently achieve a satisfactory TSS (\u0026ge;\u0026thinsp;0.6) for the TX90p and TN10p indices across all five sub-regions. This strongly suggests a significant challenge for these models in accurately representing the frequency of extreme warm and cold temperature events over the CA domain, potentially originating from issues with the simulation of the temperature distribution's tails, often registered in moderate extremes (\u003c/span\u003eZwiers et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eBesides, even ERA5 reanalysis shows low consistency with the reference CHIRTS for both TX90p and TN10p indices over the five sub-regions, raising the problem of discrepancies among different sources of observations (\u003c/span\u003eCamberlin et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ewhich often contributes to varying performance of models depending on the considered chosen dataset as reference (Fotso-Kamga et al. 2019;\u003c/span\u003e Taguela et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e"},{"header":"4 Summary and Conclusion","content":"\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThis study evaluated the representation of extreme temperature events as simulated by CMIP6 models over Central Africa (CA), with a focus on their spatial variability across five climatic sub-regions. We analyzed several percentile-based, absolute, and duration-based extreme temperature indices, comparing the simulations from individual CMIP6 models and their Multi-Model Ensemble (MME) against CHIRTS observations, with consideration of ERA5 reanalysis as well.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eOur analysis reveals that CMIP6 models generally show agreement with CHIRTS observations for the majority of extreme temperature indices across the CA sub-regions. Furthermore, the MME often demonstrates better agreement than individual models, with its accuracy frequently arising from the compensation of overestimations and underestimations across the ensemble. However, CMIP6 models, including the MME, do not consistently capture CHIRTS observations for the TX90p and TN10p indices across all sub-regions.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe number of models achieving satisfactory performance varied considerably depending on the specific extreme temperature index and the sub-region under consideration. Notably, both individual models and the MME tended to perform better over the Sudano-Sahelian (SS) and Northern Equatorial (NE) sub-regions compared to the Equatorial East (EQE), Equatorial West (EQW), and Southern Equatorial (SE) sub-regions. This disparity in performance may be linked to the complexity of climate variability and processes in these regions, such as the unimodal precipitation distribution in SS and NE versus the bimodal distribution in EQE and EQW, as suggested by previous research.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eIn conclusion, while the CMIP6 models show reasonable ability to represent many extreme temperature indices over Central Africa, challenges remain, particularly for the frequency of the percentile-based indices like TX90p and TN10p, and in specific sub-regions. These findings underscore the need for continued evaluation and potential improvements in climate models to enhance their accuracy in simulating the full spectrum of extreme temperature events across this diverse region. Improving the simulation of these extremes is crucial for better understanding potential climate change impacts and informing regional adaptation strategies.\u003c/span\u003e\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eConflicts of Interest\u003c/span\u003e\u003c/h2\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe authors confirm that there are no conflicts of interest associated with this work.\u003c/span\u003e\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eG.F-K., T.C.F-N and A.D. conceptualized the study. G.F-K, T.C.F-N., A.D., Z.D.Y. and D.A.V. defined the methodology. G.F-K, T.C.F-N., S.Z., Z.N. and A.T.T. conducted the analysis. G.F-K. and T.C.F-N. wrote the main manuscript text. All authors reviewed the manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe gratefully thank the climate modeling groups detailed in Table 1 for their essential work in producing and making their model output openly accessible through the Earth System Grid Federation (ESGF) platforms. Furthermore, we would like to express our gratitude to the LMI-NEXUS initiative for their support during the realization of this work.