A Trend Analysis of Changes in Cooling Degree Days in West Africa Under Global Warming

Abstract

Monitoring energy consumption in response to rising temperatures has become extremely important in all regions of the globe. The energy required for cooling is a major challenge in West Africa, where the climate is predominantly tropical. Among the various methods for evaluating energy requirements, the degree-day method is best known for its ability to estimate the heating, ventilation, and air-conditioning (HVAC) requirements of buildings. This study used three decades of weather station data to assess the cooling degree days (CDD) in two major West African cities, Kano and Bamako, across a range of base temperatures from 22 °C to 30 °C. The results indicate an increase in cooling degree days for Kano, whereas Bamako experienced a decrease in these parameters over the same period. Nonetheless, Bamako required a relatively higher cooling demand for all base temperatures. Furthermore, the study showed that the years 1998 and 2015 had the most significant impact on Kano and Bamako, with CDD values ranging from 2220 °C-day to 218 °C-day for Kano and from 2425 °C-day to 276 °C-day for Bamako. The study also found that a lower base temperature leads to higher energy consumption, while a higher base temperature leads to lower energy consumption. This information provides a useful reference for governments and policymakers to achieve energy efficiency and reduce greenhouse gas emissions.

1. Introduction

The ambition of limiting the global temperature rise to 1.5 °C under the Paris Agreement is getting increasingly out of reach as global warming continues to increase at a steady pace. Rising temperatures coupled with world population growth will continually increase the demand for cooling energy in several regions of the world [1]. Between 2014 and 2040, global energy demand in non-OECD countries is expected to increase by approximately 25%, in line with an increase in population [2]. Recent scientific dialogues have underscored that cities consume up to 60 and 80% of energy and account for 75% of CO₂ emissions [3]. As such, residential buildings contribute nearly 40% to the total energy consumption [4], and are therefore responsible for a significant amount of CO₂ emissions [5]. In response to rising temperatures, different economic sectors are likely to experience increased cooling needs during the hot season and decreased heating needs during the cold season.

In the past few decades, several approaches have been employed to evaluate residential heating and cooling energy consumption and assess the impacts of climate change. Such approaches include the split-degree-day method [6], hourly method [7], monthly method [8], Schoenau and Kehrig method [9], and hybrid S–K method [10]. However, the degree-day method using daily mean temperature is by far the simplest and most popular when daily data are available. Degree-day calculations require the consideration of two indices: the heating degree day (HDD) and cooling degree day (CDD). The HDD refers to the positive result of any difference between a defined base temperature and the average daily temperature when the average temperature is higher than the base temperature, while CDD represents the positive result of any difference between the same defined base temperature and the average daily temperature when the average temperature is lower than the base temperature [11]. The base temperature is the threshold at which heating or cooling is necessary in a building to keep its occupants in comfortable living conditions. Furthermore, the decision to estimate HDD, CDD, or both strongly depends on the aims of a particular study and the weather conditions of the study location. For example, relatively cold locations may require heating calculations, whereas regions with hot climates may require cooling calculations.

In West Africa, the progressive rise in global temperatures is expected to lead to significant increases in the energy needed to cool buildings for human comfort because the climate is typically quite hot. Few researchers have investigated the changes affecting the cooling energy demand within the West African context using historical station weather and global climate model (GCM) data. For instance, Akara et al. [12] estimated CDD values for different cities in West Africa using weather data from 1980 to 2014. Similarly, [13] explored the future cooling demand for several cities in West Africa using global climate model (GCM) data. Furthermore, Awolola and Olorunmaiye [14] examined the cooling energy demand using hourly meteorological data across several Nigerian cities between 1995 and 2009, using different base temperatures. Falchetta and Mistry [15] used both historical and Coupled Model Intercomparison Project Phase 6 (CMIP6) data to estimate CDD across several countries in West Africa and beyond. In addition, there have been many other studies estimating cooling degree days in various regions of the globe. For example, Bilgili [16] examined the impact of urbanization on cooling degree days in metropolitan areas in Turkey, while De Rosa et al. [17] focused on historical trends in cooling degree days across different climates in Italy. In Australia, Harvey [18] analyzed the effects of climate change on cooling needs in residential buildings. Meanwhile, Idchabani et al. [19] provided a regional assessment of cooling degree days in North Africa. Li et al. [20] studied the implications of cooling degree days on energy consumption in industrial sectors in China, while Li et al. [21] investigated the correlation between cooling degree days and public health outcomes in China. Moreover, Miranda et al. [22] explored the geographical distribution of cooling degree days in South America. Research by Scoccimarro et al. [23] explored future projections of cooling degree days under different climate scenarios in Italy. Furthermore, Shi et al. [24] evaluated the economic impacts of cooling degree days on agricultural productivity in China, while Spinoni et al. [25] provided a comprehensive review of global cooling degree-day trends in Europe. Lastly, Ukey and Rai [26] focused on the implications of cooling degree days for energy efficiency in buildings in India.

