Spatiotemporal Patterns of Extreme Precipitation Events over Jhelum River Basin

Abstract

Climate change has a wide range of impacts on Pakistan’s water resources system, followed by increased magnitude, frequency and intensity of extreme events. Therefore, investigating the variations in extreme climatic events is vital in planning and developing agriculture and water conservation projects. Hence, the present study investigated the spatiotemporal variability of extreme precipitation indices (EPIs) over the Jhelum River basin (JRB) using the Mann-Kendall (MK) and innovative trend analysis (ITA) methods. In addition, variations in the event-based extreme precipitation (EEPs) and time distribution patterns (TDPs) were analyzed because EPIs considered the daily precipitation values only and the combined effect of the extreme events that last for more than 1 day were often neglected. The results show that the fixed and station-related EPIs, i.e., R10, R20, R25, CWD, Prcpt and R95p, exhibited significant decreasing trends at 06, 05, 04, 04, 07 and 05 stations, whereas an increasing trend was evident for consecutive dry days (CDD) at nine stations with a magnitude of 0.33 days/year, indicating a rise in the drought events compared to floods. Moreover, precipitation events with an amount of 40–290 mm, duration of 4–7 days and concentration ratio of 0.7–0.95 were found to be dominant across JRB. Therefore, an increase in the CDD and a decrease in the consecutive wet days (CWD) may change the magnitude, intensity and frequency of extreme precipitation events and consequently pressure the availability of water resources. Therefore, the findings of this study may help understand the precipitation variability over JRB and play a pivotal role in future development projects.

1. Introduction

Global climate has recently led to drastic variations in precipitation and temperature patterns [1]. Various emission scenarios predict that the average global temperature will rise from 1.8 to 4 °C in the coming decades of the 21st century, a higher increase of 0.74 °C over the previous century [2]. Anthropogenic activities and the environment are strongly affected by extreme precipitation events such as floods and droughts [3]. The economic losses from extreme events in the past 60 years are more than any other natural disaster and significantly impact climate-related sectors, including agriculture and food security [4]. Furthermore, several studies indicate that Pakistan will face more frequent and intense extreme events such as heat waves, floods, and droughts [5].

Pakistan is located in an arid to semi-arid climatic region and its agriculture-based economy is heavily dependent on the water resources from the upper Indus River basin [6]. However, water resource managers and decision-makers face severe challenges in resolving water-related issues as the country’s water resources are highly prone to climate change [7]. Floods and droughts are common natural disasters in Pakistan and place the country in regions with moderate to severe threats of extreme events [8]. Moreover, severe water shortages and floods are predicted to be more intense and frequent due to climate change and large spatial variability [9]. In addition, other extreme climate-related disasters such as heat waves, landslides and excessive sediment flows are also becoming more frequent and posing widespread concerns for the socioeconomic development [10,11]. Therefore, predicting the variation of extreme events for investigating the impacts of climate change plays a vital role in devising strategies and policies for sustainable ecological development [12].

Several studies used nonparametric methods to examine the variations in the extreme precipitation indices (EPIs) because of their insensitivity to outliers and data normality [13]. For instance, the Mann-Kendall (MK) test was used to assess the spatiotemporal changes in the EPIs and detected a declining tendency in northeast Bangladesh [14]. On the contrary, an increasing trend was detected in central Asia for the six EPIs, whereas only consecutive dry days (CDD) exhibited decreasing tendency [15]. In another study, Bhatti et al. [16] found an inverse relationship in different EPIs using the nonparametric methods, which indicated a decrease in Pakistan. Xia et al. [17] detected an increase in the CDD and a decrease in the number of CWD using the MK method test over Shandong of China.

Analyzing the preceding and succeeding precipitation (PSP) is considered vital in extreme precipitation analysis because they comprise three stages, i.e., start, development and decline, and may last for several days [18]. Several studies to analyze the extreme precipitation characteristics, i.e., duration, amount and frequency, have been conducted to differentiate the extreme precipitation events from the PSP process [19]. Significant flooding may occur due to extreme precipitation, whereas PSP might accelerate the flooding process. For instance, three precipitation events with magnitudes of 40, 50 and 45 mm hit the Yangtze and Huai river basins on 8th, 13th and 14th of June 1991, respectively [20]. Both incidents could not be classified as extreme precipitation events since the precipitation occurrence of less than 50 mm reaches the threshold value and the remaining events were below the standard threshold. However, the precipitation outbreaks on 8 and 14 June were comparable to those on 13 June, and their combined effect could result in significant floods. Hence, event-based extreme precipitation (EEP) can be better evaluated by considering preceding and succeeding events.

Several studies indicated that Pakistan is experiencing widespread variations in the extreme precipitation patterns throughout the country and explicitly in the northern regions [21]. For instance, Ali et al. [22] detected a gradual spatial increase in the precipitation from the south to north with an aerial average of 455 mm, whereas there was a significant decrease in the long-term temporal precipitation for the Sindh province of Pakistan. Rafiq et al. [23] found significant declining trends in the 1-month standard precipitation index (SPI) and increasing trends in the 12-month SPI using the nonparametric methods over the Baluchistan province of Pakistan. Saddique et al. [24] detected a significant increasing trend in the high values of extreme temperature and opposite trends for the low values over a transboundary Jhelum River basin (JRB) shared between Pakistan and India. However, only monotonic trends can be detected using traditional trend analysis methods, making it impossible to find hidden trends in the time series’ low, median, and high-value groups of the timeseries.

Consequently, a novel trend analysis (ITA) technique was proposed [25,26] to detect the hidden trends in timeseries. For instance, Dong et al. [27] investigated the temperature trends in low, median and high-value clusters using the ITA method in the Jinsha river basin of China and found more obvious increasing trends in high-temperature clusters. Danbali [28] detected significant variations in the precipitation, dry days length and drought severity risk over Turkey by employing the ITA method. Alemu and Dioha [29] used the MK test with and without TFPW and variance correlation approaches to compare their results with the ITA method and found an increasing trend in high-temperature values, which could not be detected by using the MK test. In addition, several other studies used the ITA method to investigate the timeseries data variations and confirmed its enhanced capability to detect hidden trends [30,31]. However, to the best of the authors’ knowledge, variations in extreme precipitation indices, including EEPs and TDPs, have never been analyzed over JRB to detect the hidden trends using the ITA method, which are often neglected by the traditional trend analysis methods. Moreover, due to the limited application of the ITA method and contradictory reports on the traditional trend analysis methods, it is considered necessary to evaluate its performance over a transboundary river basin with diverse climatic conditions.

Therefore, the current study aims to investigate the extreme precipitation indices (EPIs) and evaluate their variations using the innovative trend analysis (ITA) method. The reliability of the ITA method was tested by comparing its results with the Mann-Kendall test. Moreover, the present study also uses the concept of PSP to analyze extreme precipitation events and examine the TDPs along with the EEPs over the transboundary Jhelum River basin.