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAbatan AA, Abiodun BJ, Lawal KA, Gutowski WJ Jr (2016) Trends in extreme temperature over Nigeria from percentile-based threshold indices. Int J Climatol 36:2527\u0026ndash;2540. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/joc.4539\u003c/span\u003e\u003cspan address=\"10.1002/joc.4539\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAguilar E, Aziz Barry A, Brunet M, Ekang L, Fernandes A, Massoukina M, Zhang X (2009) Changes in temperature and precipitation extremes in western central Africa, Guinea Conakry, and Zimbabwe, 1955\u0026ndash;2006. J Geophys Res Atmos 114:D2. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2008JD011075\u003c/span\u003e\u003cspan address=\"10.1029/2008JD011075\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAllen M, Dube O, Solecki W, Arag\u0026oacute;n-Durand F, Cramer W, Humphreys S, Kainuma M, Kala J, Mahowald N, Mulugetta Y et al (2018) Chap. 1: Framing and Context. In: Masson-Delmotte V, Zhai P, P\u0026ouml;rtner H-O, Roberts D, Skea J, Shukla PR, Pirani A, Moufouma-Okia W (eds) IPCC Special Report: Global Warming of 1.5\u0026deg;C. Cambridge University Press\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoucher O, Servonnat J, Albright AL, Aumont O, Balkanski Y, Bastrikov V, Vuichard N (2020) Presentation and evaluation of the IPSL-CM6A‐LR climate model. J Adv Model Earth Syst 12:e2019MS002010. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2019MS002010\u003c/span\u003e\u003cspan address=\"10.1029/2019MS002010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCamberlin P, Barraud G, Bigot S, Dewitte O, Makanzu Imwangana F, Maki Mateso JC, Samba G (2019) Evaluation of remotely sensed rainfall products over Central Africa. Q J R Meteorol Soc 145:2115\u0026ndash;2138. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/qj.3540\u003c/span\u003e\u003cspan address=\"10.1002/qj.3540\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDonat MG, Alexander LV, Yang H, Durre I, Vose R, Dunn RJ, Kitching S (2013) Updated analyses of temperature and precipitation extreme indices since the beginning of the twentieth century: The HadEX2 dataset. J Geophys Res Atmos 118:2098\u0026ndash;2118. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/jgrd.50150\u003c/span\u003e\u003cspan address=\"10.1002/jgrd.50150\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDonat MG, Peterson TC, Brunet M, King AD, Almazroui M, Kolli RK, Rhaz E, K (2014) Changes in extreme temperature and precipitation in the Arab region: long-term trends and variability related to ENSO and NAO. Int J Climatol 34. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/joc.3703\u003c/span\u003e\u003cspan address=\"10.1002/joc.3703\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDosio A, Jury MW, Almazroui M, Ashfaq M, Diallo I, Engelbrecht FA, Tamoffo AT (2021) Projected future daily characteristics of African precipitation based on global (CMIP5, CMIP6) and regional (CORDEX, CORDEX-CORE) climate models. Clim Dyn 57:3135\u0026ndash;3158. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00382-021-05741-z\u003c/span\u003e\u003cspan address=\"10.1007/s00382-021-05741-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eD\u0026ouml;scher R, Acosta M, Alessandri A, Anthoni P, Arneth A, Arsouze T, Zhang Q (2021) The EC-earth3 Earth system model for the climate model intercomparison project 6. Geosci Model Dev Discuss 2021:1\u0026ndash;90. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/gmd-2021-26\u003c/span\u003e\u003cspan address=\"10.5194/gmd-2021-26\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDunn RJ, Alexander LV, Donat MG, Zhang X, Bador M, Herold N, Yussof BH, M. N. A (2020) Development of an updated global land in situ-based data set of temperature and precipitation extremes: HadEX3. J Geophys Res Atmos 125. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2019JD032263\u003c/span\u003e\u003cspan address=\"10.1029/2019JD032263\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. e2019JD032263\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEbi KL, Capon A, Berry P, Broderick C, de Dear R, Havenith G, Honda Y, Kovats RS, Ma W, Malik A et al (2021) Hot weather and heat extremes: Health risks. Lancet 398:698\u0026ndash;708. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0140-6736(21)01053-0\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(21)01053-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEyring V, Bony S, Meehl GA, Senior CA, Stevens B, Stouffer RJ, Taylor KE (2016) Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. Geosci Model Dev 9:1937\u0026ndash;1958. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/gmd-9-1937-2016\u003c/span\u003e\u003cspan address=\"10.5194/gmd-9-1937-2016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFotso-Kamga G, Fotso‐Nguemo TC, Diallo I, Yepdo ZD, Pokam WM, Vondou DA, Lenouo A (2020) An evaluation of COSMO‐CLM regional climate model in simulating precipitation over Central Africa. Int J Climatol 40:2891\u0026ndash;2912. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/joc.6366\u003c/span\u003e\u003cspan address=\"10.1002/joc.6366\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFotso-Kamga G, Fotso‐Nguemo TC, Diallo I, Nyanchi GT, Yepdo ZD, Chouto S, Lenouo A (2023) Projected changes in extreme precipitation and temperature events over Central Africa from COSMO‐CLM simulations under the global warming level of 1.5\u0026deg; C and above. Int J Climatol 43:6330\u0026ndash;6351. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/joc.8183\u003c/span\u003e\u003cspan address=\"10.1002/joc.8183\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFotso-Nguemo TC, Vondou DA, Diallo I, Diedhiou A, Weber T, Tanessong RS et al (2022) Potential impact of 1.5, 2 and 3 C global warming levels on heat and discomfort indices changes over Central Africa. Sci Total Environ 804:1\u0026ndash;11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.scitotenv.2021.150117\u003c/span\u003e\u003cspan address=\"10.1016/j.scitotenv.2021.150117\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFotso-Nguemo TC, Weber T, Diedhiou A, Chouto S, Vondou DA, Rechid D, Jacob D (2023) Projected impact of increased global warming on heat stress and exposed population over Africa. Earths Future 11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2022EF003268\u003c/span\u003e\u003cspan address=\"10.1029/2022EF003268\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. e2022EF003268\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFunk C, Peterson P, Peterson S, Shukla S, Davenport F, Michaelsen J, Mata N (2019) A high-resolution 1983\u0026ndash;2016 T max climate data record based on infrared temperatures and stations by the Climate Hazard Center. J Clim 32:5639\u0026ndash;5658. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1175/JCLI-D-18-0402.1\u003c/span\u003e\u003cspan address=\"10.1175/JCLI-D-18-0402.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGolaz JC, Van Roekel LP, Zheng X, Roberts AF, Wolfe JD, Lin W, Bader DC (2022) The DOE E3SM model version 2: Overview of the physical model and initial model evaluation. J Adv Model Earth Syst 14. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2022MS003156\u003c/span\u003e\u003cspan address=\"10.1029/2022MS003156\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. e2022MS003156\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHersbach H, Bell B, Berrisford P, Hirahara S, Hor\u0026aacute;nyi A, Mu\u0026ntilde;oz-Sabater J, Th\u0026eacute;paut JN (2020) The ERA5 global reanalysis. Q J R Meteorol Soc 146:1999\u0026ndash;2049. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/qj.3803\u003c/span\u003e\u003cspan address=\"10.1002/qj.3803\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHorowitz LW, John JG, Blanton C, McHugh C, Radhakrishnan A, Rand K, Zeng Y (2018) NOAA-GFDL GFDL-ESM4 model output prepared for CMIP6 DAMIP hist-aer. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.22033/ESGF/CMIP6.13327\u003c/span\u003e\u003cspan address=\"10.22033/ESGF/CMIP6.13327\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. World Data Center for Climate (WDCC) at DKRZ\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKomelo CK, Fotso-Nguemo TC, Ngavom Z, Dessacka AK, Taguela TN, Yepdo ZD, Tchawoua C (2024) Evaluation of extreme precipitation events as simulated by CMIP6 models over Central Africa: Spatial patterns. Theor Appl Climatol 155:9579\u0026ndash;9599. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00704-024-04987-9\u003c/span\u003e\u003cspan address=\"10.1007/s00704-024-04987-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLewis SC, King AD (2015) Dramatically increased rate of observed hot record breaking in recent Australian temperatures. Geophys Res Lett 42:7776\u0026ndash;7784. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/2015GL065555\u003c/span\u003e\u003cspan address=\"10.1002/2015GL065555\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLu C, Sun Y, Zhang X (2018) Multimodel detection and attribution of changes in warm and cold spell durations. Environ Res Lett 13:074013. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1088/1748-9326/aac539\u003c/span\u003e\u003cspan address=\"10.1088/1748-9326/aac539\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMauritsen T, Bader J, Becker T, Behrens J, Bittner M, Brokopf R, Roeckner E (2019) Developments in the MPI-M Earth System Model version 1.2 (MPI‐ESM1. 2) and its response to increasing CO2. J Adv Model Earth Syst 11:998\u0026ndash;1038. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2018MS001400\u003c/span\u003e\u003cspan address=\"10.1029/2018MS001400\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMbienda AJK, Guenang GM, Kaissassou S, Sandjon AT, Tanessong RS, Lontio SLM, Djomou ZY (2022) Boundary layer schemes in the regional climate model RegCM4. 6 over Central Africa. Clim Dyn 58:691\u0026ndash;709. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00382-021-05971-4\u003c/span\u003e\u003cspan address=\"10.1007/s00382-021-05971-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMengistu AG, Woyessa YE, Tesfuhuney WA, Steyn AS, Lee SS (2024) Assessing the impact of climate change on future extreme temperature events in major South African cities. Theor Appl Climatol 155:1807\u0026ndash;1819. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00704-023-04770-0\u003c/span\u003e\u003cspan address=\"10.1007/s00704-023-04770-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMiralles DG, Teuling AJ, Van Heerwaarden CC (2014) \u0026amp; Vil\u0026agrave;-Guerau de Arellano, J. Mega-heatwave temperatures due to combined soil desiccation and atmospheric heat accumulation. Nature Geosci 7:345\u0026ndash;349. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/ngeo2147\u003c/span\u003e\u003cspan address=\"10.1038/ngeo2147\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMoron V, Oueslati B, Pohl B, Rome S, Janicot S (2016) Trends of mean temperatures and warm extremes in northern tropical Africa (1961\u0026ndash;2014) from observed and PPCA-reconstructed time series. J Geophys Res Atmos 121:5298\u0026ndash;5319. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/2015JD024103\u003c/span\u003e\u003cspan address=\"10.1002/2015JD024103\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eM\u0026uuml;ller WA, Jungclaus JH, Mauritsen T, Baehr J, Bittner M, Budich R, Marotzke J (2018) A higher-resolution version of the max planck institute earth system model (MPI‐ESM1. 2‐HR). J Adv Model Earth Syst 10:1383\u0026ndash;1413. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2017MS001211\u003c/span\u003e\u003cspan address=\"10.1029/2017MS001211\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNgavom Z, Fotso-Nguemo TC, Vondou DA, Fotso-Kamga G, Zebaze S, Yepdo ZD, Diedhiou A (2024) Projected changes in population exposure to extreme precipitation events over Central Africa under the global warming levels of 1.5\u0026deg; C and 2\u0026deg; C: insights from CMIP6 simulations. Model Earth Syst Environ 10:5753\u0026ndash;5769. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s40808-024-01968-3\u003c/span\u003e\u003cspan address=\"10.1007/s40808-024-01968-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNtoumos A, Hadjinicolaou P, Zittis G, Lelieveld J (2020) Updated assessment of temperature extremes over the Middle East\u0026ndash;North Africa (MENA) region from observational and CMIP5 data. Atmosphere 11:813. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/atmos11080813\u003c/span\u003e\u003cspan address=\"10.3390/atmos11080813\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOzturk T, Saygili-Araci FS, Kurnaz ML (2021) Projected changes in extreme temperature and precipitation indices over CORDEX-MENA domain. Atmosphere 12:622. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/atmos12050622\u003c/span\u003e\u003cspan address=\"10.3390/atmos12050622\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePark S, Shin J, Kim S, Oh E, Kim Y (2019) Global climate simulated by the Seoul National University atmosphere model version 0 with a unified convection scheme (SAM0-UNICON). J Clim 32:2917\u0026ndash;2949. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1175/JCLI-D-18-0157.1\u003c/span\u003e\u003cspan address=\"10.1175/JCLI-D-18-0157.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eParsons D, Stern D, Ndanguza D, Sylla MB (2022) Evaluation of satellite-based air temperature estimates at eight diverse sites in Africa. Climate 10:98. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/cli10070098\u003c/span\u003e\u003cspan address=\"10.3390/cli10070098\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePeel MC, Finlayson BL, McMahon TA (2007) Updated world map of the K\u0026ouml;ppen-Geiger climate classification. Hydrol Earth Syst Sci 11:1633\u0026ndash;1644. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/hess-11-1633-2007\u003c/span\u003e\u003cspan address=\"10.5194/hess-11-1633-2007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePerkins SE, Alexander LV, Nairn JR (2012) Increasing frequency, intensity and duration of observed global heatwaves and warm spells. Geophys Res Lett 39. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2012GL053624\u003c/span\u003e\u003cspan address=\"10.1029/2012GL053624\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRyti NR, Guo Y, Jaakkola JJ (2016) Global association of cold spells and adverse health effects: a systematic review and meta-analysis. Environ Health Perspect 124:12\u0026ndash;22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1289/ehp.1408794\u003c/span\u003e\u003cspan address=\"10.1289/ehp.1408794\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSeland \u0026Oslash;, Bentsen M, Olivi\u0026eacute; D, Toniazzo T, Gjermundsen A, Graff LS, Schulz M (2020) Overview of the Norwegian Earth System Model (NorESM2) and key climate response of CMIP6 DECK, historical, and scenario simulations. Geosci Model Dev 13:6165\u0026ndash;6200. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/gmd-13-6165-2020\u003c/span\u003e\u003cspan address=\"10.5194/gmd-13-6165-2020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSeneviratne SI, Nicholls N, Easterling D, Goodess CM, Kanae S, Kossin J, Luo Y, Marengo J, McInnes K, Rahimi M, Reichstein M, Sorteberg A, Vera C, Zhang X (2012) Changes in climate extremes and their impacts on the natural physical environment. In: Field C. B., Barros V., Stocker T. F., Qin D., Dokken D., Ebi K. L., Mastrandrea M. D., Mach K. J., Plattner G.-K., Allen S. K., Tignor M., Midgley P. M. (Eds.), Intergovernmental Panel on Climate Change special report on managing the risks of extreme events and disasters to advance climate change adaptation, pp. 109\u0026ndash;230. Cambridge University Press\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSillmann J, Thorarinsdottir T, Keenlyside N, Schaller N, Alexander LV, Hegerl G, Zwiers FW (2017) Understanding, modeling and predicting weather and climate extremes: Challenges and opportunities. Weather Clim Extremes 18:65\u0026ndash;74. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.wace.2017.03.003\u003c/span\u003e\u003cspan address=\"10.1016/j.wace.2017.03.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSonkou\u0026eacute; D, Monkam D, Fotso-Nguemo TC, Yepdo ZD, Vondou DA (2019) Evaluation and projected changes in daily rainfall characteristics over Central Africa based on a multi-model ensemble mean of CMIP5 simulations. Theor Appl Climatol 137:2167\u0026ndash;2186. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00704-018-2708-8\u003c/span\u003e\u003cspan address=\"10.1007/s00704-018-2708-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSwart NC, Cole JN, Kharin VV, Lazare M, Scinocca JF, Gillett NP, Sigmond M (2019) CCCma CanESM5 model output prepared for CMIP6 FAFMIP. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.22033/ESGF/CMIP6.13670\u003c/span\u003e\u003cspan address=\"10.22033/ESGF/CMIP6.13670\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. World Data Center for Climate (WDCC) at DKRZ\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTaguela TN, Vondou DA, Moufouma-Okia W, Fotso‐Nguemo TC, Pokam WM, Tanessong RS, Tchotchou D, L. A (2020) CORDEX multi‐RCM hindcast over Central Africa: Evaluation within observational uncertainty. J Geophys Res Atmos 125. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2019JD031607\u003c/span\u003e\u003cspan address=\"10.1029/2019JD031607\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. e2019JD031607\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTamoffo AT, Vondou DA, Pokam WM, Haensler A, Yepdo ZD, Fotso-Nguemo TC, Nouayou R (2019) Daily characteristics of Central African rainfall in the REMO model. Theor Appl Climatol 137:2351\u0026ndash;2368. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00704-018-2720-z\u003c/span\u003e\u003cspan address=\"10.1007/s00704-018-2720-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTamoffo AT, Weber T, Akinsanola AA, Vondou DA (2023) Projected changes in extreme rainfall and temperature events and possible implications for Cameroon's socio-economic sectors. Meteorol Appl 30:e2119. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/met.2119\u003c/span\u003e\u003cspan address=\"10.1002/met.2119\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTang Y, Rumbold S, Ellis R, Kelley D, Mulcahy J, Sellar A, Jones C (2019) MOHC UKESM1. 0-LL model output prepared for CMIP6 CMIP historical. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.22033/ESGF/CMIP6.13400\u003c/span\u003e\u003cspan address=\"10.22033/ESGF/CMIP6.13400\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. World Data Center for Climate (WDCC) at DKRZ\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTatebe H, Ogura T, Nitta T, Komuro Y, Ogochi K, Takemura T, Kimoto M (2019) Description and basic evaluation of simulated mean state, internal variability, and climate sensitivity in MIROC6. Geosci Model Dev 12:2727\u0026ndash;2765. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/gmd-12-2727-2019\u003c/span\u003e\u003cspan address=\"10.5194/gmd-12-2727-2019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTaylor KE (2001) Summarizing multiple aspects of model performance in a single diagram. J Geophys Res Atmos 106:7183\u0026ndash;7192. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2000JD900719\u003c/span\u003e\u003cspan address=\"10.1029/2000JD900719\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evan Der Walt AJ, Fitchett JM (2021) Trend analysis of cold extremes in South Africa: 1960\u0026ndash;2016. Int J Climatol 41:2060\u0026ndash;2081. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/joc.6828\u003c/span\u003e\u003cspan address=\"10.1002/joc.6828\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVoldoire A, Saint-Martin D, S\u0026eacute;n\u0026eacute;si S, Decharme B, Alias A, Chevallier M, Waldman R (2019) Evaluation of CMIP6 deck experiments with CNRM‐CM6‐1. J Adv Model Earth Syst 11:2177\u0026ndash;2213. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2018MS001683\u003c/span\u003e\u003cspan address=\"10.1029/2018MS001683\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVolodin E, Mortikov E, Gritsun A, Lykossov V, Galin V, Diansky N, Emelina S (2019) IPCC DDC: INM INM-CM5-H model output prepared for CMIP6 HighResMIP control-1950. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.22033/ESGF/CMIP6.13444\u003c/span\u003e\u003cspan address=\"10.22033/ESGF/CMIP6.13444\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. World Data Center for Climate (WDCC) at DKRZ\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWilliams C, Lunt D, Singarayer J, Guarino MV (2021) CMIP6 PMIP NERC HadGEM3-GC31-LL lig127k. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.22033/ESGF/CMIP6.13601\u003c/span\u003e\u003cspan address=\"10.22033/ESGF/CMIP6.13601\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. World Data Center for Climate (WDCC) at DKRZ\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWu T, Zhang F, Zhang J, Jie W, Zhang Y, Wu F, Hu A (2020) Beijing Climate Center Earth System Model version 1 (BCC-ESM1): model description and evaluation of aerosol simulations. Geosci Model Dev 13:977\u0026ndash;1005. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/gmd-13-977-2020\u003c/span\u003e\u003cspan address=\"10.5194/gmd-13-977-2020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZebaze S, Jain S, Salunke P, Shafiq S, Mishra SK (2019) Assessment of CMIP5 multimodel mean for the historical climate of Africa. Atmos Sci Lett 20:e926. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/asl.926\u003c/span\u003e\u003cspan address=\"10.1002/asl.926\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZwiers FW, Alexander LV, Hegerl GC, Knutson TR, Kossin JP, Naveau P, Zhang X (2013) Climate extremes: challenges in estimating and understanding recent changes in the frequency and intensity of extreme climate and weather events. Climate science for serving society: research, modeling and prediction priorities. Springer, pp 339\u0026ndash;389. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-94-007-6692-1_13\u003c/span\u003e\u003cspan address=\"10.1007/978-94-007-6692-1_13\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"theoretical-and-applied-climatology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"taac","sideBox":"Learn more about [Theoretical and Applied Climatology](https://www.springer.com/journal/704)","snPcode":"704","submissionUrl":"https://submission.nature.com/new-submission/704/3","title":"Theoretical and Applied Climatology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Extreme temperature events, Central Africa, CMIP6","lastPublishedDoi":"10.21203/rs.3.rs-7155223/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7155223/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eExtreme temperature events pose significant risks to both natural environments and communities across Central Africa (CA).\u003c/span\u003e Gaining deeper insight into how these events vary is crucial to inform effective climate change mitigation and adaptation plans. \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe present study evaluates the suitability of global climate models from the Coupled Model Intercomparison Project Phase 6 (CMIP6), along with their multi-model ensemble mean (MME), in simulating the recent past spatial variations of extreme temperature events in CA. For this purpose, under the period 1985\u0026ndash;2014, we assessed seven relevant indicators based on daily minimum and maximum temperatures, recommended by the Expert Team on Climate Change Detection and Indices (ETCCDI). We examined the spatial patterns of these extreme temperature events as simulated by sixteen CMIP6 models and their MME against CHIRTS and ERA5 observational and reanalysis datasets, focusing on percentile, absolute, and duration-based indices. The results showed that both individual models and the MME demonstrated reasonable skill in reproducing the spatial patterns of most extreme temperature indices, with the MME often showing better agreement with observations. However, the models faced challenges in accurately simulating the frequency of the percentile-based indices, notably TX90p and TN10p, across all sub-regions. Furthermore, the model performance varies depending on the specific index and the climatic sub-region. This study thereby elucidates the capabilities and shortcomings of CMIP6 models in representing extreme temperatures across Central Africa, providing valuable information for developing more reliable regional climate projections and assessing the future impacts of climate change on temperature extremes.\u003c/span\u003e\u003c/p\u003e","manuscriptTitle":"Evaluation of extreme temperature events as simulated by CMIP6 models over Central Africa","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-20 13:14:49","doi":"10.21203/rs.3.rs-7155223/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-09-19T18:50:47+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-12T13:57:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"149466915042026246755958887898577846452","date":"2025-08-19T03:13:40+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-14T10:24:42+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"5976496423702150030723762427136309765","date":"2025-08-12T15:38:41+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-12T12:09:20+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-20T22:19:59+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-20T22:19:10+00:00","index":"","fulltext":""},{"type":"submitted","content":"Theoretical and Applied Climatology","date":"2025-07-18T07:50:59+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"theoretical-and-applied-climatology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"taac","sideBox":"Learn more about [Theoretical and Applied Climatology](https://www.springer.com/journal/704)","snPcode":"704","submissionUrl":"https://submission.nature.com/new-submission/704/3","title":"Theoretical and Applied Climatology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"245cffb8-81ce-48d7-980a-8c9580b00417","owner":[],"postedDate":"August 20th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-03-09T16:01:58+00:00","versionOfRecord":{"articleIdentity":"rs-7155223","link":"https://doi.org/10.1007/s00704-026-06065-6","journal":{"identity":"theoretical-and-applied-climatology","isVorOnly":false,"title":"Theoretical and Applied Climatology"},"publishedOn":"2026-03-05 15:58:40","publishedOnDateReadable":"March 5th, 2026"},"versionCreatedAt":"2025-08-20 13:14:49","video":"","vorDoi":"10.1007/s00704-026-06065-6","vorDoiUrl":"https://doi.org/10.1007/s00704-026-06065-6","workflowStages":[]},"version":"v1","identity":"rs-7155223","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7155223","identity":"rs-7155223","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}