Blessing Funmbi Sasanya

Theoretical and Applied Climatology -:118 https://doi.org/10.1007/s- RESEARCH Spatio-temporal analysis of rainfall over Chad River Basin, Nigeria Blessing Funmbi Sasanya1 · Sunday Olufemi Adesogan2 · Akeem Abiodun Ademola2 Received: 8 October 2024 / Accepted: 27 December 2024 © The Author(s), under exclusive licence to Springer-Verlag GmbH Austria, part of Springer Nature 2025 Abstract The substantial reduction in the size of Lake Chad, approaching 90% of its original size in the 1960s, has raised global concerns. This research is designed to identify critical junctures in the time series of annual precipitation data, leading to abrupt shifts in trends across various segments of the Lake Chad River Basin. Annual and seasonal rainfall data spanning forty-one years -) were collected from twenty-five monitoring stations, sourced from the National Aeronautics and Space Administration (NASA) website. The time series data were assessed for serial correlation and non-parametric statistical tests, (Mann-Kendall (MK), Modified Mann-Kendall (MMK), and Sequential Mann-Kendall (SMK)) at 5% significance level. Results revealed that 28% of the locations exhibited statistically significant inter-annual increases in rainfall, while 20% displayed statistically significant inter-annual decreases. Furthermore, 32% of the locations recorded statistically significant upward trends in inter-seasonal rainfall, while 8% indicated statistically significant declines. The application of the SMK test unveiled that 48% of the locations experienced statistically significant sudden alterations in inter-annual rainfall, with a similar abrupt shift occurring in 40% of the locations for inter-seasonal rainfall. These erratic precipitation patterns are significantly impacting water resources management and this has global implications. 1 Introduction Global warming has led to increasing or reducing annual rainfall amount, frequency and intensity (Croitoru et al. 2013). The frequency of phenomena which includes droughts and floods can be attributed to extreme changes in rainfall patterns (Srivastava et al. 2015). The consistent alteration of precipitation quantities, occurrences and types is attributable to climate change and global warming. Focus on climate change subjects is crucial for a healthy earth environment. Discuss and research on the subject must be global, from all regions (Marques et al. 2020). Climate change, caused by natural and anthropogenic activities, has led to a build-up of heat-trapping substances, marked by rising temperatures, altered rainfall patterns, and increased occurrence of extreme weather events (UNFCCC 2007). Such extreme events may include wind storms, heavy rainfall, floods, * Blessing Funmbi Sasanya-1 Department of Crop and Soil Science, Faculty of Agriculture, University of Port Harcourt, Port Harcourt, Nigeria 2 Department of Civil Engineering, Faculty of Technology, University of Ibadan, Ibadan, Nigeria rising sea levels, droughts, melting glaciers and heat waves (UNFCCC 2007). Rainfall is dynamic in Nigeria and it is subject to significant temporal and spatial fluctuations. Such fluctuations have become more evident with climate change (Itiowe et al. 2019). Adequate understanding of these would facilitate the modelling, simulation and design of hydraulic structures and water facilities, including surface water storage reservoirs, drainages, irrigation appurtenances and dams for flood regulation (Terêncio et al. 2018). Hence, statistical analysis of distribution patterns and changes of precipitation at different space and time dimensions is crucial in devising a potential action that will reduce the impacts of extreme events of precipitation and evaluate the ability of an ecosystem to recover from it. Precipitation patterns may vary seasonally, annually and over decades across the globe. Seasonal patterns of rainfall occur in Nigeria, with the south receiving a peak amount two times a year, while the northern part experiences a maximum amount just once annually (Odjugo 2006). Studying the pattern of rainfall change and trends over time and location is crucial for the responsible exploitation and utilization of water resources and the prevention or mitigation of floods and droughts (Grose et al. 2019; Li et al. 2019). Climate variations have been studied in recent Vol.:-) 118 Page 2 of 16 decades, and findings from such studies have been useful for various areas of knowledge, where better land preservation and water resources management are of great importance. Researchers across the world have developed several methods for the establishment of precipitation analysis either by utilizing statistical methods that accept specific distribution assumptions (parametric) or the ones that do not make such an assumption (non-parametric) (Rowell and Chadwick 2018; Creese et al. 2019). Examples of such non parametric approaches include Spearman’s rho (SR) test Mann-Kendall (MK) test, Theil-Sen’s slope (SS), liner regression, Modified Mann-Kendall (MMK) test, coefficient of variation and cumulative sum. The Mann-Kendall (MK) test is utilized to identify increasing or decreasing trends within time series data (Sa’adi et al. 2017; Wang et al. 2020). The Modified Mann-Kendall (MMK) test is an extension of the MK test specifically designed to consider serial correlation within time series data (Sa’adi et al. 2017; Alashan 2020). The Theil-Sen’s slope (SS) method, on the other hand, provides an estimate of the median slope of a dataset while being robust to outliers (Dagnachew et al. 2020). Linear regression, models the linear relationship between two variables by determining the best-fit line (Xu et al. 2019; Kibria and Lukman 2020). Additionally, the coefficient of variation is a metric used to determine the relative variability within a dataset by comparing the standard deviation to the mean, while the cumulative sum is employed for continuous monitoring and the detection of shifts or alterations within a sequential dataset over time (Woodall et al. 2020). MK trend test and SS method together with statistical and interpolation techniques were employed by Da Silva et al. (2020) to analyse the variability of yearly rainfall over both space and time in a catchment area in Brazil. A declining trend in average rainfall within the catchment region was observed. Groleau et al. (2007) have detected a noteworthy increase in winter precipitation trend in certain areas of Canada using rainfall indices, MK test and trend free prewhitening procedure. The Mann-Whitney test was used by Caloiero et al. (2011) to analyse 50 years of rainfall data in southern Italy. Negative trends during winter and positive trends during summer were reported. AlSubih et al. (2021) recorded a statistically meaningful decreasing trend in most of the rainfall stations in Saudi Arabia, when a trend analysis was conducted on five decades rainfall data. Waha et al. (2022) have predicted a positive summer rainfall and a negative trend for winter rainfall in Western Australia. The study also projected a strong negative trend for future climate scenarios (Waha et al. 2022). In another study in Ethiopia, Harka et al. (2021) have