2. Study Area

The Jhelum River basin (JRB) originates from the western Himalayas and flows through the Jhelum, Poonch, Kanshi, Neelum and Kunhar rivers to feed the Mangla reservoir. The elevation of the Mangla watershed varies between 6285 m and 216 m from north to south, with a catchment area of 33,490 km². The catchment area is shared between Pakistan (45%) and India (55%), with population density varying from 350 to 1000 per km² in the mountainous and agricultural regions, respectively. The watershed contains large icecaps and snow-fed areas, ultimately contributing to the Mangla reservoir. More than 600,000 km² of agricultural lands depend on this reservoir for irrigation to meet the food and fiber requirements of the country. In addition, the dam generates 1000 MW of hydropower as a by-product to fulfill the electricity demand and supply gap and is considered the lifeline of Pakistan’s agriculture-based economy. However, an increase in the frequency and intensity of extreme events, i.e., floods and droughts, have been witnessed in recent years and are predicted to be more severe in the future due to climate change [11]. Therefore, predicting climate extremes is critical for analyzing the effects of climate change on water resource availability [32].

The precipitation data over the JRB are distributed among different organizations: Pakistan Meteorological Department (PMD), Water and Power Development Authority (WAPDA) and Indian Meteorological Department (IMD). The daily precipitation data from 1971 to 2017 of 12 stations were carefully chosen based on homogeneity, extent and completeness. The geographical location of different meteorological stations and mean monthly variations in the precipitation timeseries over the Mangla watershed are presented in Figure 1 and Figure 2, respectively. The maximum precipitation was observed during July and August due to the south Asian summer monsoon climatic regime. However, flood events are often observed in these regions due to the limited storage capacity of the Mangla reservoir and vice versa for the dry season. Therefore, spatiotemporal variations in extreme events should be investigated to develop management strategies in this climate change hotspot region [33].

3. Materials and Methods

The overall research framework is presented in Figure 3. The daily precipitation timeseries data at various climatic stations over JRB were collected from 1971 to 2017. Daily precipitation data of only 12 stations were carefully selected based on the records’ extent, homogeneity, and completeness to ensure the data quality. The selected stations’ data range from 34 to 47 years. Precipitation timeseries data were tested for homogeneity before applying extreme value analysis. For this purpose, four tests, i.e., standard normal homogeneity test (SNHT), cumulative deviation (CD), Buishand’s range (BR) and Pettit’s tests, were employed to assess the homogeneity of the precipitation timeseries. Different types of precipitation indices were considered, i.e., indices based on duration, threshold, percentile, and absolute precipitation, and evaluated using RClimDex.

The MK and ITA methods were used for trend analysis to investigate the variability in the various EPIs. The ITA method developed by Sen [25,26] was employed for estimating variations in the precipitation timeseries because it does not require the data to be normally distributed and is independent of serial correlation. Moreover, their graphical representation can effectively detect high, median and low values variations. Finally, the EEP, designated as a precipitation event with at least one daily precipitation extreme, is considered a proceeding and succeeding (PSP) of extreme daily events and time distribution patterns (TDPs).

3.1. Homogeneity Tests

Homogeneity analyses of the hydrometeorological timeseries data are considered essential for historical trend analysis, water resource planning and management and climate change studies. Long-term homogeneous precipitation data are required for hydrometeorological decision-making as some non-climatic factors may influence the data homogeneity, leading to inconsistent trends, jumps and shifts. The sources of inhomogeneity in hydrometeorological timeseries data may include the relocation of the station, instrument change, instrument inaccuracy, and errors in calculation procedures [34]. Consequently, several methods for detecting homogeneity in timeseries have been suggested and employed in previous studies. The standard normal homogeneity (SNHT), cumulative deviation (CD), Buishand’s range (BR) and Pettitt tests are the most commonly used methods and for a homogenous timeseries, their critical values should be lower than 11.38 [35], 1.52, 1.78 [36] and 293 [37], respectively. The present study analyzed the homogeneity of precipitation data by employing the tests mentioned above at a 10% significance level.

3.2. Extreme Precipitation Indices (EPIs)

Various indices have been extensively used to investigate extreme precipitation events’ intensity, frequency and duration in previous studies worldwide [38,39]. The present study considered the ETCCDI-recommended eleven (11) key EPIs described in Table 1. These indices are classified based on the precipitation intensity, frequency and duration. A portion of the wide range of indices has been harmonized by integrating the datasets from different regions to demonstrate how extreme events have shifted worldwide [39].

These EPIs can be categorized into different subclasses based on the duration (CDD and CWD), threshold (R10, R20 and R25), percentile (R95p and R99p), absolute (Rx1d and Rx5d) and others (Prcpt and SDII) as defined in Table 1. The combined use of these indices enables the integrated assessment of long-term multi-decadal variation in the extreme precipitation events at spatial and temporal scales in any region of the world.

3.3. Statistical Methods

Trend analysis has often been used to detect variations in the hydrometeorological data to investigate whether the timeseries are upward, downward, or no-trend. For this purpose, several studies used parametric and nonparametric methods to detect the trends in hydrometeorological timeseries. The parametric methods are considered more potent for detecting trends; however, they require the data to be normally distributed, which is rarely true in the case of hydrometeorological timeseries analysis. Therefore, nonparametric tests have been generally used to identify trends in timeseries because they are independent of such limitations, i.e., normal distribution and data length, but data should be serially independent. The nonparametric statistical trend detection methods are discussed in subsequent sections.

3.3.1. Mann Kendall Test

MK is a nonparametric rank-based test [40,41] that is insensitive to outliers and does not require the data to be normally distributed [42]. Hence, this method has been widely used to identify variations in hydrometeorological timeseries in various studies [29,30,43,44]. The MK statistics, S, are given as:

(1)$S=\sum _{k=1}^{n-1}\sum _{j=k+1}^{n}sgn(Y_j-Y_k)$

(2)$sgn(Y_j-Y_k)=\left\{\begin{array}{ccc}if(Y_j-Y_k)<0;& then& -1\\ if(Y_j-Y_k)=0;& then& 0\\ if(Y_j-Y_k)>0;& then& 1\end{array}\right\}$

where Y_k and Y_j are the consecutive data values over time k and j, n denotes the number of data points, and sgn denotes the function that takes the values −1, 0 and 1 as of Y_j < Y_k, Y_j = Y_k, and Y_j > Y_k, respectively. In the hydrometeorological timeseries, a positive value of S indicates an increasing trend, whereas a negative value depicts a decreasing trend. The test probability (E) and variance (Var) can be calculated as follows:

(3)$E\left[S\right]=0$

(4)$Var(S)=\frac{n(n-1)(2n+5)-{\displaystyle {\sum }_{p=1}^q{t}_p(t_p-1)(2t_p+5)}}{18}$

The standardized test statistics (Z_MK) value is determined by using the following equation after calculating the variance Var(S) from Equation (4): where q denotes the tied groups, which represent observations of the same value but do not include the location of distinctive rank numbers, t_p denotes the number of data values in the pth group, and symbol (∑) denotes the sum of all tied groups. This sequence description can be ignored if the data contain no tied classes.

(5)$Z_{MK}=\{\begin{array}{ccc}\frac{S-1}{\sqrt{VAR(S)}},& if& S>0\\ 0,& if& S=0\\ \frac{S+1}{\sqrt{VAR(S)}},& if& S<0\end{array}$

This study calculates the trends in the precipitation timeseries data at 1%, 5%, and 10% significance.

