https://orcid.org/0000-0002-1970-1662
Abstract
The determination of rainfall characteristics of a basin is obligatory for mitigation planning and adaptation. This study presents the spatiotemporal analysis of gridded precipitation to understand the variability and seasonality over the Rapti Basin. This basin is affected by frequent flooding; hence it becomes necessary to understand the variability of rainfall. To carry out this analysis, daily precipitation data is extracted for 22 grids for 30 years (1990–2020) of high resolution. Extreme rainfall indices are computed as per the threshold decided by the Indian Meteorological Department (IMD). To determine the trend of the rainfall, Mann Kendall (MK) test, Modified MK test and Innovative Trend Analysis was used and magnitude of trend was determined using slope. Trends of daily rainfall, annual, and seasonal rainfall were determined. Annual rainfall has shown significant decreasing trend over the basin. Monsoon season has a noticeable tendency, but other seasons have only seen small shifts. From all the rainfall indices, PRCPTOT and SDII showed a decreasing trend with Sen Slope of −2.56 and 3.00 respectively. While examining the variance in precipitation across the basin, this study will be helpful.
HIGHLIGHTS
Rainfall characteristics at Rapti basin are determined using gridded data.
Significant decreasing trend was observed in annual rainfall of the basin.
PRCPTOT and SDII showed major decreasing trend with Sen Slope of −2.56 and −3.00 respectively.
Gorakhpur district has highest SDII in the basin.
Flooding in the basin has increased after the change point of the rainfall.
INTRODUCTION
Precipitation is the crucial source of water on earth serving humankind. Understanding rainfall variability is critical for controlling the available water resources and budgeting for them. Owing to spatial and temporal variation in rainfall, severe consequences may arise due to excessive scarcity or surplus of rainfall (Swain et al. 2021b). Presently, climate change has impacted variability of rainfall which is predominantly caused by anthropogenic activities and global warming. Change in climate has global effects on variation, intensity and distribution of the rainfall (Zhang et al. 2019). More researchers are focusing on precipitation variability and the implications for water resource management, flood or drought predictions, given the differences in global warming that are triggering changes in the hydrological phenomena (Gu et al. 2017).
As per the Intergovernmental Panel on Climate Change (IPCC), as global warming conditions increase, the changes occurring in hydrological cycles will dominate, henceforward making extreme rainfall events more recurrent and intense (Solomon et al. 2007). The overall management of the water resources depends upon the rainfall pattern of the region and in India, monsoon holds the utmost importance as most of the population of India depends upon agriculture for their livelihood (Roy & Balling 2004; Nageswararao et al. 2018). Significant long-term change in the rainfall pattern in India has been observed due to climate changes (Pandey & Khare 2018; Maharana et al. 2020). Heat and cold waves, snowfall, monsoon depressions, cyclones, floods and droughts are major events affecting India (Dash et al. 2007). Therefore, it turns out to be essential to comprehend the capriciousness of precipitation at diverse time scales.
Trend analysis is used to determine rate of change occurring in time series of the variables taken into consideration. Non-parametric tests used for trend detection have less assumptions and more advantages than parametric tests (Kumar et al. 2019). Parametric tests are dependent on the population of the dataset. Linear regression is one of the parametric tests used to detect trends in time series. Parametric methods are grounded on conventions and non-parametric tests more robust as they do not depend on distribution of the data. Effect of outliers is also seen on the trend detection using parametric tests while this is not the case in non-parametric tests. The Mann-Kendall (MK) test and modified Mann-Kendall (MMK) test are widely accepted non-parametric tests which are used for trend detection (Duhan & Pandey 2013). Results of MK tests are altered by the autocorrelation present in the data and hence the MMK test is used to overcome this effect, so it gives more convenient results (Hamed & Ramachandra Rao 1998; Pour et al. 2020; Swain et al. 2021a). Rajasthan's rainfall trend is assessed using the MK test and Sen's slope (Mehta & Yadav 2021, 2022). Chadee et al. (2023), Anushka et al. (2020) and Khaniya et al. (2020) have used trend tests to understand the climate variability in Sri Lanka and the West Indies using the Mann-Kendall test. Patel & Mehta (2023), Mehta & Yadav (2023) and Mehta (2022) have used trend detection tests to analyse temporal variation of rainfall in Pali, Jalore district of Rajasthan. Mehta & Yadav (2021) have analysed rainfall variability over Barmer district of Rajasthan and concluded that there is a decreasing winter trend while pre-monsoon, post-monsoon and south-west monsoon trends are increasing.