discovered a rising trend in the high precipitation season and a declining trend during the season of low precipitation. Itiowe et al. (2019) have used the coefficient of variation method to analyse 31 years of daily rainfall data -) in Abuja, Nigeria. The study B. F. Sasanya et al. revealed decreasing trends of precipitation in the study area. In the same vein, Abe et al. (2022) have analysed 31 years of annual rainfall data of the Kumadugu-Yobe river basin and the result indicated a rising trend between 1981 and 2017, there was a general inclination towards increased precipitation based on the analysis performed for the study period. Having reviewed these studies, most researchers have investigated and reported increased trends of rainfall in varying river basins. However, to the best of our knowledge, none of the studies investigated the potential turning or abrupt changing points of long-term rainfall data. Therefore, this study aims to establish the turning points within sets of 41 years -) of total annual precipitation data, leading to abrupt trend changes in different parts of the Lake Chad River Basin, within the Nigeria boundary. Forecasts of rainfall intensities for the next twenty years were also made for each location. By examining and forecasting the long-term rainfall trends, the study seeks to provide insights into the changing climate of the region, which can inform strategies for mitigating the effects of climate change on agriculture and water resources. Methods used to achieve these were the Modified Mann Kendall (MMK) test, Mann Kendall (MK) test, Theil’s slope (TS) and Sequential Mann Kendall (SMK) test. Each of these methods have their peculiar advantages and characteristics. Unlike MK, MMK adapts to the presence of serial correlation in time series data when investigating trends. SMK detect abrupt turning points in time series data, while TS are also useful for trend investigations but it is less sensitive to outliers in data (Dagnachew et al. 2020). The findings of this research will offer valuable insight into the behaviour of rainfall patterns in the Chad Basin and inform strategies for managing water resources and agricultural production in the region. Specifically, this study will identify regions of the basin that are most vulnerable to changes in rainfall patterns and provide adequate guidance to engineers, farmers, policymakers and water managers on the most effective approaches for managing water resources in the area, thus providing a scientific basis for decision-making and a roadmap for sustainable agricultural management in the Chad River Basin. 2 Methodology 2.1 Study area and data collection The Chad River Basin is a critical region in Nigeria, providing water resources for millions of people and serving as a major agricultural hub. However, the region has been facing significant challenges in recent years due to changes in rainfall patterns, which have had significant impacts on agricultural productivity, food security and water availability (Pham-Duc et al. 2020; Goni et al. 2021). The understanding Page 3 of 16 Spatio-temporal analysis of rainfall over Chad River Basin, Nigeria of rainfall dynamics over the Chad basin is therefore critical for developing strategies for sustainable management of water resources and agricultural production. The Chad basin covers an area of approximately 470,000 k­ m2, including parts of Nigeria, Chad, Cameroon and Niger. This study is however limited to the part of the Chad Basin within the Nigeria boundary. Figure 1 shows the total land area within the Nigeria boundary, drained by Lake Chad. This amounted to 116,400 k­ m2. Within the Nigeria boundary, the climate of the Chad River basin varies from semi-arid to arid and the basin experiences dry and wet seasons. The total annual rainfall amount within the basin varies highly but averages 611.18 mm and the maximum temperature average ranges between 35 and 40 0C as well as an average temperature of 28.53 0C. The relative humidity in the basin is always low over a long period of the year (41.14%) because long dry seasons and short rainy seasons are usually experienced. Forty-one years -) of annual and seasonal (June to September) rainfall data were gathered from 25 locations within the Chad River Basin Authority catchment in Nigeria. This period was selected as it represented the most recent available dataset at the time of the study. The characteristics of these locations are presented in Table 1. Data for each location were obtained from the National Aeronautics and Space Administration (NASA) website (https://​power.​larc.​nasa.​gov/​data-​access-​viewer/), which is widely regarded for its reliability and accuracy. The dataset is organised on a global grid with a spatial resolution of 0.1° × 0.1° in latitude and longitude, equivalent to a grid size of approximately 10 km (Sasanya et al. 2024). Similar datasets 118 have been employed in previous studies, such as those by Espinoza-Dávalos et al. (2015), Sharma et al. (2020), and Sasanya et al. (2024). The use of satellite data was necessitated by the scarcity of gauging stations and the general unavailability of reliable data across many African regions. 2.2 Data analysis All collected data were analysed using the R studio and statistical analysis packages including ‘modified mk’, ‘trend’, ‘trend change’ and ‘t series’. These enabled the deployment of autocorrelation, Mann Kendall, modified Mann Kendall, the Sequential Mann Kendall tests and Theil’s Sen slope. 2.3 Autocorrelation Autocorrelation also referred to as serial correlation is a statistical concept that describes the measure of the degree of correlation or relationship between a time series and a lagged or shifted version of itself (Box and Jenkins 2013). It is also important for statistical inference in regression analysis to determine the strength of the relationship between two or more variables (Damodar and Gujarati 2009). It provides insights into the trends of variation in time series data to detect any systematic cycle (Brockwell and Davis 2002). Autocorrelation could be positive, negative or zero. The autocorrelation function (ACF), was used in this study to identify the presence of any significant correlations between the time series and its past values at lag-1 for both annual and Fig. 1  Map of the lake chad basin, delineating the lake Chad River Basin, Nigeria 118 Page 4 of 16 B. F. Sasanya et al. Table 1  Geographical coordinates and averages of climate data -) of study area Locations Coordinates 0 - Elevation (m) Average Annual Rainfall (mm/yr) Average Annual Air Temperature (0C) Average Annual Relative Humidity (%) - - - - 0 Latitude ( E) Longitude ( N) - - seasonal time series. The ACF for this study was estimated from Eq. 1. −� −� � ∑ n−k � X − X − X X t+k t t=1 (ACF)k = (1) −� 2 ∑n � X − X t t=1 at time t, Xt+k is Where: Xt is the value of the time series − the value of the time series at time t + k, X is the mean of the time series, n is the number of observation which is 41 for this study and k is the lag = 1. A high ACF value at a specific time lag suggests there is a robust correlation between the time series and the past values at that lag (Durbin and Watson 1950; Chatfield and Xing 2019). A significant ACF in time series data violates the assumption of independent observations which is important for the accuracy of several statistical tests including the Mann Kendall tests, t-test, ANOVA and linear regression (Box et al. 2008; Chatfield and Xing 2019). Autocorrelations in datasets can also result in biased parameter estimates and can affect the precision of confidence intervals (Hamilton 1994; Enders 2014). 