3.3.2. Innovative Trend Analysis Method

The ITA method has been extensively used in combination with other trend analysis methods in many studies across the globe to investigate the variations in the hydrometeorological timeseries due to its advantages over other nonparametric approaches. The data are divided into two halves and organized in ascending order. Figure 4 displays the first half of the data along the X-axis and the second half along the Y-axis in a Cartesian coordinate system. The timeseries data in each half can be divided into subgroups to calculate the variations between low, median and high values. If all of the data values in a scatter diagram fall on the 45° (1:1) line, the hydrometeorological timeseries data exhibited no trend; furthermore, if the data points fall in the lower and upper triangular areas of the 1:1 graph, the trend is downward or upward, respectively [25]. The following equation calculates a trend’s magnitude in a data series [45].

(6)$D=\frac{1}{n}{\displaystyle \sum _{i=1}^n\frac{10\left(Y_j-Y_i\right)}{\mu }}$

where D is the trend statistic, n denotes the number of data values, Y_i and Y_j represent the data values in the first and second halves of the timeseries, respectively, and µ is the mean of the first subseries.

3.4. Definition of EEP

The first step in categorizing the EEPs from precipitation timeseries to determine the extreme events is establishing a threshold. The precipitation values of less than 1 mm are often missed and registered as no precipitation event because 1 mm is generally considered a threshold to categorize dry and wet days [46]. Hence, a precipitation event is characterized as daily if more than 1 mm of precipitation occurs consecutively for more than one day. The next step is to determine the magnitude and frequency of precipitation on a regular basis. According to earlier studies, the percentile-based technique has been frequently utilized to identify extreme precipitation events [47]. Furthermore, the 99th percentile has often been used to identify extremely heavy precipitation events [48].

Therefore, extreme daily precipitation is defined as precipitation that is greater than the 99th percentile among all the wet days. In the final step, an EEP can be calculated by combining the results of the first two steps with at least one precipitation extreme over the course of a precipitation event. The EEPs are separated by at least one-day intervals when there is less than 1 mm of precipitation. It is important to note that an EEP that lasts for only one day (1-day EEP) will make the same event extreme daily precipitation. The concept of EEPs with four representative EEP scenarios from a precipitation event is presented in Figure 5. The EEP can be defined using four indicators, i.e., amount, concentration ratio, duration, and frequency, as given as follows:

(7)$d_{i+1}=t_{i+1,e}-t_{i+1,s}+1$

(8)$V_{i+1}={{\displaystyle \int }}_{t_{i+1,s}}^{t_{i+1,e}}P\left(t\right)dt$

(9)$e_{i+1}=\frac{\left[{{\displaystyle \sum }}_{k=1}^{n}P(t_{i+1,k)}\right]}{V_{i+1}}, n=1,2,3$

The starting and ending days of the event are denoted by t_i+1,S and t_i+1,e, respectively, d_i+1, E_i+1, and V_i+1 represent the duration, concentration ratio, and amount of the ratio of EEP_i+1. T_i+1,k represents the time of the kth extreme daily precipitation of the event (for EEP_i+1, n = 1), and n denotes the total number of daily precipitation extremes over the course of the precipitation event. The timeseries of precipitation is depicted by P(t). The duration and concentration ratio are equal to 1 for a 1-day EEP, as shown in Figure 5).

The EEP and TDP definitions are explained in Figure 5. An event is declared as EEP if the total sum of precipitation over consecutive days is more than 1 mm with at least one daily precipitation extreme above the 99th percentile. The horizontal brown and black lines denote the 99th percentiles and the 1 mm precipitation threshold, respectively. The vertical dashed lines should be considered as the EEPs’ median time obtained by dividing the event period into two parts. TDP1 (EEP_i+1) and TDP2 (EEP_i+2) represent the EEP with daily precipitation extremes in the first and second halves of the precipitation event, respectively. In contrast, TDP3 (EEP_i+3) represents EEP with daily precipitation extremes lying in both halves of the event.

3.5. Time Distribution Patterns

The extreme precipitation events in the real world are dynamic and complex, making it difficult to determine which time distribution pattern (TDP) can be used to characterize temporal processes of extreme precipitation [49]. Hence, a systematic categorization of TDPs of EEPs is inevitable because this process is computationally challenging and complex. Therefore, the duration of multiday EEP is divided into two equal parts, and the event TDP is classified according to the duration during which an extreme precipitation event is distributed. This study used the daily precipitation data to analyze the TDPs of EEPs on a large regional scale.

This approach was developed based on previous research that divided the duration of extreme precipitation into multiple identical sections and categorized TDPs according to peak intensity positions [50]. The defined method yields three TDPs, i.e., TDP1, TDP2 and TDP3, demonstrating that all extreme daily precipitation for a multiday EEP lies in the first half, second half and middle of the event duration.

A multiday EEP may occasionally have only one daily extreme precipitation, which occurs on the middle day of the event. In this case, the daily precipitation extreme is not correlated with either the first or second half of the event period. Therefore, if the cumulative precipitation in the first/second half of the event period is greater than the second/first half portion, such EEP is observed as TDP1/TDP2. It is also worth noting that 1-day EEP has no TDP because it only lasts one day. There are four types of EEPs, including a one-day EEP and three TDPs. The TDP1, TDP2, TDP3, and 1-day EEP are designated as EEP_i, EEP_i+1, EEP_i+2, and EEP_i+3, respectively (Figure 5).

4. Results and Discussion

4.1. Homogeneity Analysis

In this study, four homogeneity tests, i.e., standard normal homogeneity test (SNHT), cumulative deviation (CD), Buishand’s range (BR) and Pettitt tests, were used to evaluate the daily precipitation timeseries data over JRB at a significance level of 1% and their results are presented in Table 2. The results indicate that 35.71% of daily precipitation timeseries were inhomogeneous using the Pettitt test compared to 21.14% of timeseries by CD, SNHT and BR tests. All four tests detected two timeseries, Plandri and Khandar stations, as inhomogeneous. Several factors, such as gauge relocation, instrument replacement, instrument malfunctioning, and changes in the observation or computation methods, could be the reasons behind this inhomogeneity in precipitation timeseries. Moreover, no established standards are available for making reasonable decisions based on the findings of various studies. An incorrect decision to include an inhomogeneous timeseries in analysis or exclude a homogeneous timeseries can result from an inaccurate representation of hydrometeorological conditions. Wijngaard et al. [51] conducted several experiments that rejected the null hypothesis of data homogeneity, and the findings of four tests were divided into three categories: useful, doubtful, and suspect.

Out of a total of four (04) homogeneity tests, Pettitt’s test was deemed to be more sensitive. The results of these tests display inconsistencies in some instances because the homogeneity tests have different sensitivities to the variability and changes in the station’s precipitation timeseries. Based on the four (04) homogeneity analyses, the overall classification and qualitative interpolation of the precipitation timeseries sequence revealed that 85.7% and 14.3% of the timeseries were categorized as useful and suspect, respectively. These results indicate that most stations over JRB represent the homogeneous precipitation timeseries.

4.2. Variations in the Extreme Precipitation Indices (EPIs)

The results of mean annual trends of precipitation indices and the number of stations exhibiting positive, negative and no-trends over JRB are presented in Table 3. The fixed threshold-based EPIs, i.e., R10, R20 and R25, exhibited decreasing trends at 9, 8 and 8 stations, respectively. Both positive and negative trends were dominant for CDD and CWD, respectively, indicating an increased frequency of drought and flooding [24,52]. In contrast, decreasing trends were found to be dominant over the JRB for the percentile (i.e., R95p and R99p), absolute (Rx1d and Rx5d) and other (Prcpt and SDII) EPIs. However, most of these decreasing trends were insignificant except for the Prcpt EPI, which exhibited significant decreasing trends at seven (07) stations.