However, despite the rising global temperatures, the availability of reliable cooling degree day (CDD) records remains limited in several locations, thereby making it more difficult for policymakers to strengthen energy policies and advance climate action. This lack of data is particularly challenging in Africa, where acquiring station data involves considerable financial resources. Furthermore, while a single base temperature offers a basic measure, it often fails to reflect varying comfort levels, climate zones, or energy needs accurately. To address these gaps, our study analyzes CDD across multiple base temperatures using three decades of daily meteorological data for two major West African cities, Kano and Bamako. To the best of our knowledge, comprehensive assessments of CDD in these cities have not been conducted. This study aims to evaluate three-decade trends in cooling degree days (CDD) for Kano and Bamako across multiple base temperatures, with implications for enhancing regional energy policies and climate adaptation.

2. Materials and Methods

2.1. Study Area and Data

The research was conducted in two West African cities, Kano and Bamako, located in Nigeria and Mali, respectively (Figure 1). Kano, the capital of Kano State, is located within UTM Zone 32 N at 12°03′29″ N and 8°33′54″ E, with an elevation of 472 m above sea level. Meanwhile, Bamako is the capital and largest city of the Republic of Mali, and is situated at 380 m above the sea at coordinates 12°35′26″ N and 7°56′15″ W in UTM Zone 29 N. Both cities belong to the Sudanian zone of West Africa, with the annual precipitation of approximately 900 mm recorded in Bamako and about 700 mm recorded in Kano, and the average monthly temperature in the rainy season ranges from 26.5 to 31.5 °C in Bamako, compared with 26.5 to 33 °C in Kano [27].

For each city, 30 years of daily maximum and minimum temperatures were acquired in Microsoft Excel for both Kano and Bamako through the Nigerian Metrological Office (NiMet) and Malian National Agency of Meteorology, respectively. Over the years, these government weather forecasting agencies have been responsible for gathering and disseminating accurate meteorological data across the respective countries. In light of the size of the data, thorough quality control and data homogenization verification were performed following the recommendations of Aguilar et al. [28]. This process was used to essentially check for missing values, and the days, months, and years for all data, along with the coherence of the data, for example, the maximum temperature exceeded the minimum temperature. The daily average temperature for each day was estimated for both cities based on the daily records for the entire study period.

2.2. Methodology

For this study, all data were computed in Python using the Jupiter notebook. The cooling degree days for each year were obtained by summing the positive values derived from the difference between the threshold temperature and the average temperature for each day of the year. As mentioned previously, the average temperature was calculated from the difference between the maximum and minimum temperatures on each day during the entire study period. Furthermore, although Kano and Bamako share the same climatic zone, differences in building design and energy planning may affect the need for domestic energy between the two locations. To account for these variations, we used nine base temperatures (22 °C to 30 °C) in calculating CDD for each city. This range covers typical cooling thresholds in energy demand studies [29,30,31,32,33], and was selected to reflect diverse cooling needs in similar climates. This approach minimizes subjectivity and offers a detailed view of cooling demand variability across comfort and energy use patterns in both locations. Therefore, CDD was computed using the following equation:

$$CDD=\sum _{k=1}^{n}f\left(\phi \right)$$

(1)

with

$$f\left(\phi \right)=T_m-T_b,if{T}_b<T_m$$

(2)

Otherwise, $f\left(\phi \right)=0$.

$$T_m={\displaystyle \frac{\mathrm{T}max-\mathrm{T}min}{2}}$$

(3)

where CDD is the number of cooling degree days; n is the number of days in a year.

• T_m (°C): Mean temperature of the day k;
• T_b (°C): Base temperature;
• Tmax (°C): Maximum temperature of the day k;
• Tmin (°C): Minimum temperature of the day k.