2.4 Mann‑kendall test The Mann Kendal (MK) test is a principal non-parametric test or method for pinpointing uniform or monotonic trends of data collected at regular intervals over time. It has the advantage of being able to analyse data with complex or unknown distributions because it does not necessitate making any presumption about the distribution of data (Mann 1945; Kendall 1975). It is capable of handling outliers and missing values without making assumptions about residual distribution. This study utilized the Mann-Kendall test to ascertain the trend in annual and seasonal rainfall data over the period under consideration. The Mann-Kendall Statistic (S) which measures the difference between the count of rising and declining trends in the temporal sequence was estimated from Eqs. 2 and 3. The MK test was performed by calculating the Kendall rank correlation coefficient (τ), which measures the strength and direction of the monotonic relationship between each pair of Page 5 of 16 Spatio-temporal analysis of rainfall over Chad River Basin, Nigeria observations between the time series and time index. τ was computed from Eq. 4, by first arranging the time data and the paired rainfall amount in ascending order of magnitude. ) ∑ n−1 (∑ n sgn(xi − xj ) S= (2) j=i+1 I=1 ⎧ √S−1 , if S > 0 2 ⎪ 𝜎 (S) if S = 0 Z = ⎨ 0, ⎪ √S+1 � if S < 0 ⎩ 𝜎 2 (S) ⎧ 1 ⎪ sgn(xi − xj ) = ⎨ 0 ⎪ −1 ⎩ 2.5 Modified mann‑kendall test 𝜏 = if xj > xi if xj = xi if xj < xi (3) S n(n − 1)∕2 (4) Where n is the data set length or years and ­xi and ­xj are the subsequent data values. The Kendall correlation coefficient ranges from − 1 (showing decreasing or negative trend) to 1 (showing a positive or an increasing trend). Values of τ close to zero indicate no trend. The MK statistics can be approximated to the normal distribution, with a specific mean of 0 and variance of 1 as described by Eqs. 5 and 6, respectively. The variance was estimated by Eq. 6 based on the assumption that some of the data have equal values. Equation 7 was employed to estimate variance when each data in the set are distinct. These mean and variance enable the determination of how significant the trends of time series data are. The null hypothesis ­(H0) of the MK test for this study is that there is an absence of a trend, while the alternative hypothesis ­(H1) indicated either an upward or downward trend in a one-sided or two-sided test (Pohlert 2020). For this study, significance was determined from the p-value corresponding to 95% level of confidence. When the size of the sample (n) exceeds 8 or the null hypothesis is true, the Mann-Kendall test statistic (S) adheres to an approximately normal distribution. The value of the Z statistic was estimated from Eq. 8. A positive trend is possible if Z is greater than 0 and negative trend if Z is less than zero. (5) E(S) = 0 𝜎 2 (S) = n(n − 1)(2n + 5) − ∑m 18 t=1 t(t − 1)(2t + 5) (6) t is the number of data that are equal in a particular group. 𝜎 2 (S) = n(n − 1)(2n + 5) 18 (7) 118 (8) The Modified Mann-Kendall (MMK) method is a statistical method utilized for detecting trends in time series data exhibiting autocorrelation or other forms of dependence. The MMK test differs from the MK test as it deals with the problem of data dependence and non-identical distribution by eliminating the effects of autocorrelation through prewhitening of the time series data before conducting the MK test. This study made use of MMK to evaluate the significance of the trend. as outlined by Hamed and Rao (1998); Yue et al. (2002); and Sharma and Saha (2017). The first step was pre-whitening, which was done by fitting an autoregressive (AR) model (Eq. 9) to the data and determining the values of the model’s parameters by utilizing maximum likelihood estimation (Eqs. 10 and 11). (9) Xt = ∅ Xt−1 + ∈ t Where: ∅ is the autoregressive parameter which indicates the strength of relationship between time series Xt and Xt−1 and ∈ t is a random error for the time t. ) ( ) 1 ∑ n−1 ( log 2𝜋 𝜎 2 − logL ∅ 𝜎 2 = 2 2𝜎 2 n t=2 ( )2 Xt − ∅ X t−1 (10) The log-likelihood function (Eq. 10) was maximised to determine the autoregressive function. Maximizing the loglikelihood function is equivalent to minimising the Sum of Squared Residual (SSR) (Vijayvargia et al. 2023), which resulted to Eq. 11. ∑n Xt Xt−1 ∅ = ∑ t=2� �2 (11) n Xt−1 t=2 Equation 12 was applied to estimate the variance σ2 of the residual 𝜎2= 1 ∑ n−1 n t=2 ( )2 Xt − ∅ X t−1 (12) The pre-whitened series were obtained from Eq. 13. The pre-whitened series got rid of all serial correlation in the original data before the MK test was executed on the newly obtained pre-whitened series, using Eqs. 2–4. The modified test statistics that considered the correlation structure of the data were utilized to account for any residual serial 118 Page 6 of 16 B. F. Sasanya et al. correlation. The modified test statistic followed the normal distribution which enables the estimation of p-values and confidence interval estimation (Patakamuri and O’Brien 2021). The modified variance was computed from Eqs. 13 and 14. 𝜎 2adj = 𝜎 2 . (ESS) (13) 𝜎 2adj is the modified variance, 𝜎 2 is the estimated variance of the test statistics from Eq. 7 and ESS is the Effective Sample Size estimated from Eq. 14. ) ( ∑ 2 n−1 (n − k)rk ESS = n. 1 − (14) k=1 n(n − 1 rk is the autocorrelation at lag k = 1 and n is the total number of tome series = 41. The MK test statistics Z given as Eq. 8 was standardised under the MMK test using the 𝜎 2adj in place of the 𝜎 2. 2.6 Sequential mann‑kendall test The Sequential Mann-Kendall (SMK) test was first initiated by Sneyers (1990) and it is an extension of the MK test. The SMK test is commonly used for establishing trends that already exist in a set of time series data, taking into account the sequential dependence between observations and thus identifying sudden changes in long-term data. Studies have demonstrated that it is more effective than the standard MK test in identifying trends in time series data that exhibit high auto or serial correlation because it provides insights into various aspects of trend changes, including their nature, direction, magnitude, and timing. This makes it particularly useful for detecting abrupt changes or turning points in trends (Hirsch et al. 1982; Yue and Wang 2004). The SMK test involves applying the MMK to sub-divided time series data in sub-periods. The SMK test for this study was conducted following the steps specified by Nasri and Modarres (2009); Sharma and Saha (2017) and Patakamuri and Das (2022). A progressive (u(t) and retrogressive (u� (t)) series were computed to determine the abrupt change in trend or the turning point where both series intersects and continued beyond the 5% level of significance. The