4.3. Spatiotemporal Variations in EPIs

The MK and ITA methods were applied to analyze the variations in various EPIs, and their results are presented in Table 4. A combination of both positive and negative trends was detected at different stations over the JRB. Significant increasing trends were evident at the Naran station in 09 EPIs [24], whereas CWD and Rx1d exhibited decreasing trends [53]. The CWD and Prcpt indices exhibited widespread decreasing trends for threshold-based EPIs of R10m, R20m, and R25m; however, CDD displayed increasing trends in the lower reaches of JRB, consistent with the findings of [54]. Table 4 showed that most stations indicated an insignificant decreasing trend, and similarly, non-threshold EPIs exhibited diverse trends at most stations. Figure 6 presents the graphical representation of the ITA method for various EPIs. For each station, increasing and decreasing trends are evident in different EPIs for low, median and high values. Most stations’ data points fall within the 10% range of the 1:1 line (no trend) and represent insignificant trends at a 90% confidence interval. However, more obvious increasing trends were evident in high values for CDD EPI compared to low values and vice versa for CWD, indicating more intense extreme precipitation events over JRB. Therefore, it represents the enhanced ability of the ITA method to detect the trends within different ranges of EPIs [31].

The duration-based indices of CDD and CWD exhibited opposite trends. For instance, increasing trends for CDD were detected at the high-altitude stations, whereas there were decreasing trends at lower altitudes stations and vice versa for CWD [54]. At Rawalakot, CDD showed a significant increase. In contrast, its neighboring Bagh station showed decreasing trends because at Rawalakot station, the majority of days in several years are dry and seldom have extreme events. The major portion of the JRB showed an insignificant increasing and decreasing trend, whereas there were significant decreasing trends for CWD as shown in Figure 7. This indicates the tendency of increased frequency of extreme events in the form of floods and droughts as predicted in previous studies [53].

Widespread significant trends were detected for the threshold-based EPIs, i.e., R10, R20, and R25 over JRB, whereas only the Naran station exhibited a significant increasing trend with a magnitude of 0.45 to 0.74 days/year, as shown in Figure 7. The results of the significant increasing trend at Naran station are consistent with the findings of [24]. Moreover, the number of days with very heavy precipitation (R20) exhibited decreasing trends in the south-east (i.e., Gulmarg and Kupwara) at a magnitude of 0.98 to 1.2 days/year and in the western (i.e., Muzaffarabad, Bagh and Rawalakot) parts at the magnitude of −0.34 to 0.087 days/year. In comparison, a significant decrease in the frequency of extremely heavy precipitation days (R25) at the rate of 0.29–0.038 days/year was detected in the eastern part of the JRB. These widespread increasing and decreasing trends were consistent with the findings of its neighboring upper Indus basin [55].

The percentile-based (i.e., R95p, R99p) and absolute EPIs (i.e., Rx1d, Rx5d) exhibited widespread decreasing trends of varying magnitudes over JRB as shown in Table 4 and Figure 7. The precipitation on very wet days (R95p) exhibited various positive and negative trends with varying magnitude. For instance, a significant reduction of 6.09 mm/year was observed for R95p in the eastern region with an overall decline in most of the areas of the JRB. A drastic decrease in the magnitude of R95p was observed on the Indian side of the watershed of 4.56 mm/year, whereas the most significant increase of 3.8 mm/year was detected at Naran station [24].

The trend of the maximum one-day amount (Rx1day) depicts decreasing trends at most stations. A significant decreasing trend has been observed at Muzaffarabad, Bagh and Rawalakot stations, with magnitudes varying from −1 to −0.71 mm/year (Figure 7). GujarKhan station exhibited a significant increasing trend having a magnitude of 0.75 mm/year while Kupwara station showed no trend. The maximum five-day amount (Rx5day) across the JRB has decreased significantly in the eastern and western regions, including Gulmarg, Muzaffarabad, Bagh and Kotli stations. The stations located on the Indian side of the JRB exhibited decreasing trends. An insignificant increasing trend was observed in the Mangla watershed’s northern and southern regions, with magnitudes varying from 0.21–0.26 mm/year.

The EPI of total annual precipitation (Prcpt) exhibited a decreasing trend at most stations with a magnitude of −19.5 to −1.76 mm/year (Figure 7). However, Naran station showed a significant increasing trend of magnitude varying from 7.14 to 33.8 mm/year. In southern parts of the JRB majority of stations, including Muzaffarabad, Bagh, Rawalakot and Kotli, a decreasing trend was exhibited. The significant decreasing trends were dominated at the stations located on the Indian side of Kashmir. The simple daily intensity index (SDII) reflects a decreasing trend across the major parts of the JRB, explicitly in the eastern and western parts, with a magnitude of −0.1 to 0.07 mm/year. In contrast, significant trends were detected in the northern and southern parts at a magnitude of 0.15–0.21 mm/year. These results of various precipitation indices over JRB indicate an increase in the dry spells over JRB, which may impact the water resources availability of Mangla reservoir to meet the agricultural water requirements of the downstream users. It was also witnessed that the Mangla reservoir dropped to its dead level in 2018 with no water to operate for either irrigation or hydropower generation purposes [56].

4.4. Comparison of Results

The comparison of MK and ITA results of different EPIs over JRB is presented in Figure 8. At an annual scale, 132 timeseries were investigated for detecting variations in various precipitation indices, out of which 40.15% and 31.1% exhibited significant trends by using ITA and MK methods, respectively. The results of both statistical tests, ITA and MK, are comparable for the CDD as both tests showed increasing trends at Rawalakot and Gulmarg stations and a decrease at Srinagar, GujarKhan and Kotli stations. The annual total precipitation EPI showed consistent trends using ITA and MK methods. Similarly, an increasing trend at Naran, Mangla, Kallar and GujarKhan stations and decreasing trend at Gulmarg, Bagh and Muzaffarabad stations were detected for CWD, R10, R20, R25, R95p, R99p, Rx1day, Rx5day and SDII EPIs by using both MK and ITA methods as shown in Figure 8, which proves the reliability of the results of the ITA method. However, using the ITA method is advantageous because of the enhanced ability to detect the hidden trends in the timeseries by its graphical representation. Moreover, the ITA method has a general application, unlike the MK method, which has some restrictive measures of serial correlation and time series duration of timeseries.

4.5. Different Patterns of EEPs

Investigation of preceding and succeeding precipitation (PSP) of an extreme event is considered vital because a precipitation event comprises three stages, i.e., start, development and decline, and may last for several days; however, the above-mentioned EPIs only include the one-day precipitation as the extreme event for analysis. Therefore, the amount, duration, and concentration ratio of various EEP patterns were also investigated and their results, along with spatial distribution across JRB, are presented in Figure 9. An increase in the amount of all 1-day EEP patterns was detected in the northern region of the JRB. The distribution of 1-day EEP events with an amount of less than 50 mm was found obvious across southern parts; however, 1-day EEP events with amounts varying between 80 and 90 mm were detected in the northern regions. The event amount of TDP1 is generally higher than 1-day EEP with more than a 1-day EEP event amount. TDP is higher towards the northern part of the JRB, having an amount of 120–140 mm. The event amount of TDP3 is more significant than 215 mm across most parts of the JRB while its amount is increasing towards the northern region.