Using the Expert Team on Climate Change Detection and Indices (ETCCDI) indexes, the characteristics of extreme rainfall are assessed. Twenty-seven core indices were given by them to determine the characteristics of precipitation and temperature (Alexander et al. 2019). Out of all indices given to measure characteristics of precipitation, total number of wet days, consecutive dry days, maximum 1 day rainfall, maximum 5-day rainfall, and total precipitation in wet days of the year are the significant indices used. Threshold for wet days is taken as 1 mm/day while 10 mm/day, 20 mm/day and 40 mm/day are taken as thresholds for heavy rainfall. These thresholds will not remain the same throughout the globe. During south-west monsoon, the Indian region receives an average of 1,200 mm annual rainfall (Jain & Kumar 2012). Because rainfall has a high spatiotemporal variability, it is best to examine the severe precipitation features using Indian-specific criteria.
Analysing and understanding the precipitation variability is necessary for the rivers originating from Himalayan region as they are the source of water for one fifth of the total population (Immerzeel et al. 2020). It becomes vital to understand the effect of the behavioral pattern of rainfall on the rivers originating from Himalayas like Ghaghara, Rapti, Brahmaputra, Yamuna and others because in the Indian context, Himalayan mountains have an important role in controlling climate (Dimri 2014; Dimri et al. 2017). However, climatic unpredictability has had a considerable impact on the Himalayan area, providing difficulties for natural resource sustainability (Negi et al. 2016). Rapti River is a main tributary of Ghaghara River. Rapti River is prone to floods and flooding occurs as a result of excessive rainfall caused by the break-in-monsoon situation and the migration of low-pressure bands during the monsoon season (Kumar 2019). Floods in the basin occurred in 1992, 1998, 2000, 2008, 2014, 2017, 2018, 2019 and 2020. Frequent flooding in the basin had a tremendous impact on the lives of people. Rainfall becomes the major source of flooding in the basin; hence it becomes necessary to understand the variability of rainfall in the basin.
Therefore, this study aims to understand the rainfall characteristics of Rapti River Basin. This study includes analysis of daily, seasonal and annual rainfall of the gridded rainfall for Rapti Basin. It also focuses on examining the temporal variability of rainfall using ETCCDI indices, analysing trends in gridded rainfall, and understanding the rainfall characteristics of the Rapti River Basin. Present research focuses on determining the change point of the rainfall in the basin and relating the change point with flooding in the basin. The study highlights that flooding in the basin increases after the change point. Gorakhpur is the most flooded region of the basin since it lies in the downstream area of the basin, yet it has the maximum number of consecutive dry days. Rainfall in the upstream area leads to flooding in the Gorakhpur.
MATERIAL AND METHODS
Study area
Index map of study area.
The climate of Rapti Basin is sub-tropical, monsoonal. Winters (October to February) are cool and dry with occasional fogs and light showers; summers (March to early June) are hot and dry; and the monsoon season (mid-June to September) is warm and humid, with frequent heavy rainfall. The average annual rainfall in Rapti Basin is 1,093.44 mm. It can be seen that rainfall in the Rapti Basin is dominated by monsoon rainfall (87.5% of total rainfall) in the months of June to September, with July and August being the wetter months in which most of the rainfall occurs (65% of monsoon rainfall and 57% of total rainfall). The highest monthly mean maximum temperature is in the month of June (35.64 °C) while the lowest monthly mean minimum temperature is in the month of January (10.85 °C).
Data used
Daily precipitation data for the current study is derived from a coarse resolution (0.25° × 0.25°) gridded dataset made available by the Indian Meteorological Department (IMD), Pune, India (Pai et al. 2014). The gridded dataset is evolved using 6,955 rain gauges which are spread all over India. The aim of development of high resolution spatial data was to study climate change and its variability, rainfall-runoff modeling, etc.
METHODOLOGY
This section describes the trend detection methods used to determine the variation in rainfall of Rapti Basin.
Mann-Kendall test
is the number of tied groups in jth group. The statistic S is almost normally distributed if Z-transformation is determined using Equation (5).
Here, d indicates the slope and Xj and Xi are values at times i and j respectively, and n is the number of variables.