progressive series was prepared starting from the beginning of the time series data, while the retrogressive series was estimated starting from the end of the time series data. The test statistic (S) was calculated from Eq. 15, while the reverse statistics was estimated from Eq. 16. These involves comparing the annual mean time series values of ­xj (from j = i + 1 to j = n) with ­xi (from i = 1 to i = n − 1). For each comparison, a count was made for cases where ­xj is greater than ­xi as represented in Eqs. 8 and 9. The parameter k in Eq. 9 represents the time step of the time series data. S= S� = ∑ n−1 ∑ n i=1 ∑n i=k+11 j=i+1 ∑n ) ( sgn xj − xi j=i+1 ) ( sgn xj − xi (15) (16) The means (E(S)) and variances of the test statistics (σ2(S)) were computed from Eqs. 17 and 18. E(S) = n(n − 1) 4 𝜎 2 (S) = n(n − 1)(2j + 5) 18 (17) (18) While the estimations of sequential values of the statistics U(t) or reverse statistics U� (t) were computed from Eqs. 19 and 20. S − E(S) U(t) = √ 𝜎 2 (S)) (19) S� − E(S) U� (t) = √ 𝜎 2 (S)) (20) 3 Results 3.1 Autocorrelation, MMK and MK of rainfall Following a comprehensive analysis of historical precipitation data spanning the preceding 41 years, from 1981 to 2021, within various regions of the Chad River Basin situated within the Nigerian boundary, this study has derived certain empirical findings. Notably, the annual precipitation dataset acquired from specified geographical coordinates, specifically locations 1, 2, 3, 9, 10, 16, 18, 21, 23, and 25, displayed discernible traits of both serial correlation, as shown in Table 2. However, with respect to the seasonal dataset, serial correlations were only observed at locations 9, 10, 16, 18, 21 and 25 (Table 2). In order to mitigate the identified issue of serial correlations, the Pre-whitened Sen’s Slope was computed through the utilization of the Modified Mann Kendall’s (MMK) tests. The findings, subsequently uncovered through the analysis, evince a statistically significant increase in both interannual and inter-seasonal rainfall at specific geographic coordinates, including 1, 9, 10, 16, 18, 21, and 25 (Figs. 2 and 3). It is pertinent to emphasize that seasonal rainfall constitutes a substantial component of the aggregate annual Page 7 of 16 Spatio-temporal analysis of rainfall over Chad River Basin, Nigeria precipitation (Harka et al. 2021). Invariably, 44% of the inter-seasonal rainfall locations exhibited negative trends, but only 8% indicated a significant decrease in rainfall. In the context of Nigeria’s Chad River Basin, it was observed that certain locations, specifically 2, 3, 5, 12, 14, 20, and 23, exhibited an upward trend in both annual and seasonal rainfall (Figs. 2 and 3). However, these trends were not statistically significant. This phenomenon was particularly noticeable in areas situated along the basin’s boundaries, where the presence of water bodies were noted. The Jama’are River was found to be located near regions experiencing a significant increase in rainfall, whereas the Yedseram River lies closer to areas with a non-statistically significant but upward trend in rainfall. The Yedseram River, which flows through 118 Adamawa and Borno States, originates from the Mandara Mountains and discharges into Lake Chad (FAO Fisheries and Aquaculture 2012). In contrast, the Jama’are River originates in Bauchi State and contributes to the Hadejia-Nguru Wetlands (Ibrahim et al. 2022). The Yedseram River exhibits significant seasonal flow variations, largely influenced by the dry season, while the Jama’are River is comparatively less affected by seasonal dryness (FAO Fisheries and Aquaculture 2012). The flow variability of both rivers in response to seasonal changes may partly explain the observed significant and non-significant upward rainfall trends in their respective regions. Similar increasing rainfall trends were reported by Abe et al. (2022) in Gombe, a state in north-eastern Nigeria. Table 2  Autocorrelation, Pre-whitened Sen’s slope, Mann Kendall and Sequential Mann Kendall tests of Annual and Seasonal Rainfall in Chad River Basin, Nigeria Annual Rainfall Locations ACF PSS Seasonal Rainfall (June to September) MK τ Turning Points ACF PSS MK τ -*- −0.23 −0.01 2007, 2008, 2010, 2013,-,-, 1988, 1990, 1994, 1996 - −1.28 −0.15 -, 1985, 1991, 1995, 1996, 2010, 2014, 2016,2017,2018, 2019 0.12 −3.87 −0.20 0.09 −2.63 −-*-* 0.12 −1.21 −0.09 1999, 2000,-, 1986, 1992, 1994, 1999, 2010, 2011,-, 1993,-, 2002, 2007, 2008, 2010, 2011, 2012, 2013,-, 1985, 1992, 1994, 1997, 2008, 2009, 2010, 2015, 2016,-, 200, 2007, 2008, 2010, 2013,-, 1985, 1992, 1994,1997, 2008, 2009, 2010, 2014, 2016,-, 1983, 2012, 2013, 2015, 2019,-, 2000, 2001, 2019,- 1 2 3 4 -*- −0.80 −0.01 5 6 - −1.78 −0.17 - 0.22 −4.09 −0.22* 0.18 −4.59 −0.22*-*-* 0.17 −1.25 −0.13 2006, 2009, 2010, 2013,-,-,1984,1986, 1987, 1988, 1990, 1994, 1996, 1982,-, 1985, 1992, 1994, 1996, 2008, 2009, 2010, 2015, 2016, 2017, 2018,-,1986, 1993, 1994, 2000, 2012, 12 13 14 - −2.01 −- 1982, 1983, 1988,-, 2002, 2007, 2008, 2019 - −1.27 −-* 15 0.22 −2.10 −0.17 0.19 −1.68 −0.15 16 17 -* 0.23 −1.58 −0.18 - -* 0.25 −4.74 −0.27*-* 0.22 −4.09 −0.22*- −3.56 −0.24*-* 1984, 1985, 1993, 1994, 1998, 2008, 2009, 2010, 2015, 2016,-, 1985, 1992, 1994, 1997, 2010, 2014, 2016, 2017, 2019,-, 1983, 1988, 1989,-,- -* 0.20 −0.93 −-* 0.14 −3.74 −0.24*-* 0.13 −3.87 −- −2.39 −0.22*-* Turning Points * means significant trends existed., ACF is autocorrelation, PSS is the pre-whitened Sen’s Slope from MMK tests 118 Page 8 of 16 B. F. Sasanya et al. Fig. 2  Mann-Kendall trends for inter-annual rainfall in Chad River Basin, Nigeria Fig. 3  Mann-Kendall trends for inter-seasonal rainfall in Chad River Basin, Nigeria Conversely, regions within the Chad River Basin that experienced decreasing rainfall trends were typically located within the basin, away from its boundaries and distant from existing water bodies. This observation aligns with the research of Jajere et al. (2022), who reported erratic and consistently declining inter-annual rainfall trends in the north-eastern part of Nigeria. In concurrence with Jajere et al. (2022) findings, geographical locations 4, 6, 11, 13, 15, and 17 displayed statistically insignificant decreasing inter-annual rainfall amounts over a 41-year period (Figs. 2 and 3). The lack of statistical significance in these declining trends can be attributed to the substantial variability in both inter-annual and inter-seasonal rainfall (Bekele et al. 2017). Furthermore, it is worth noting that certain locations, specifically 7, 8, 19, 22, and 24, recorded statistically significant declines in rainfall amounts throughout the studied periods (Figs. 2 and 3). 