To further investigate the changes in the CDD time series, we applied three statistical tools to the data for each station. Initially, we used linear regression to model the overall trend in CDD across the study period. Linear regression is a commonly used approach in climatology for assessing relationships and estimating temporal changes. This technique has been widely applied to evaluate temperature-related indices, such as CDD, due to its simplicity and ability to illustrate general trends [34,35]. The linear regression formula is an affine function, expressed as follows:

$$Y=ax+b$$

(4)

where Y is the values in each year between 1992 and 2022 and x is the year over the study period 1992–2022.

Subsequently, the Mann–Kendall test [36,37], a non-parametric method, was applied to identify trends in the CDD time series at a 95% confidence level. This method is well documented in CDD studies as it allows for a trend analysis without the need for data transformation or distribution assumptions, making it particularly suitable for meteorological datasets [38,39]. The test was calculated as follows:

$$S=\sum _{k=1}^n\sum _{i=k+1}^{n}sgn(x_i-x_k)$$

(5)

where n is the sample size, $x_i$ and $x_k$ being the values of individual time series i and k, respectively. The term $\mathrm{s}\mathrm{g}\mathrm{n}\left(x_i-x_k\right)$ represents the sign function applied to the difference between $x_i$ and $x_k$, indicating whether this difference is positive, negative, or zero. The sign function $\mathrm{s}\mathrm{g}\mathrm{n}\left(x_i-x_k\right)$ is given by the following formula:

$$\mathrm{s}\mathrm{g}\mathrm{n}\left(x_i-x_k\right)=\left\{\begin{array}{c}+1,if{x}_i-x_k>0\\ 0,if{x}_i{-x}_k=0\\ -1,if{x}_i-x_k<0\end{array}\right.$$

(6)

Furthermore, to quantify the magnitude and direction of the trends, we applied Sen’s slope estimator [40], a robust median-based method that is ideal for non-normally distributed data. In contrast to least-squares regression, which may be sensitive to anomalies, Sen’s estimator provides a more stable trend estimate by calculating a median slope. This approach has been recommended for environmental studies due to its resilience to extreme values and its ability to yield consistent trend slopes in temperature and CDD series [41]. The Sen’s slope equation is as follows:

$$Q_i={\displaystyle \frac{x_i-x_k}{i-k}}fori=1,\dots ,N,$$

(7)

where $x_i$ and $x_k$ represent the sample data measured at times i and k (i > k), respectively, and N is the number of slopes.

3. Results

The evolution of cooling degree days over the years for nine base temperatures covering 22 °C to 30 °C for Kano and Bamako is shown in

Table 1. Yearly cooling degree days for Kano at different base temperatures.

	Base Temperature
Year	22 °C	23 °C	24 °C	25 °C	26 °C	27 °C	28 °C	29 °C	30 °C
1992	1628	1331	1054	803	585	409	272	170	99
1993	1824	1504	1206	932	685	478	323	215	137
1994	1725	1406	1101	817	569	377	243	150	87
1995	1733	1403	1097	823	589	399	259	163	96
1996	1867	1517	1196	912	669	464	314	210	130
1997	2007	1681	1365	1062	786	548	365	231	134
1998	1933	1613	1304	1012	748	534	376	263	179
1999	1902	1582	1275	984	736	548	404	287	184
2000	1753	1445	1154	888	661	476	333	227	148
2001	1651	1341	1052	792	571	394	254	155	92
2002	1941	1638	1347	1070	825	614	446	318	215
2003	1881	1557	1249	967	825	517	357	238	159
2004	1927	1609	1309	1029	779	557	377	239	142
2005	2026	1709	1402	1119	861	639	452	313	204
2006	1854	1527	1215	926	663	444	274	157	82
2007	1814	1495	1194	911	651	433	282	179	111
2008	1758	1434	1133	859	614	418	268	155	79
2009	1952	1619	1309	1016	743	506	326	198	117
2010	2089	1760	1450	1154	881	645	457	315	206
2011	1833	1533	1245	971	722	509	339	221	142
2012	1889	1565	1254	964	708	506	352	238	158
2013	1872	1552	1251	976	738	540	384	260	162
2014	1654	1374	1112	867	646	465	323	211	125
2015	2220	1884	1557	1247	961	711	500	338	218
2016	1938	1639	1355	1088	843	632	468	339	239
2017	1809	1487	1192	923	683	485	338	235	160
2018	1863	1563	1279	1009	762	555	391	265	167
2019	1851	1540	1245	968	719	517	362	249	163
2020	1764	1460	1187	942	729	552	412	307	217
2021	1882	1573	1282	1006	758	562	410	287	184
2022	1427	1159	912	692	511	372	267	188	118

and

Table 2. Yearly cooling degree days for Bamako at different base temperatures.