Pettitt test
Pettitt (1979) is the opportune procedure to find unforeseen alteration in time series data. Suppose, X1, X2, … Xt, is a set of random variables having a change point at position τ (Xt for t = 1, 2, … , τ) and follows a mutual distribution function F1(x) and Xt for t = τ + 1, … , T has a mutual distribution function F1(x) where both functions are not similar. It tests the null hypothesis (H0) that if the variable follows one or other distribution it has the same position constraint, i.e., no change point; while the alternative hypothesis (Ha) states that there exists a change point.
has statistically significant results at p ≤ 0.05 and is obtained using Equation (11) where p is the associated probability,
Modified Mann-Kendall test
is substituted by 
. Later, Equation (5) is used to calculate z which relies on the value of S.Sen's slope estimator
) is estimated from Equation (14), which is developed by Sen (1968) and Theil (1950).
Innovative trend analysis
Precipitation indices for the evaluation of rainfall characteristics
In the present study, nine indices given by ETCCDI are used to evaluate rainfall characteristics in Rapti Basin. Threshold of 1 mm is used as wet days. The criteria of wet days are selected as per definition given by ETCCDI. Descriptions of the indices used are given in Table 1.
Description of precipitation indices
| Sr no. | Extreme precipitation indices | Description |
|---|---|---|
| 1 | Consecutive dry days (CDD) | Maximum number of days that have passed with less than 1 mm of rain |
| 2 | Consecutive wet days (CWD) | The greatest number of days in a row with rainfall of at least 1 mm |
| 3 | Maximum 1-day rainfall (R × 1Day) | Maximum of daily precipitation in a year |
| 4 | Maximum 5-day rainfall (R × 5Day) | Maximum consecutive 5 days of rainfall |
| 5 | Number of rainy days (N × Rainy) | Annual count of days when rainfall is greater than or equal to 1 mm |
| 6 | Total precipitation in rainy days (PRCPTOT) | Sum of the wet days in a year |
| 7 | Simple daily intensity index (SDII) | Average intensity of daily rainfall in a year. It is the ratio of total precipitation in wet days (PRCPTOT) to the total number of wet days (N × Rainy) |
| 8 | Days with more than 10 mm rainfall (R10mm) | Days with rain totalling more than or equal to 10 mm |
| 9 | Days with more than 20 mm rainfall (R20mm) | Days with rain totalling more than or equal to 20 mm |
| Sr no. | Extreme precipitation indices | Description |
|---|---|---|
| 1 | Consecutive dry days (CDD) | Maximum number of days that have passed with less than 1 mm of rain |
| 2 | Consecutive wet days (CWD) | The greatest number of days in a row with rainfall of at least 1 mm |
| 3 | Maximum 1-day rainfall (R × 1Day) | Maximum of daily precipitation in a year |
| 4 | Maximum 5-day rainfall (R × 5Day) | Maximum consecutive 5 days of rainfall |
| 5 | Number of rainy days (N × Rainy) | Annual count of days when rainfall is greater than or equal to 1 mm |
| 6 | Total precipitation in rainy days (PRCPTOT) | Sum of the wet days in a year |
| 7 | Simple daily intensity index (SDII) | Average intensity of daily rainfall in a year. It is the ratio of total precipitation in wet days (PRCPTOT) to the total number of wet days (N × Rainy) |
| 8 | Days with more than 10 mm rainfall (R10mm) | Days with rain totalling more than or equal to 10 mm |
| 9 | Days with more than 20 mm rainfall (R20mm) | Days with rain totalling more than or equal to 20 mm |
RESULTS AND DISCUSSION
Trend results of rainfall using MK and MMK test
Z values for MK and MMK test: (a) daily rainfall, (b) annual rainfall, (c) monsoon season, (d) post monsoon season, (e) winter season, (f) summer season, (g) annual maximum rainfall.
Z values for MK and MMK test: (a) daily rainfall, (b) annual rainfall, (c) monsoon season, (d) post monsoon season, (e) winter season, (f) summer season, (g) annual maximum rainfall.
Pettitt test
The annual rainfall data for all the grids are assessed to determine the change point of the rainfall in the basin. Table 2 shows the grid-wise change point observed in the 30 years. The change point for most of the grids varies between 2005 and 2009. Of the 22 grids 10 have a change point between 2005 and 2009 indicating the change in pattern of rainfall over the Rapti Basin. Highest frequency was observed for year 2008 which is observed over six grids. Flooding in the basin has increased after occurrence of the change point. Therefore, 2008 is considered as a change point for the whole basin.