3.2 Sequential mann kendall tests Figures 4 and 5 present the outcomes derived from the sequential Mann-Kendall (SMK) test, which was employed to identify abrupt alterations in the historical patterns of both annual and seasonal precipitation across a dataset comprising 25 locations. The detected abrupt shifts in the trends of seasonal and annual precipitation records revealed multiple turning points for specific locations (1, 4, 6, 11, 12, 14, 15, 17, 20, and 23), while others exhibited only a single turning point over the extensive 41-year examination period (refer to Table 2; Figs. 4 and 5 and Spatio-temporal analysis of rainfall over Chad River Basin, Nigeria supplementary Fig. 1 and supplementary Fig. 2). Notably, the turning points identified at most locations with multiple abrupt changes were not found to be statistically significant, except for location 1. It was noted that the majority of locations exhibiting multiple abrupt changes are situated in the western part of the basin in Borno State, Nigeria. In contrast, locations with a single abrupt change were found in Yobe State, located at the north eastern end of the basin. The occurrence of multiple abrupt changes in precipitation experienced in Borno State can be attributed to extreme weather events linked to climate change (Ilarri et al. 2022). Additionally, these multiples changes may be driven by anthropogenic factors such as land-use practices, deforestation, and alterations in hydrological Page 9 of 16 118 regimes, often resulting from irrigation projects or dam construction (Stephens et al. 2021; Kayitesi et al. 2022). The statistically significant abrupt changes in the positive precipitation trends at location 1 can be attributed to its proximity to Lake Chad, which has experienced a significant reduction in size, up to 10% of its original extent in the 1960s, due to factors such as climate change, population growth and suboptimal irrigation practices (Zhu et al. 2019). Figures 4 and 5 offer a comprehensive depiction of the turning points in historical precipitation data for inter-annual and inter-seasonal rainfall, respectively. For location 1, the turning points were statistically significant, with the first abrupt change in trends occurring in 2006 for inter-annual precipitation (Fig. 4) and in 2007 Fig. 4  Sequential Mann Kendall Progression and Regression Series of Annual Rainfall Intensities in Chad River Basin, Nigeria 118 Page 10 of 16 B. F. Sasanya et al. Fig. 4  (continued) for seasonal rainfall (Fig. 5). Conversely, the statistically significant abrupt shifts towards decreasing trends in interannual historical rainfall were observed at locations 7, 8, 19, 22, and 24 between the years 1999 and 2001 (Table 2; Fig. 4). In contrast, the turning points indicating significantly positive trends in historical inter-annual rainfall at locations 9, 10, 16, 18, 21, and 25 transpired between 2016 and 2018 (Fig. 4 and supplementary Fig. 1). Moreover, the inter-seasonal precipitation data displayed significant abrupt shifts towards increased trends at locations 9, 10, 14, 16, 18, 21, and 25 during the years 2016, 2015, 1998, 2005, 2015, 2018, and 2017, respectively (Table 2; Fig. 5 and supplementary Fig. 2). Notably, locations 14 and 16 exhibited multiple significant abrupt changes in their increasing inter-seasonal rainfall trends. Conversely, only inter-seasonal rainfall at locations 19 and 24 displayed significantly decreasing trends during the years 2011 and 1999, respectively. The computed percentage changes in inter-annual rainfall revealed that alterations in 24 out of 25 locations were less than 10%, while for inter-seasonal data, all 25 locations demonstrated changes of less than 10%. The most substantial negative change was observed at location 20, amounting to −7.24, while the smallest change was registered at location 3, with a value of −1.36, within the annual series. In the inter-seasonal rainfall series, the highest negative percentage change occurred at location 20, totalling − 7.29, while the lowest change was noted at location 13, amounting to −1.93. 4 Discussions This research endeavour delved into the spatio-temporal attributes and pivotal moments associated with annual and seasonal precipitation patterns within the Chad River Basin, situated in Nigeria, utilizing data collected from 25 distinct geographical locations. A noteworthy finding of this study was the presence of substantial serial correlation within portions of the annual and seasonal precipitation datasets. These findings closely align with those reported by Wang et al. (2013) and Tigabu et al. (2020). Wang et al. (2013) undertook a comprehensive analysis of precipitation records spanning 12 distinct time intervals -) in Shouguang city, Shandong, China, employing the autocorrelation function as a key analytical tool. Their investigation revealed pronounced autocorrelation patterns within the seasonal precipitation time series. Moreover, as outlined by Tigabu et al. (2020), the temporal progression of rainfall within the Lake Spatio-temporal analysis of rainfall over Chad River Basin, Nigeria Tana Basin in Ethiopia exhibited statistically significant autocorrelation coefficients, indicative of non-random distribution and linear interdependence over specific temporal intervals. The underlying factors contributing to these pronounced autocorrelation patterns were ascribed to prevailing meteorological conditions and substantial water vapour movement within the atmosphere (Tigabu et al. 2020). This study’s investigation further substantiates the notion that distinct segments within the basin exhibit varying rainfall characteristics, which can be categorized into four Page 11 of 16 118 fundamental groups. These categories encompass regions displaying statistically significant upward trends in rainfall, characterized by either a single or multiple turning points; areas with rainfall trends that do not display statistical significance but exhibit zero or multiple turning points; locations marked by substantial declines in rainfall with one turning point; and locations where the reduction in rainfall lacks statistical significance and features either one or multiple turning points. These fluctuations in rainfall characteristics can be attributed to a confluence of factors, including Fig. 5  Sequential Mann Kendall Progression and Regression Series of Seasonal Rainfall Intensities in Chad River Basin, Nigeria 118 Page 12 of 16 B. F. Sasanya et al. Fig. 5  (continued) climate change, disparities in geographical attributes such as topography, elevation above sea level, climate attributes, temperature increases leading to uneven precipitation patterns, encroachment of desertification, escalating atmospheric pollution (Rao et al. 2001; Gupta et al. 2005; Ramanathan et al. 2005; Sharma and Saha 2017), and variations in seasonal climatic patterns (Tigabu et al. 2020). The notable increase in rainfall trends observed at points 1, 9, 10, 16, 18, and 25 can be attributed to their proximity to Lake Chad (13.10°N and 14.45°E) (Jedwab et al. 2022; Fougou and Lemoalle 2022). Similar patterns of statistically significant rainfall increase were documented by Bekele et al. (2017), Weldegerima et al. (2018), and Alemu and Bawoke (2020) in river basins across Ethiopia during the major rainy seasons. Moreover, Sridhar and Raviraj (2017) reported predominantly rising rainfall trends in the Amaravathi River Basin, India, specifically during the North-East Monsoon, employing the Mann Kendall and Sen’s slope estimator. The findings of both increasing and decreasing rainfall trends in different areas within the studied basins were further supported by Deka (2021). Conversely, Deka (2021) examined the long-term rainfall