For the concentration ratio, the TDP1 values are higher toward the southern side of the Mangla watershed, which is greater than 0.7, and TDP2 values were also found to be similar but higher than TDP1. TDP3 showed higher values of 0.9 in the northern and southern parts but low values of 0.77 in the middle of the JRB. The event duration for TDP1, TDP2 and TDP3 showed that the TDP3 could last longer than TDP2 and TDP1 in general. Overall, the event duration of TDP1 and TDP2 could last less than four days. In contrast, TDP3 could last five days across JRB in many cases. Overall, the duration of highly extended events was mainly observed in the northern regions of the JRB.

5. Conclusions

Spatiotemporal variations in the extreme precipitation indices (EPIs) over the Jhelum River basin (JRB) were investigated at 12 stations to detect their impacts on floods and droughts using the Mann-Kendall (MK) and innovative trend analysis (ITA) methods. At an annual scale, 132 timeseries were investigated for detecting variations in various precipitation indices, of which 40.15% and 31.1% exhibited significant trends using ITA and MK methods, respectively, indicating the ability of the ITA method to detect the hidden trends. The following specific conclusions are drawn from the results of this study:

• Fixed and station-related threshold indices, i.e., R10, R20, R25, CWD, Prcpt, and R95p, exhibited significant decreasing trends at 06, 05, 04, 04, 07 and 05 stations, respectively, are decreasing and CDD is increasing at most parts of JRB.

• An increase in the CDD was found at most stations (i.e., nine stations) with a magnitude of 0.33 days/year, whereas CWD decreased at eight stations with a magnitude of 0.03 days/year. These results show a significant increase in drought events compared to floods.

• Naran station showed a significant increasing trend for most extreme precipitation indices, including Prcpt, R10, R20, R25, Rx5day and SDII. In contrast, CWD also showed a decreasing trend, which means more probability of flood events.

• The precipitation data established the EEPs concept with the event amount (40–290 mm), duration (2–7 days) and the concentration ratio (0.6–0.95).

The daily precipitation data at 12 out of 14 stations were found useful, while two (02) were suspicious. These stations were not considered for further analysis in this study. However, the inhomogeneity of these stations needs to be further investigated in future studies based on data, e.g., relocation of recording station, instrument change and changes in measurement or observation procedures. Moreover, in future development projects, the results of this study can present valuable information because drought and flood events have often been observed in this region.

Author Contributions

Conceptualization, I.A.; methodology, I.A. and F.A.; software, F.A. and L.W.; validation, I.A. and L.W.; formal analysis, I.A. and F.A.; investigation, F.A.; resources, L.W. and F.Z.; data curation, F.A.; writing—original draft preparation, I.A.; writing—review & editing, F.Z.; supervision, F.Z.; project administration, L.W. and F.Z.; funding acquisition, L.W. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not Applicable.

Informed Consent Statement

Not Applicable.

Data Availability Statement

The data used to support the findings of this study are available from the corresponding author upon reasonable request.

Acknowledgments

The authors thank the Pakistan Meteorological Department (PMD) and Water and Power Development Authority (WAPDA) for providing data to conduct this research.

Conflicts of Interest

The authors declare no conflict of interest.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figures and Tables

Figure 1. Location and domain of study area along with climatic stations.

Figure 2. Mean monthly precipitation variations over the Jhelum River basin (JRB).

Figure 3. Overview of the research framework.

Figure 4. Representation of increasing, decreasing and no trend areas in ITA methodology (a) increasing and decreasing trend regions, (b) low, medium and high value trend regions (c) monotonic trend representation and (d) ITA method explanation.

Figure 5. Definition of EEPs and TDPs patterns.

Figure 6. Variations in the EPIs using the ITA method.

Figure 7. Spatial and temporal distribution of EPIs over JRB.

Figure 8. Comparison of ITA and MK results at different stations over JRB.

Figure 9. Spatial distribution of amount, duration and concentration ratio for EEPs and TDPs.

Table 1

Extreme precipitation indices and their definitions.

Type	Index Name	ID	Definition	Unit
Duration	Consecutive dry days	CDD	No. of consecutive days with RR * < 1 mm	Days
	Consecutive wet days	CWD	No. of consecutive days with RR ≥ 1 mm	Days
Threshold	No. of heavy precipitation days	R10	Annual count of days with RR ≥ 10 mm	Days
	No. of very heavy precipitation days	R20	Annual count of days with RR ≥ 20 mm	Days
	No. of extremely heavy precipitation days	R25	Annual count of days with RR ≥ 25 mm	Days
Percentile	Precipitation on very wet days	R95p	Annual total precipitation when RR > 95th percentile	mm
	Precipitation on extremely wet days	R99p	Annual total precipitation when RR > 99th percentile	mm
Absolute	1-day maximum precipitation	Rx1d	Annual maximum 1-day precipitation	mm
	5-day maximum precipitation	Rx5d	Annual maximum consecutive 5-day precipitation	mm
Others	Annual total wet-day precipitation	Prcpt	Total annual precipitation during wet days (≥1 mm)	mm
	Simple daily intensity index	SDII	Annual total precipitation divided by the number of wet days (RR ≥ 1 mm) in a year	mm/day

* Rainfall.

Table 2

Summary of homogeneity tests’ results of selected stations in daily precipitation timeseries.

Station	Data Length(Years)	Pettitt’s Test(Xk)	CD Test	BR Test	SNHT(To)	Remarks
Kotli	47	221	1.12	1.15	7.52	Useful
Plandri	47	324 **	2.3 **	1.24	12.1 **	Suspect
Rawalakot	47	335 **	1.45	1.71	10.9	Useful
GujarKhan	47	169	0.9	1.2	7.17	Useful
Kallar	47	92	0.51	1.02	2.67֗	Useful
Mangla	47	110	0.7	1.06	3.1	Useful
Muzaffarabad	47	156	0.86	1.38	4.47	Useful
Srinagar	34	129	0.91	1.44	4.37	Useful
Kupwara	34	129	0.9	1.43	4.36	Useful
Gulmarg	34	227	1.64	1.71	14.65 **	Useful
Balakot	47	148	0.77	1.63	7.52	Useful
Khandar	47	316 **	1.82 **	1.94 **	13.1 **	Suspect
Bagh	47	314 **	2.08 **	1.65	10.3	Useful
Naran	47	404 **	1.49	2.02 **	11.2	Useful

*, ** and *** represents the trends at the 10%, 5% and 1% significance levels.

Table 3

Number of stations with positive or negative trends in different EPIs.

Type	ID	Positive Trend			Negative Trend			No Trend
		Total	SS	NS	Total	SS	NS
Duration	CDD	9	1	8	3	0	3	0
	CWD	4	0	4	8	4	4	0
Threshold	R10	3	1	2	9	6	3	0
	R20	4	1	3	8	5	3	0
	R25	4	1	3	8	4	4	0
Percentile	R95p	2	1	1	10	5	5	0
	R99p	4	1	3	8	1	7	0
Absolute	Rx1d	1	1	0	11	3	7	1
	Rx5d	2	0	2	10	4	6	0
Others	Prcpt	2	1	1	10	7	3	0
	SDII	4	3	1	8	1	7	0

Table 4

MK and ITA methods results of different EPIs.