	Base Temperature
Year	22 °C	23 °C	24 °C	25 °C	26 °C	27 °C	28 °C	29 °C	30 °C
1992	2272	1908	1551	1211	900	629	411	255	155
1993	2349	1996	1652	1324	1023	763	541	361	233
1994	2156	1795	1448	1128	841	603	424	294	191
1995	2143	1785	1438	1118	829	583	393	253	152
1996	2306	1941	1580	1227	905	642	436	277	173
1997	2196	1834	1482	1142	822	543	329	194	111
1998	2425	2062	1707	1370	1060	788	568	400	276
1999	2076	1721	1380	1071	803	605	462	339	234
2000	2137	1775	1422	1097	808	575	400	275	181
2001	2286	1921	1563	1218	904	640	441	291	185
2002	2374	2010	1650	1303	985	724	520	361	243
2003	2257	1897	1546	1223	936	699	513	369	252
2004	2216	1853	1502	1168	865	615	419	270	162
2005	2318	1954	1600	1260	939	666	461	311	200
2006	2009	1655	1317	1002	726	507	344	229	146
2007	2313	1960	1616	1287	986	733	531	378	257
2008	2077	1735	1408	1096	816	581	401	267	167
2009	2286	1932	1589	1264	976	720	517	354	234
2010	2292	1935	1590	1266	974	721	518	363	243
2011	2070	1714	1376	1064	791	576	411	286	188
2012	1923	1572	1233	925	648	431	283	182	107
2013	2220	1871	1540	1223	939	693	504	362	249
2014	2143	1785	1434	1103	803	556	368	235	148
2015	2114	1764	1433	1124	841	601	420	286	181
2016	2278	1916	1560	1219	904	638	438	296	196
2017	2255	1897	1546	1215	921	675	479	336	223
2018	2166	1807	1456	1126	828	584	398	257	160
2019	2098	1747	1412	1096	806	567	394	273	182
2020	2058	1703	1359	1038	762	538	378	260	171
2021	2267	1903	1543	1197	880	610	399	248	149
2022	2093	1736	1396	1083	804	575	400	276	177

, respectively. Our analysis of CDD values across the two locations, using statistical methods such as linear regression, the Mann–Kendall test, and Sen’s slope estimator, revealed significant discrepancies in cooling degree days between the two locations. Kano consistently exhibited lower CDD values over the period from 1992 to 2022.

In terms of years with the highest cooling demand, it was found that 2015 recorded the maximum number of cooling degree days in Kano with 2220 °C-day, 1884 °C-day, 1557 °C-day, 1247 °C-day, 961 °C-day, 711 °C-day, 500 °C-day, 338 °C-day, and 218 °C-day for the lowest to the highest base temperatures, respectively. On the other hand, Bamako experienced higher cooling degree days in 1998 with 2425 °C-day, 2062 °C-day, 1707 °C-day, 1370 °C-day, 1060 °C-day, 788 °C-day, 568 °C-day, 400 °C-day, and 276 °C-day across the multiple base temperatures. The rise in CDD in Kano and Bamako during 1998 and 2015 could be linked to global warming, with strong El Niño events further intensifying these effects in West Africa, thereby increasing cooling demand.

Furthermore, the results revealed that Bamako experienced the lowest cooling energy demand in 2012 across all base temperatures with a number of cooling degree days ranging from just 1923 °C-day for the coldest threshold temperature (e.g., 22 °C) to 107 °C-day for the hottest threshold temperature (e.g., 30 °C). Meanwhile, Kano recorded its lowest CDD values in 2022 with 1427 °C-day for the lowest threshold temperature (e.g., 22 °C) to 188 °C-day for the highest threshold temperature (e.g., 30 °C).

The temporal evolution of cooling degree days in Kano is shown in Figure 2. The findings indicated a positive correlation between CDD and threshold temperatures over time. However, only CDD calculated using thresholds 28 °C, 29 °C, and 30 °C showed a significant increasing trend with a p-value below 0.05, whereas CDD values for thresholds between 22 °C and 27 °C also indicated an increase, but the trend was not statistically significant. This discrepancy therefore demonstrates that a base temperature below 28 °C is not sufficient to have an impact on the need for cooling in Kano. In earlier years, Kofar-Bai and Zheng [42] recorded cooling degree hours for Kano as 6104.8, 5490.1, and 4746.1 using base temperatures of 24 °C, 26 °C, and 28 °C, respectively. When converted to cooling degree days (e.g., 254.37 °C-day, 228.75 °C-day, and 197.75 °C-day), these values were considerably lower than those found in this study (e.g., 38,283 °C-day, 22,221 °C-day, 10,928 °C-day, and 4654 °C-day) for the same base temperature, reflecting an increase in cooling energy demand.