trend in Cherrapunji, Meghalaya, India, utilizing Sen’s estimator and the Mann Kendall Z test of significance. The declining patterns in rainfall can be attributed to several factors, including a notable increase in dry spells between 1980 and 1990 (Sarr 2012), the Sahel drought spanning from 1970 to 1980 (Dong and Sutton 2015; Nkiaka et al. 2017), and the influence of climate change driven by human activities and land degradation (Epule et al. 2014). Similar declines in rainfall trends were also documented by Khavse et al. (2015) and Akhoury and Avishek (2019). Khavse et al. (2015) observed diminishing trends in mean monthly rainfall for the months of February, March, May, August, September, October, and November when investigating temperature and rainfall patterns in Raipur district, Chhattisgarh, India. The most significant reduction in total mean monthly rainfall was recorded in August, with a decrease of 1.439 mm per year between 1971 and 2013. Akhoury and Avishek (2019) conducted a statistical analysis of rainfall in Indian sub-divisions and explored the Page 13 of 16 Spatio-temporal analysis of rainfall over Chad River Basin, Nigeria relationship between rainfall data and the Southern Oscillation Index. Their findings revealed a substantial downward trend in rainfall between July and October from 1949 to 2016. In a related study, Sharma and Saha (2017) examined rainfall trends in the Damodar River Basin, India. They have utilized various analytical tools, including the lag-1 autocorrelation coefficient to identify serial correlation, nonparametric MK and MMK tests to assess trends, and the SMK test to pinpoint potential turning points. Their results indicated significant decreases in both annual and seasonal precipitation across a significant portion of the basin, with the most pronounced declines occurring in the north-western region and the mildest reductions observed in the northeastern region. The SMK test identified significant and nonsignificant annual and seasonal rainfall trends in various segments of the basin. 5 Conclusion This research sought to illuminate critical junctures within the time series of seasonal and annual precipitation data, thereby providing insights into abrupt shifts in trends across various regions within the Lake Chad River Basin within Nigeria. Our findings unveiled significant revelations regarding the erratic precipitation patterns and trends within the basin. The irregular rainfall patterns were observed in different parts of the Chad River Basin, Nigeria and these patterns can be attributed to climate change. This pattern may also have contributed to the reported diminution of the Lake Chad dimensions, a development which has attracted global attention. Regions proximate to the basin’s boundaries witnessed substantial increases in rainfall, because of the presence of some other water bodies along those boundaries. However, locations farther from the basin boundaries experienced declines in precipitation trends. The escalating rainfall trends in the regions closer to water bodies are indicative of elevated evaporation rates resulting from rising temperatures which is also attributable to climate change and global warming. The compounded effects of these phenomena may result to adverse consequences for water resource management and other vital socio-economic sectors heavily reliant on water resources, particularly agriculture. Given these findings, it is imperative for both regional and international stakeholders to acknowledge the gravity of the situation and take concerted action. Addressing the challenges posed by shifting rainfall patterns necessitates a multifaceted approach encompassing climate mitigation, adaptive water resource management, and sustainable environmental policies. Implementing sound water conservation techniques, such as promoting the adoption of drip irrigation 118 over surface irrigation systems, can also prove beneficial. The preservation of water as a critical natural resource is not merely a regional concern but a matter of global significance, given its far-reaching ecological, agricultural, socioeconomic and geopolitical implications. To advance this study, it is recommended that pertinent drought and flood monitoring indices be utilized to track and predict drought and flood occurrences in the Lake Chad River Basin. Furthermore, Africa, as a continent, should prioritise the installation of gauging stations across its catchments and river basins to enable the in situ measurement of rainfall. This would reduce the reliance on satellite data, which, although useful, was the sole data source utilised in this study. Supplementary information The online version contains supplementary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 00704-0​ 24-0​ 5338-2. Acknowledgements Not Applicable. Author and co‑author's contribution SASANYA, Blessing Funmbi: Conceptualization, data collection, data analysis and original writeup. ADESOGAN, Sunday Olufemi: Data curation, final review before submission. ADEMOLA, Akeem Abiodun: Methodology, writing, review and editing. Funding sources Not applicable. Data availability The raw data are available on reasonable request from the authors. Declarations All authors have read, understood, and complied as applicable with the statement on "Ethical responsibilities of Authors" as found in the Instructions for Authors and are aware that with minor exceptions, no changes can be made to authorship once the paper is submitted. Ethical approval Not Applicable. Consent to participate Not Applicable. Consent to publish The author and Co-author agreed to publish this version of the research article. Conflict of interest The Author and Co-author declare no conflict of interest in this research article. Informed consent on Studies with Human and Animals Subjects Not applicable. References Abe AO, Adeniji QA, Rabiu JA, Adegboyega O, Raheem IO, Rasaki MG, Sada SM, Fidelis LF (2022) Statistical analysis and forecasting of rainfall patterns and trends in Gombe North-Eastern Nigeria. Iraqi J Phys (IJP) 20(2):64–77. https://​doi.​org/​10.​30723/​ ijp.​v20i2.​989 Aditya Satrio CB, Darmawan W, Nadia BU, Hanafiah N (2021) Time series analysis and forecasting of coronavirus disease in Indonesia 118 Page 14 of 16 using ARIMA model and PROPHET. Procedia Comput Sci 179:524–532.https://​doi.​org/​10.​1016/j.​procs.​2021.​01.​036 Akhoury G, Avishek K (2019) Statistical analysis of Indian rainfall and its relationship with the Southern Oscillation Index. Arab J Geosci 12:255 https://​doi.​org/​10.​1007/​s12517-​019-​4415-z Alashan S (2020) Combination of modified Mann-Kendall method and Şen innovative trend analysis. Eng Rep. https://​doi.​org/​10.​1002/​ eng2.​12131 Alemu MM, Bawoke GT (2020) Analysis of spatial variability and temporal trends of rainfall in Amhara region, Ethiopia. J Water Clim Change 11(4):- AlSubih M, Kumari M, Mallick J, Ramakrishnan R, Islam S, Singh CK (2021) Time series trend analysis of rainfall in the last five decades and its quantification in the Aseer Region of Saudi Arabia. Arab J Geosci 14(6). https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 12517-0​ 21-0​ 6935-5 Awe O, Okeyinka A, Fatokun JO (2020) An alternative algorithm for ARIMA model selection. 