Station	Test	CDD	CWD	Prcptot	R10	R20	R25	R95p	R99p	Rx1day	Rx5day	SDII
Bagh	ITA	1.56	0.67	−2.61 *	−2.28 *	−3.02 *	−3.45 *	−4.32 *	−2.64 *	−2.97 *	−1.66 *	−0.14
	MK	−0.28	0.8	−2.74 *	−2.86 *	−2.86 *	−3.09 *	−2.94 *	−1.45	−2.5 *	−1.85 *	−1.56
Balakot	ITA	0.14	−1.63	−1.13	−1.08	−0.54	−0.73	−1.42	−2.73 *	−0.53	−0.77	−0.87
	MK	−0.02	−2.61 *	−1.47	−1.19	−0.27	−0.30	−0.83	−1.08	−0.90	−1.16	−0.53
G.Khan	ITA	0.18	−0.95	0.07	−0.10	−0.23	−0.01	0.78	4.55 *	3.65 *	1.27	0.55
	MK	0.29	−1.52	0.57	0.29	0.53	1.07	1.16	1.41	2.72 *	1.34	1.70 *
Gulmarg	ITA	0.85	−2.19 *	−2.10 *	−1.96 *	−1.85 *	−2.21 *	−3.24 *	−5.18 *	−1.91 *	−2.23 *	−1.43
	MK	1.16	−2.45 *	−2.49 *	−2.54 *	−1.95 *	−1.76 *	−0.95	−1.77 *	−1.07	−2.31 *	−2.24 *
Kallar	ITA	0.07	−1.03	−0.38	−0.35	0.26	0.76	−1.70 *	6.51 *	0.78	−0.09	0.91
	MK	0.92	−1.31	−0.05	−0.46	0.08	0.08	−0.73	1.41	−0.03	−0.71	2.10 *
Kotli	ITA	−1.72 *	−1.30	−1.40	−1.16	−0.54	−0.83	−3.00 *	−3.35 *	−0.54	−1.05	0.01
	MK	0.06	−2.19 *	−2.29 *	−1.64 *	−0.55	−1.23	−1.97 *	−1.00	−1.27	−1.87 *	0.17
Kupwara	ITA	−0.84	−2.27 *	0.28	−0.41	0.63	0.67	2.63 *	1.94 *	2.13 *	0.93	0.35
	MK	0.48	−2.07	−1.81 *	−2.05 *	−2.66 *	−0.47	−0.80	−0.91	0.01	−0.68	−1.36
Mangla	ITA	−0.70	1.03	−0.51	−0.18	−0.21	0.49	−2.50 *	1.04	0.14	−0.93	−0.26
	MK	0.16	0.89	−0.69	0.94	0.47	0.29	−2.80 *	0.05	−0.28	−1.10	−0.84
M.Abad	ITA	2.02 *	0.67	−2.61 *	−2.28 *	−3.28 *	−3.45 *	−4.32 *	−2.64 *	−2.97 *	−1.66 *	−0.14
	MK	−0.31	0.80	−2.74 *	−1.73 *	−2.86 *	−3.09 *	−2.94 *	−1.45	−2.50 *	−1.85 *	−1.56
Narran	ITA	2.94 *	−3.25 *	3.83 *	3.50 *	9.69 *	9.07 *	4.54 *	6.23 *	−0.63	1.33	5.72 *
	MK	1.67 *	−0.02	4.53 *	3.63 *	5.13 *	4.95 *	1.65 *	2.02 *	−0.09	2.51 *	5.54 *
Rawalakot	ITA	3.91 *	0.73	−2.0 *	−1.94 *	−3.23 *	−3.87 *	−2.76 *	2.42 *	−0.93	0.01	−0.61
	MK	3.00 *	0.58	−3.47 *	−3.31 *	−3.65 *	−3.86 *	−2.59 *	−1.62	−2.08 *	−1.16	−1.83
Srinagar	ITA	0.31	−2.19 *	−1.48	−1.04	−2.19 *	−3.03 *	−2.81 *	−4.36 *	−2.08 *	−1.79 *	−0.79
	MK	0.71	−0.71	−2.02 *	−1.09	−1.41	−0.60	−0.53	−0.79	−0.56	−0.67	−0.85

*, ** and *** represents the trends at the 10%, 5% and 1% significance levels.

References

1. Bozkurt, D.; Bromwich, D.H.; Carrasco, J.; Rondanelli, R. Temperature and precipitation projections for the Antarctic Peninsula over the next two decades: Contrasting global and regional climate model simulations. Clim. Dyn.; 2021; 56, pp. 3853-3874. [DOI: https://dx.doi.org/10.1007/s00382-021-05667-2]

2. IPCC. Summary for Policymakers. Climate Change 2013—The Physical Science Basis. Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change; Cambridge University Press: Cambridge, UK, 2013; 33. [DOI: https://dx.doi.org/10.1017/CBO9781107415324]

3. Tabari, H. Climate change impact on flood and extreme precipitation increases with water availability. Sci. Rep.; 2020; 10, 13768. [DOI: https://dx.doi.org/10.1038/s41598-020-70816-2] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32792563]

4. Nicholls, N.; Easterling, D.; Goodess, C.; Kanae, S.; Kossin, J.; Luo, Y.; Marengo, J.; McInnes, K.; Rahimi, M.; Reichstein, M. et al. Changes in Climate Extremes and their Impacts on the Natural Physical Environment. Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation; Field, C.; Barros, V.; Stocker, T.; Qin, D.; Dokken, D.; Ebi, K.; Mach, K.; Plattner, G.; Allen, S.; Tignor, M. et al. Cambridge University Press: Cambridge, UK, New York, NY, USA, 2012; pp. 109-230.

5. Lutz, A.F.; Immerzeel, W.W.; Kraaijenbrink, P.D.A.; Shrestha, A.B.; Bierkens, M.F.P. Climate change impacts on the upper indus hydrology: Sources, shifts and extremes. PLoS ONE; 2016; 11, e0165630. [DOI: https://dx.doi.org/10.1371/journal.pone.0165630] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27828994]

6. Janjua, S.; Hassan, I.; Muhammad, S.; Ahmed, S.; Ahmed, A. Water management in Pakistan’s Indus Basin: Challenges and opportunities. Water Policy; 2021; 23, pp. 1329-1343. [DOI: https://dx.doi.org/10.2166/wp.2021.068]

7. Pakmehr, S.; Yazdanpanah, M.; Baradaran, M. How collective efficacy makes a difference in responses to water shortage due to climate change in southwest Iran. Land Use Policy; 2020; 99, 104798. [DOI: https://dx.doi.org/10.1016/j.landusepol.2020.104798]

8. NDMA. Annual Report 2018, National Disaster Management Authority; Ministry of Climate Change, Government of Pakistan: Islamabad, Pakistan, 2019.