On the other hand, the findings have shown that the CDD has decreased over the years in Bamako (Figure 3), indicating a decreasing need for cooling (negative Sen’s slope values). Nonetheless, none of the CDD showed a statistically significant change in Bamako with a p-value greater than 0.05 for all base temperatures.

Furthermore, the results also showed that CDD values decrease significantly whenever the base temperature rises and increase whenever the base temperature used decreases for both cities (Figure 4).

Additionally, Figure 5 and Figure 6 illustrate the linear relationship between the calculated cooling degree days for each base temperature and the mean annual temperature for Kano and Bamako, respectively. For both cities, CDD values increase as mean temperature rises, indicating a consistent trend across all base temperatures. For example, at the lowest threshold (e.g., 22 °C), the correlation coefficient is R² = 0.740 for Kano and R² = 0.458 for Bamako. As the base temperature increases (e.g., to 30 °C), the correlation weakens, with R² values of 0.202 for Kano and 0.315 for Bamako. This suggests that lower thresholds are more effective for assessing cooling demands in Kano, while Bamako exhibits a relatively steady correlation across thresholds.

Furthermore, temperatures in Kano are generally lower than those in Bamako, with annual averages ranging from 25 °C to 28 °C in Kano (

Table 3. Annual temperature data for Kano.

Year	Maximum Temperature (°C)	Minimum Temperature (°C)	Annual Average Temperature (°C)
1992	33	20	26
1993	34	20	27
1994	33	20	27
1995	33	20	27
1996	34	20	27
1997	34	21	27
1998	33	21	27
1999	34	20	27
2000	33	20	27
2001	33	19	26
2002	34	20	27
2003	24	20	27
2004	34	20	27
2005	34	21	27
2006	33	21	27
2007	33	20	27
2008	33	20	27
2009	33	21	27
2010	34	21	28
2011	34	20	27
2012	34	20	27
2013	34	19	27
2014	34	19	26
2015	35	21	28
2016	34	20	27
2017	34	20	27
2018	34	20	27
2019	34	20	27
2020	34	20	27
2021	35	19	27
2022	33	18	25

) compared to 27 °C to 29 °C in Bamako (

Table 4. Annual temperature data for Bamako.

Year	Maximum Temperature (°C)	Minimum Temperature (°C)	Annual Average Temperature (°C)
1992	34	22	28
1993	35	22	28
1994	34	22	28
1995	34	22	28
1996	35	22	28
1997	34	22	28
1998	35	22	29
1999	33	22	28
2000	34	33	28
2001	35	22	28
2002	35	22	29
2003	35	22	28
2004	35	21	28
2005	35	22	28
2006	35	20	27
2007	35	22	28
2008	35	20	28
2009	35	22	28
2010	35	21	28
2011	34	21	28
2012	34	20	27
2013	35	21	28
2014	35	21	28
2015	35	21	28
2016	35	22	28
2017	35	21	28
2018	35	21	28
2019	35	21	28
2020	35	21	28
2021	35	21	28
2022	35	21	28

). This difference in mean temperatures directly contributes to the higher CDD in Bamako compared to Kano.

4. Discussion

In this study, the cooling energy demand of two major West African cities was estimated, using multiple base temperatures. As global temperatures rise, monitoring cooling requirements is of paramount importance, especially in regions with a warmer climate. Cooling degree days was found to be higher in Bamako than in Kano. This aligns with the findings of [43], which showed that from 1964 to 2013, Nigeria had a cooling demand of 3496 °C-days, while Mali had a higher demand of 4104 °C-days. Notably, despite the overall difference, we observed a positive trend in Kano, contrasting with a negative trend in Bamako. Similar increases in cooling degree days have been observed in other regions, including South Africa [44], China [45], Portugal [46], and Bangladesh [38]. In Kano, 2015 marked the most significant impact on cooling energy demand. These findings are consistent with studies such as those by Karl et al. [47] and Tollefson [48], describing 2015 as one of the hottest years on record.