2020 International Conference in Mathematics, Computer Engineering and Computer Science (ICMCECS).https://​doi.​org/​10.​1109/​icmce​cs476​90.​2020.​24 Bekele D, Alamirew T, Kebede A, Zeleke G, Melese AM (2017) Analysis of rainfall trend and variability for agricultural water management in awash river Basin, Ethiopia. J Water Clim Change 8(1). https://​doi.​org/​10.​2166/​wcc.​2016.​044 Box GEP, Jenkins GM (2013) Time series analysis, forecasting and control. In: Mills TC (ed) A Very British Affair. Palgrave Advanced Texts in Econometrics. Palgrave Macmillan, London. https://​doi.​org/​10.​1057/​97811​37291​264_6 Box GEP, Jenkins GM, Reinsel G (2008) Time series analysis, forecasting and control. 4th edn. John Wiley and Sons Incorporation, New Jersey Brockwell PJ, Davis RA (2002) Introduction to time series and forecasting. In: Springer Texts in Statistics. Springer International Publisher, New York. https://​doi.​org/​10.​1007/0-​387-​21657-X_8 Caloiero T, Coscarelli R, Ferrari E, Mancini M (2011) Trend detection of annual and seasonal rainfall in Calabria (Southern Italy). Int J Climatol 31(1):44. https://​doi.​org/​10.​1002/​joc.​2055 Chatfield C, Xing H (2019) The analysis of time series. In: The Analysis of Time Series. https://​doi.​org/​10.​1201/​97813​51259​446 Creese A, Washington R, Munday C (2019) The plausibility of September-November Congo Basin rainfall change in coupled climate models. J Geophys Research: Atmos. https://​doi.​org/​10.​1029/​ 2018j​d0298​47 Croitoru AE, Chiotoroiu BC, Ivanova Todorova V, Toricǎ V (2013) Changes in precipitation extremes on the Black Sea Western Coast. Glob Planet Change 102. https://​doi.​org/​10.​1016/j.​glopl​ acha.​2013.​01.​004 Da Silva RM, Santos CAG, da Costa Silva JFCB, Silva AM, Brasil Neto RM (2020) Spatial distribution and estimation of rainfall trends and erosivity in the Epitácio Pessoa reservoir catchment, Paraíba, Brazil. Nat Hazards 102(3). https://​doi.​org/​10.​1007/​ s11069-​020-​03926-9 Dagnachew M, Kebede A, Moges A, Abebe A (2020) Effects of Climate Variability on Normalized Difference Vegetation Index (NDVI) in the Gojeb River Catchment, Omo-Gibe Basin, Ethiopia. Adv Meteorol 2020:1–16. https://d​ oi.o​ rg/1​ 0.1​ 155/2​ 020/8​ 2632​ 46 Damodar N, Gujarati, DCP (2009) Basic econometrics. 5th edn. Tata McGraw Hill Education Publisher, New York Deka S (2021) Statistical analysis of long-term rainfall trends in Cherrapunji, Meghalaya, India. J Appl Nat Sci 13(1):170–177. https://​ doi.​org/​10.​31018/​jans.​v13i1.​2442 Dong B, Sutton R (2015) The dominant role of greenhouse-gas forcing in the recovery of Sahel rainfall. Nat Clim Change 5(8). https://​ doi.​org/​10.​1038/​nclim​ate26​64 B. F. Sasanya et al. Durbin J, Watson GS (1950) Testing for serial correlation in least squares regression: I. Biometrika 37(3/4):409. https://​doi.​org/​10.​ 2307/​23323​91 Enders W (2014) Stationary time-series models. In: Applied Econometric Time Series, 4th edn. John Wiley and Sons Incorporation, New Jersey Epule ET, Peng C, Lepage L, Chen Z (2014) The causes, effects and challenges of Sahelian droughts: a critical review. Reg Envriron Chang 14(1). https://​doi.​org/​10.​1007/​s10113-​013-​0473-z Espinoza-Dávalos GE, Arctur DK, Teng W, Maidment DR, GarcíaMartí I, Comair G (2015) Studying soil moisture at a national level through statistical analysis of NASA NLDAS data. J Hydroinformatics 18(2):277–287. https://​d oi.​o rg/​1 0.​2 166/​ hydro.​2015.​23 FAO Fisheries and Aquaculture (2012) Hydrology of the Lake Chad Basin. In: De Young C, Sheridan S, Davies S, Hjort AA (eds) Climate change implications for fishing communities in the Lake Chad Basin: FAO/Lake Chad Basin Commission Workshop Proceedings 2011(25):2–6 Fougou HK, Lemoalle J (2022) Variability of Lake Chad. In: book: Congo Basin Hydrology, Climate, and Biogeochemistry https://​ doi.​org/​10.​1002/​97811​19657​002.​ch26 Goni IB, Taylor RG, Favreau G, Shamsudduha M, Nazoumou Y, Ngounou Ngatcha B (2021) Groundwater recharge from heavy rainfall in the southwestern Lake Chad Basin: evidence from isotopic observations. Hydrol Sci J 66(8):-. https://​doi.​org/​10.​ 1080/​02626​667.​2021.​19376​30 Groleau A, Mailhot A, Talbot G (2007) Trend analysis of winter rainfall over outhern Québec and New Brunswick (Canada). Atmos Ocean 45(3):153. https://​doi.​org/​10.​3137/​ao.​450303 Grose MR, Syktus J, Thatcher M, Evans JP, Ji F, Rafter T, Remenyi T (2019) The role of topography on projected rainfall change in mid-latitude mountain regions. Clim Dyn. https://d​ oi.o​ rg/1​ 0.1​ 007/​ s00382-​019-​04736-x Gupta A, Thapliyal P, Pal P, Joshi P (2005) Impact of deforestation on Indian monsoon—a GCMsensitivity study. J Ind Geophys Union 9:97–104 Hamed KH, Rao RA (1998) A modified Mann-Kendall trend test for auto-correlated data. J Hydrol 204(1–4):182–196. https://​doi.​org/​ 10.​1016/​s0022-​1694(97)​00125-x Hamilton JD (1994) Time series analysis, 2nd edn. Princeton University Press Harka AE, Jilo NB, Behulu F (2021) Spatial-temporal rainfall trend and variability assessment in the Upper Wabe Shebelle River Basin, Ethiopia: application of innovative trend analysis method. J Hydrology: Reg Stud 37:100915. https://d​ oi.o​ rg/1​ 0.1​ 016/J.E ​ JRH.​ 2021.​100915 Hirsch RM, Slack JR, Smith RA (1982) Techniques of trend analysis for monthly water quality data. Water Resour Res 18(1):107. https://​doi.​org/​10.​1029/​WR018​i001p​00107 Ibrahim UA, Dan’azumi S, Bdliya HH, Bunu Z, Chiroma MJ (2022) Comparison of WEAP and SWAT models for streamflow prediction in the Hadejia-Nguru wetlands, Nigeria. Model Earth Syst Environ 8(4):- Ilarri M, Souza AT, Dias E, Antunes C (2022) Influence of climate change and extreme weather events on an estuarine fish community. Sci Total Environ 827:154190 Itiowe T, Hassan SM, Agidi VA (2019) Analysis of rainfall trends and patterns in Abuja, Nigeria. Curr J Appl Sci Technol. https://​doi.​ org/​10.​9734/​cjast/​2019/​v34i4​30139 Jajere AA, Sawa AB, Kibon UA, Muhammed BU, Babagana-Kyari M (2022) Spatio-temporal variability analysis of Rainfall in Kumadugu-Yobe River Basin, Nigeria. Geosfera Indonesia 7(1). https://​ doi.​org/​10.​19184/​geosi.​v7i1.​24302 Jedwab R, Haslop F, Rodr C (2022) The Real effects of climate change in the poorest countries: evidence from the permanent shrinking Spatio-temporal analysis of rainfall over Chad River Basin, Nigeria of Lake Chad, Economics. Environ Sci. https://​doi.​org/​10.​1596/​ 1813-​9450-​10561 Kayitesi NM, Guzha AC, Mariethoz G (2022) Impacts of land use land cover change and climate change on river hydro-morphology-a review of research studies in tropical regions. J Hydrol 615:128702 Kendall MG (1975) Rank correlation methods, 4th edn. Charles Griffin Publisher, San Francisco Khavse R, Deshmukh R, Manikandan N, Chaudhary JL, Kaushik D (2015) Statistical analysis of temperature and Rainfall trend in Raipur District of Chhattisgarh. Curr World Environ 10(1):305–312 Kibria BMG, Lukman AF (2020) A new ridge-type estimator for the linear regression model: simulations and applications. Scientifica 2020:1–16. https://​doi.​org/​10.