9. Hussain, M.; Butt, A.R.; Uzma, F.; Ahmed, R.; Irshad, S.; Rehman, A.; Yousaf, B. A comprehensive review of climate change impacts, adaptation, and mitigation on environmental and natural calamities in Pakistan. Environ. Monit. Assess.; 2020; 192, 48. [DOI: https://dx.doi.org/10.1007/s10661-019-7956-4]

10. Shahzad, N.; Amjad, M. Climate Change and Food Security in Pakistan. Sustainable Agriculture and Food Security, World Sustainability Series; Leal Filho, W.; Kovaleva, M.; Popkova, E. Springer: Cham, Switzerland, 2022; pp. 579-594. [DOI: https://dx.doi.org/10.1007/978-3-030-98617-9_33]

11. Shah, S.M.H.; Mustaffa, Z.; Teo, F.Y.; Imam, M.A.H.; Yusof, K.W.; Al-Qadami, E.H.H. A review of the flood hazard and risk management in the South Asian Region, particularly Pakistan. Sci. African; 2020; 10, e00651. [DOI: https://dx.doi.org/10.1016/j.sciaf.2020.e00651]

12. Raymond, C.; Horton, R.M.; Zscheischler, J.; Martius, O.; AghaKouchak, A.; Balch, J.; Bowen, S.G.; Camargo, S.J.; Hess, J.; Kornhuber, K. et al. Understanding and managing connected extreme events. Nat. Clim. Chang.; 2020; 10, pp. 611-621. [DOI: https://dx.doi.org/10.1038/s41558-020-0790-4]

13. Patakamuri, S.K.; Muthiah, K.; Sridhar, V. Long-Term homogeneity, trend, and change-point analysis of rainfall in the arid district of ananthapuramu, Andhra Pradesh State, India. Water; 2020; 12, 211. [DOI: https://dx.doi.org/10.3390/w12010211]

14. Basher, M.A.; Stiller-Reeve, M.A.; Saiful Islam, A.K.M.; Bremer, S. Assessing climatic trends of extreme rainfall indices over northeast Bangladesh. Theor. Appl. Climatol.; 2018; 134, pp. 441-452. [DOI: https://dx.doi.org/10.1007/s00704-017-2285-4]

15. Zhang, M.; Chen, Y.; Shen, Y.; Li, Y. Changes of precipitation extremes in arid Central Asia. Quat. Int.; 2017; 436, pp. 16-27. [DOI: https://dx.doi.org/10.1016/j.quaint.2016.12.024]

16. Bhatti, A.S.; Wang, G.; Ullah, W.; Ullah, S.; Hagan, D.F.T.; Nooni, I.K.; Lou, D.; Ullah, I. Trend in Extreme Precipitation Indices Based on Long Term In Situ Precipitation Records over Pakistan. Water; 2020; 12, 797. [DOI: https://dx.doi.org/10.3390/w12030797]

17. Xia, J.; Yang, X.; Liu, J.; Wang, M.; Li, J. Dominant change pattern of extreme precipitation and its potential causes in Shandong Province, China. Sci. Rep.; 2022; 12, 858. [DOI: https://dx.doi.org/10.1038/s41598-022-04905-9] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35039594]

18. White, R.H.; Battisti, D.S.; Skok, G. Tracking precipitation events in time and space in gridded observational data. Geophys. Res. Lett.; 2017; 44, pp. 8637-8646. [DOI: https://dx.doi.org/10.1002/2017GL074011]

19. Ma, S.; Zhou, T.; Dai, A.; Han, Z. Observed changes in the distributions of daily precipitation frequency and amount over China from 1960 to 2013. J. Clim.; 2015; 28, pp. 6960-6978. [DOI: https://dx.doi.org/10.1175/JCLI-D-15-0011.1]

20. Lu, E.; Zhao, W.; Zou, X.; Ye, D.; Zhao, C.; Zhang, Q. Temporal-spatial monitoring of an extreme precipitation event: Determining simultaneously the time period it lasts and the geographic region it affects. J. Clim.; 2017; 30, pp. 6123-6132. [DOI: https://dx.doi.org/10.1175/JCLI-D-17-0105.1]

21. Khan, F.; Ali, S.; Mayer, C.; Ullah, H.; Muhammad, S. Climate change and spatio-temporal trend analysis of climate extremes in the homogeneous climatic zones of Pakistan during 1962–2019. PLoS ONE; 2022; 17, e0271626. [DOI: https://dx.doi.org/10.1371/journal.pone.0271626]

22. Ali, G.; Sajjad, M.; Kanwal, S.; Xiao, T.; Khalid, S.; Shoaib, F.; Gul, H.N. Spatial–temporal characterization of rainfall in Pakistan during the past half-century (1961–2020). Sci. Rep.; 2021; 11, 6935. [DOI: https://dx.doi.org/10.1038/s41598-021-86412-x]

23. Rafiq, M.; Li, Y.C.; Cheng, Y.; Rahman, G.; Ullah, I.; Ali, A. Spatial and temporal fluctuation of rainfall and drought in Balochistan province, Pakistan. Arab. J. Geosci.; 2022; 15, 214. [DOI: https://dx.doi.org/10.1007/s12517-022-09514-4]

24. Saddique, N.; Khaliq, A.; Bernhofer, C. Trends in temperature and precipitation extremes in historical (1961–1990) and projected (2061–2090) periods in a data scarce mountain basin, northern Pakistan. Stoch. Environ. Res. Risk Assess.; 2020; 34, pp. 1441-1455. [DOI: https://dx.doi.org/10.1007/s00477-020-01829-6]

25. Şen, Z. Innovative trend analysis methodology. J. Hydrol. Eng.; 2012; 17, pp. 1042-1046. [DOI: https://dx.doi.org/10.1061/(ASCE)HE.1943-5584.0000556]

26. Şen, Z. Trend Identification Simulation and Application. J. Hydrol. Eng.; 2014; 19, pp. 635-642. [DOI: https://dx.doi.org/10.1061/(ASCE)HE.1943-5584.0000811]

27. Dong, Z.; Jia, W.; Sarukkalige, R.; Fu, G.; Meng, Q.; Wang, Q. Innovative trend analysis of air temperature and precipitation in the jinsha river basin, china. Water; 2020; 12, 3293. [DOI: https://dx.doi.org/10.3390/w12113293]

28. Dabanli, İ. A Climate Change Impact: Variation in Precipitation Patterns, and Increased Drought Risk in Turkey. Sak. Univ. J. Sci.; 2019; 23, pp. 193-202. [DOI: https://dx.doi.org/10.16984/saufenbilder.467119]

29. Alemu, Z.A.; Dioha, M.O. Climate change and trend analysis of temperature: The case of Addis Ababa, Ethiopia. Environ. Syst. Res.; 2020; 9, 27. [DOI: https://dx.doi.org/10.1186/s40068-020-00190-5]

30. Jenifer, M.A.; Jha, M.K. Assessment of precipitation trends and its implications in the semi-arid region of Southern India. Environ. Chall.; 2021; 5, 100269. [DOI: https://dx.doi.org/10.1016/j.envc.2021.100269]

31. Wang, Y.; Xu, Y.; Tabari, H.; Wang, J.; Wang, Q.; Song, S.; Hu, Z. Innovative trend analysis of annual and seasonal rainfall in the Yangtze River Delta, eastern China. Atmos. Res.; 2020; 231, 104673. [DOI: https://dx.doi.org/10.1016/j.atmosres.2019.104673]

32. Mahmood, R.; Babel, M.S.; Jia, S. Assessment of temporal and spatial changes of future climate in the Jhelum river basin, Pakistan and India. Weather Clim. Extrem.; 2015; 10, pp. 40-55. [DOI: https://dx.doi.org/10.1016/j.wace.2015.07.002]