The global rise in cooling degree days is causing energy demands to increase in several regions. Yet, the Sahel region is reported to have the highest levels of unmet cooling needs, highlighting a stark inequality in cooling accessibility [15]. Conversely, situations do exist where despite global warming, cooling degree days decrease, a trend likely influenced by regional adaptations, seasonal shifts, or improvements in energy efficiency. For example, our findings have shown a decreasing trend in cooling needs in Bamako. Similar occurrences have also been observed in Shahrekord, Iran [39]. The cooling demand is naturally responsive to temperature shifts. Therefore, the slight decrease in the number of cooling degree days for Bamako may suggest a decrease in temperature over the last three decades. Corroborating our findings, Touré et al. [49] observed a decline in warming trends for both day and night temperatures in Bamako from 1961 to 2014. Our observations also align with the study by Sawadogo et al. [50], which also reported decreasing temperature trends under the SSP1–2.6 scenario in Burkina Faso, a trend that could contribute to a reduction in cooling energy requirements.

Global climate change has evidently caused disruptions in the temperatures of both cities, thereby affecting their demand for cooling energy differently. Given the sensitivity of energy demand to weather conditions, several studies have shown that some regions of West Africa experience more intense warming than others [51,52,53]. This suggests that a higher base temperature is required to optimize HVAC systems and improve the energy efficiency of buildings, while lower base temperatures could increase electricity consumption. Moreover, it also emerged that although both cities belong to the same climate zone, Bamako has higher energy requirements than Kano, when the same air-conditioning setpoint is used. For this reason, efforts to maintain ambient indoor temperature are likely to be more demanding in Bamako than in Kano.

Furthermore, the relationship between cooling degree days and mean annual temperature showed stronger correlations at lower base temperatures in both locations, weakening as the base temperature increased. Similar patterns were documented by Bilgili et al. [54] in Türkiye and Mourshed [7], who established the relationship between cooling degree days and base temperature across various locations. Indeed, when the base temperature is low, a broader range of temperatures may contribute to the cooling degree days, resulting in a stronger correlation with the mean annual temperature. Similarly, when the base temperature is high, only days with temperatures well above the threshold temperature contribute to the cooling degree days, resulting in a weaker correlation.

In light of the observed trends in cooling degree days and their subsequent relationship with temperature changes, specific policy measures are necessary to meet the cooling energy demands at different locations. A city like Bamako, where cooling needs are pronounced, needs policymakers to take action to improve the energy efficiency of buildings, promote passive cooling techniques, and adopt energy-efficient cooling technologies. Additional financial incentives, such as tax breaks or subsidies for energy-efficient appliances and solar air-conditioning systems, may help to reduce cooling energy consumption. Furthermore, changes in construction practices and energy standards are essential to ensure better energy efficiency in new and existing buildings. This will help to mitigate the demand for cooling energy on a long-term basis, enabling cities to adapt to rising temperatures without overburdening the energy infrastructure.

5. Conclusions

The present study investigated the changes in annual cooling degree days and temperature for Kano and Bamako, two large cities in West Africa, using long-term station data for a period of 30 years. Various base temperatures (e.g., 22 °C to 30 °C) were used to perform the analysis. Based on the findings, both cities have witnessed significant variations in cooling degree days since 1992, with an observed increase in the number of CDD for Kano and a slight decrease in cooling degree days for Bamako over the years. Both regions have been adversely affected by the unequal effects of global warming, notwithstanding being situated in the same Sudan Savannah climate zone. However, during the study period, Bamako experienced significantly higher cooling degree days than Kano at all the measured base temperatures. The years 1998 and 2015 were when the cooling degree-day values peaked in Bamako and Kano, respectively. In addition, the lowest base temperatures contributed to the highest cooling degree-day values, whereas the number of cooling degree days became considerably lower when the base temperature reached 30 °C.

In light of these findings, policymakers and building landlords are encouraged to adopt region-specific policies that promote energy-efficient and smart cooling technologies to meet local climate demands. For instance, Bamako may require efficient designs to manage its high cooling needs, while Kano could benefit from smart solutions to address its rising demand. However, this study provides valuable insights into the trends and relationships of cooling degree days (CDD) across multiple base temperatures; one key limitation should be acknowledged. Our analysis relied solely on historical CDD data, without incorporating climate projection models. Consequently, the findings do not account for potential shifts in CDD under future climate scenarios, which could influence temperature trends, seasonal patterns, and energy demands. Future research could benefit from incorporating climate projection models to simulate anticipated changes in CDD under different greenhouse gas scenarios, which would provide a clearer picture of potential future conditions.