​1155/​2020/​97583​78 Li F, Chen J, Liu Y, Xu P, Sun H, Engel BA, Wang S (2019) Assessment of the impacts of land use/cover change and rainfall change on Surface Runoff in China. Sustainability 11(13):3535. https://​doi.​org/​10.​3390/​su111​33535 Mann HB (1945) Nonparametric tests against Trend. Econometrica 13(3):245. https://​doi.​org/​10.​2307/​19071​87 Marques R, Augusto C, Santos G, Flávio J, Braga C, Silva AM, Moura R, Neto B (2020) Spatial distribution and estimation of rainfall trends and erosivity in the Epitácio Pessoa reservoir catchment. Nat Hazards 102(3):829–849. https://​d oi.​o rg/​1 0.​ 1007/​s11069-​020-​03926-9 Nasri M, Modarres R (2009) Dry spell trend analysis of Isfahan Province, Iran. Int J Climatol 29(10). https://​doi.​org/​10.​1002/​ joc.​1805 Nkiaka E, Nawaz NR, Lovett JC (2017) Analysis of rainfall variability in the Logone catchment, Lake Chad basin. Int J Climatol 37(9):3553. https://​doi.​org/​10.​1002/​joc.​4936 Odjugo PA (2006) An analysis of rainfall patterns in Nigeria. Global J Environ Sci 4(2). https://​doi.​org/​10.​4314/​gjes.​v4i2.​2455 Patakamuri SK, Das B (2022) Innovative trend analysis and timeseries change point analysis. R package version 1.2. https://​ CRAN.R-​proje​ct.​org/​packa​ge=​trend​change. Accessed 14 Aug 2023 Patakamuri SK, O’Brien N (2021) Modified versions of Mann Kendall and Spearman’s Rho trend tests. R package version 1.6.https://​CRAN.R-​proje​ct.​org/​packa​ge=​modif​i edm. Accessed 12 Sep 2023 Pham-Duc B, Sylvestre F, Papa F, Frappart F, Bouchez C, Crétaux J-F (2020) The Lake Chad hydrology under current climate change. Sci Rep 10(1):5498. https://d​ oi.o​ rg/1​ 0.1​ 038/s​ 41598-0​ 20-6​ 2417-w Pohlert T (2020) Non-parametric trend tests and change-point detection. R package version 1.1.4. https://​CRAN.R-​proje​ct.​org/​packa​ ge=​trend​1–18. Accessed 12 July 2023 Ramanathan Vet al et al (2005) Atmospheric brown clouds: impacts on south Asian climate and hydrological cycle. Proc Natl Acad Sci U S A 102:- Rao DB, Naidu C, Da BS (2001) Trends and fluctuations of the cyclonic systems over North Indian Ocean. Mausam 52:37–46 Rowell DP, Chadwick R (2018) Causes of the uncertainty in projections of tropical terrestrial rainfall change: East Africa. Journal of Climate 31(15):-. https://​doi.​org/​10.​1175/​jcli-d-​17-​0830.1 Sa’adi Z, Shahid S, Ismail T, Chung E-S, Wang X-J (2017) Trends analysis of rainfall and rainfall extremes in Sarawak, Malaysia using modified Mann–Kendall test. Meteorol Atmos Phys. https://​ doi.​org/​10.​1007/​s00703-​017-​0564-3 Sarr B (2012) Present and future climate change in the semi-arid region of West Africa: a crucial input for practical adaptation in agriculture. Atmospheric Sci Lett 13(2). https://​doi.​org/​10.​1002/​asl.​368 Sasanya BF, Awodutire PO, Ufuoma OG (2024) Modelling rainfall in selected agricultural hubs in Nigeria: a comparative probability Page 15 of 16 118 distributions study. Theoret Appl Climatol 1–14:1. https://d​ oi.o​ rg/​ 10.​1007/​s00704-​024-​04832-x Sharma S, Saha AK (2017) Statistical analysis of rainfall trends over Damodar River Basin, India. https:// ​ d oi. ​ o rg/ ​ 1 0. ​ 1 007/​ s12517-​017-​3096-8 Sharma P, Singh AK, Agrawal B, Sharma A (2020) Correlation between weather and COVID – 19 pandemic in India: an empirical investigation. J Public Affairs. https://d​ oi.o​ rg/1​ 0.1​ 002/p​ a.2​ 222 Sneyers R (1990) On the statistical analysis of a series of observations. World Meteorological Organization Sridhar SI, Raviraj A (2017) Statistical trend analysis of rainfall in amaravathi River Basin using mann-kendall test. Curr World Environ 12(1):89–96.https://​doi.​org/​10.​12944/​CWE.​12.1.​11 Srivastava PK, Mehta A, Gupta M, Singh SK, Islam T (2015) Assessing the impact of climate change on Mundra mangrove forest ecosystem, Gulf of Kutch, western coast of India: a synergistic evaluation using remote sensing. Theoret Appl Climatol 120:3–4. https://​doi.​org/​10.​1007/​s00704-​014-​1206-z Stephens CM, Lall U, Johnson FM, Marshall LA (2021) Landscape changes and their hydrologic effects: interactions and feedbacks across scales. Earth Sci Rev 212:103466 Terêncio DPS, Sanches Fernandes LF, Cortes RMV, Moura JP, Pacheco FAL (2018) Rainwater harvesting in catchments for agro-forestry uses: a study focused on the balance between sustainability values and storage capacity. Sci Total Environ 613–614. https://​doi.​org/​ 10.​1016/j.​scito​tenv.​2017.​09.​198 Tigabu TB, Wagner PD, Hörmann G, Fohrer N (2020) Modelling the spatio-temporal flow dynamics of groundwater-surface water interactions of the Lake Tana Basin, Upper Blue Nile, Ethiopia. Hydrology Research 51(6):-. https://​doi.​org/​10.​2166/​ nh.​2020.​046 UNFCCC (2007) Climate Change: impacts, vulnerabilities and adaptation in developing countries. United Nations Framew Convention Clim Change 68:68. https://​doi.​org/​10.​1029/​2005J​D0062​89 Vijayvargia A, Sharma KC, Bhakar R (2023) VARMA model parameterization using MLLE approach for intraday wind power forecasting application. Int J Numer Model Electron Networks Devices Fields 36(6):e3119 Waha K, Clarke J, Dayal K, Freund M, Heady C, Parisi I, Vogel E (2022) Past and future rainfall changes in the Australian midlatitudes and implications for agriculture. Clim Change 170:3–4. https://​doi.​org/​10.​1007/​s10584-​021-​03301-y Wang S, Feng J, Liu G (2013) Application of seasonal time series model in the precipitation forecast. Math Comput Modelling 58:677–683. https://​doi.​org/​10.​1016/j.​mcm.​2011.​10.​034 Wang F, Shao W, Yu H, Kan G, He X, Zhang D, Wang G (2020) Reevaluation of the power of the Mann-Kendall test for detecting monotonic trends in hydrometeorological time series. Front Earth Sci 8:14. https://​doi.​org/​10.​3389/​feart.​2020.​00014 Weldegerima TM, Zeleke TT, Birhanu BS, Zaitchik BF, Fetene ZA (2018) Analysis of Rainfall trends and its relationship with SST signals in the Lake Tana Basin, Ethiopia. Adv Meteorol 2018. https://​doi.​org/​10.​1155/​2018/​58690​10 Woodall WH, Rakovich G, Steiner SH (2020) An overview and critique of the use of cumulative sum methods with surgical learning curve data. Stat Med 40(6):-. https://​doi.​org/​10.​1002/​ sim.​8847 Xu W, Peng H, Zeng X, Zhou F, Tian X, Peng X (2019) A hybrid modelling method for time series forecasting based on a linear regression model and deep learning. Appl Intell. https://​doi.​org/​ 10.​1007/​s10489-​019-​01426-3 Yue S, Pilon P, Phinney B, Cavadias G (2002) The influence of autocorrelation on the ability to detect trends in hydrological series. Hydrol Process 16(9). https://​doi.​org/​10.​1002/​hyp.​1095 Yue S, Wang CY (2004) The Mann-Kendall test modified by effective sample size to detect trends in serially correlated hydrological 118 Page 16 of 16 series. Water Resour Manage 18(3). https://​doi.​org/​10.​1023/B:​ WARM.​00000​43140.​61082.​60 Zhu W, Jia S, Lall U, Cao Q, Mahmood R (2019) The relative contribution of climate variability and human activities on the water loss of the Chari/Logone River discharge into Lake Chad: a conceptual and statistical approach. J Hydrol 569:519–531. https://d​ oi.o​ rg/1​ 0.​ 1016/j.​jhydr​ol.​2018.​12.​015 Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. B. F. Sasanya et al. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.