33. De Souza, K.; Kituyi, E.; Harvey, B.; Leone, M.; Murali, K.S.; Ford, J.D. Vulnerability to climate change in three hot spots in Africa and Asia: Key issues for policy-relevant adaptation and resilience-building research. Reg. Environ. Chang.; 2015; 15, pp. 747-753. [DOI: https://dx.doi.org/10.1007/s10113-015-0755-8]

34. Domonkos, P. Homogenization of precipitation time series with ACMANT. Theor. Appl. Climatol.; 2015; 122, pp. 303-314. [DOI: https://dx.doi.org/10.1007/s00704-014-1298-5]

35. Alexandersson, H.; Moberg, A. Homogenization of Swedish Temperature Data Part I: Homogeneity Test for Linear Trends. Int. J. Climatol.; 1997; 17, pp. 25-34. [DOI: https://dx.doi.org/10.1002/(SICI)1097-0088(199701)17:1<25::AID-JOC103>3.0.CO;2-J]

36. Buishand, T.A. Some methods for testing the homogeneity of rainfall records. J. Hydrol.; 1982; 58, pp. 11-27. [DOI: https://dx.doi.org/10.1016/0022-1694(82)90066-X]

37. Pettitt, A.N. A Non-Parametric Approach to the Change-Point Problem. Appl. Stat.; 1979; 28, 126. [DOI: https://dx.doi.org/10.2307/2346729]

38. Mondal, S.K.; Huang, J.; Wang, Y.; Su, B.; Kundzewicz, Z.W.; Jiang, S.; Zhai, J.; Chen, Z.; Jing, C.; Jiang, T. Changes in extreme precipitation across South Asia for each 0.5 °C of warming from 1.5 °C to 3.0 °C above pre-industrial levels. Atmos. Res.; 2022; 266, 105961. [DOI: https://dx.doi.org/10.1016/j.atmosres.2021.105961]

39. Yang, X.; Wu, J.; Liu, J.; Ye, X. Changes of extreme precipitation and possible influence of enso events in a humid basin in China. Atmosphere; 2021; 12, 1522. [DOI: https://dx.doi.org/10.3390/atmos12111522]

40. Kendall, M. Rank Correlation Methods; 4th ed. Charles Griffin: London, UK, 1975.

41. Mann, H.B. Nonparametric tests against trend. Econom. J. Econom. Soc.; 1945; 13, pp. 245-259. [DOI: https://dx.doi.org/10.2307/1907187]

42. Tabari, H.; Marofi, S.; Aeini, A.; Talaee, P.H.; Mohammadi, K. Trend analysis of reference evapotranspiration in the western half of Iran. Agric. For. Meteorol.; 2011; 151, pp. 128-136. [DOI: https://dx.doi.org/10.1016/j.agrformet.2010.09.009]

43. Heureux, A.M.C.; Alvar-Beltrán, J.; Manzanas, R.; Ali, M.; Wahaj, R.; Dowlatchahi, M.; Afzaal, M.; Kazmi, D.; Ahmed, B.; Salehnia, N. et al. Climate Trends and Extremes in the Indus River Basin, Pakistan: Implications for Agricultural Production. Atmosphere; 2022; 13, 378. [DOI: https://dx.doi.org/10.3390/atmos13030378]

44. Chen, T.; Ao, T.; Zhang, X.; Li, X.; Yang, K. Climate Change Characteristics of Extreme Temperature in the Minjiang River Basin. Adv. Meteorol.; 2019; 2019, 1935719. [DOI: https://dx.doi.org/10.1155/2019/1935719]

45. Wu, H.; Qian, H. Innovative trend analysis of annual and seasonal rainfall and extreme values in Shaanxi, China, since the 1950s. Int. J. Climatol.; 2017; 37, pp. 2582-2592. [DOI: https://dx.doi.org/10.1002/joc.4866]

46. Oueslati, B.; Camberlin, P.; Zoungrana, J.; Roucou, P.; Diallo, S. Variability and trends of wet season temperature in the Sudano-Sahelian zone and relationships with precipitation. Clim. Dyn.; 2018; 50, pp. 1067-1090. [DOI: https://dx.doi.org/10.1007/s00382-017-3661-6]

47. Wang, Y.; Zhou, L. Observed trends in extreme precipitation events in China during 1961-2001 and the associated changes in large-scale circulation. Geophys. Res. Lett.; 2005; 32, L09707. [DOI: https://dx.doi.org/10.1029/2005GL023769]

48. She, D.; Shao, Q.; Xia, J.; Taylor, J.A.; Zhang, Y.; Zhang, L.; Zhang, X.; Zou, L. Investigating the variation and non-stationarity in precipitation extremes based on the concept of event-based extreme precipitation. J. Hydrol.; 2015; 530, pp. 785-798. [DOI: https://dx.doi.org/10.1016/j.jhydrol.2015.10.029]

49. Huff, F. Time Distributions of Heavy Rainstorms in Illinois Time Distributions of Heavy Rainstorms in Illinois; Illinois State Water Survey: Campaign, IL, USA, 1990.

50. Wu, X.; Guo, S.; Yin, J.; Yang, G.; Zhong, Y.; Liu, D. On the event-based extreme precipitation across China: Time distribution patterns, trends, and return levels. J. Hydrol.; 2018; 562, pp. 305-317. [DOI: https://dx.doi.org/10.1016/j.jhydrol.2018.05.028]

51. Wijngaard, J.B.; Klein Tank, A.M.G.; Können, G.P. Homogeneity of 20th century European daily temperature and precipitation series. Int. J. Climatol.; 2003; 23, pp. 679-692. [DOI: https://dx.doi.org/10.1002/joc.906]

52. Hussain, A.; Cao, J.; Hussain, I.; Begum, S.; Akhtar, M.; Wu, X.; Guan, Y.; Zhou, J. Observed Trends and Variability of Temperature and Precipitation and Their Global Teleconnections in the Upper Indus Basin, Hindukush-Karakoram-Himalaya. Atmosphere; 2021; 12, 973. [DOI: https://dx.doi.org/10.3390/atmos12080973]

53. Sajjad, H.; Ghaffar, A. Observed, simulated and projected extreme climate indices over Pakistan in changing climate. Theor. Appl. Climatol.; 2019; 137, pp. 255-281. [DOI: https://dx.doi.org/10.1007/s00704-018-2573-7]

54. Abbas, S.; Yaseen, M.; Latif, Y.; Waseem, M.; Muhammad, S.; Leta, M.K.; Sher, S.; Imran, M.A.; Adnan, M.; Khan, T.H. Spatiotemporal Analysis of Climatic Extremes over the Upper Indus Basin, Pakistan. Water; 2022; 14, 1718. [DOI: https://dx.doi.org/10.3390/w14111718]

55. Hartmann, H.; Andresky, L. Flooding in the Indus River basin—A spatiotemporal analysis of precipitation records. Glob. Planet. Chang.; 2013; 107, pp. 25-35. [DOI: https://dx.doi.org/10.1016/j.gloplacha.2013.04.002]

56. Zulfiqar, M.A. Water crisis in Pakistan: Facts and solutions. Nation. 2020; Available online: https://www.nation.com.pk/07-Aug-2020/water-crisis-in-pakistan-facts-and-solutions (accessed on 1 November 2022).