Change point year at different grids
| Grid | Change point | Grid | Change point |
|---|---|---|---|
| G1 | 1996 | G12 | 2008 |
| G2 | 2019 | G13 | 2001 |
| G3 | 2001 | G14 | 2007 |
| G4 | 2008 | G15 | 2009 |
| G5 | 2008 | G16 | 2008 |
| G6 | 1996 | G17 | 2018 |
| G7 | 2008 | G18 | 2018 |
| G8 | 2005 | G19 | 2008 |
| G9 | 2005 | G20 | 2017 |
| G10 | 2003 | G21 | 2017 |
| G11 | 2018 | G22 | 2018 |
| Grid | Change point | Grid | Change point |
|---|---|---|---|
| G1 | 1996 | G12 | 2008 |
| G2 | 2019 | G13 | 2001 |
| G3 | 2001 | G14 | 2007 |
| G4 | 2008 | G15 | 2009 |
| G5 | 2008 | G16 | 2008 |
| G6 | 1996 | G17 | 2018 |
| G7 | 2008 | G18 | 2018 |
| G8 | 2005 | G19 | 2008 |
| G9 | 2005 | G20 | 2017 |
| G10 | 2003 | G21 | 2017 |
| G11 | 2018 | G22 | 2018 |
Seasonal and annual precipitation trend
Results of ITA and MMK test for seasonal rainfall
| Annual rainfall | Summer season | Winter season | Monsoon season | Post monsoon season | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Grid | MMK | ITA | MMK | ITA | MMK | ITA | MMK | ITA | MMK | ITA |
| G1 | − 8.98 | − 11.9 | 1.24 | 1.593 | 0.017 | − 0.36 | − 13.2 | − 12.6 | − 0.05 | − 0.54 |
| G2 | − 0.13 | − 4.71 | 1.2 | 1.464 | − 0.02 | − 0.34 | − 7.14 | − 5.8 | − 0.024 | − 0.03 |
| G3 | − 10.5 | − 4.98 | − 0.501 | 0.1199 | − 0.33 | − 0.87 | − 7.3 | − 3.87 | − 0.03 | − 0.35 |
| G4 | − 9.59 | − 9.12 | 1.46 | 1.614 | − 0.46 | − 1.30 | − 11.1 | − 7.89 | − 0.07 | − 1.54 |
| G5 | − 8.86 | − 10.3 | 1.372 | 1.484 | − 0.405 | − 1.2 | − 7.99 | − 9.35 | − 0.129 | − 1.23 |
| G6 | − 8.4 | − 7.11 | 0.446 | 0.476 | − 0.005 | − 0.50 | − 7.36 | − 7.21 | 0 | 0.129 |
| G7 | − 8.32 | − 9.94 | 0.254 | 0.719 | − 0.371 | − 0.70 | − 6.46 | − 8.55 | − 0.419 | − 1.4 |
| G8 | − 20.5 | − 28.2 | 0.405 | 0.586 | − 0.05 | − 1.4 | − 19.6 | − 25.1 | 0.0025 | − 2.17 |
| G9 | − 6.73 | − 14 | 0.925 | 1.18 | − 0.106 | − 1.21 | − 6.78 | − 12.1 | 0 | − 1.89 |
| G10 | − 14.5 | − 15.6 | − 0.469 | − 1.19 | − 0.487 | − 1.11 | − 10.8 | − 12.1 | − 0.382 | − 1.21 |
| G11 | − 0.80 | 0.047 | 0.933 | 1.73 | − 0.02 | 0.051 | − 0.80 | − 2.28 | − 0.016 | − 0.54 |
| G12 | − 12.3 | − 15.6 | − 0.111 | 0.315 | − 0.631 | − 1.01 | − 10.2 | − 13.0 | − 0.63 | − 1.85 |
| G13 | − 14.6 | − 21.2 | 0.6 | 1.101 | − 0.188 | − 1.11 | − 13.2 | − 19.5 | − 0.237 | − 1.68 |
| G14 | − 2.35 | − 6.64 | 1.55 | 1.88 | − 0.097 | − 1.1 | − 1.06 | − 6.35 | − 0.015 | − 1.11 |
| G15 | − 5.01 | − 5.63 | 0.235 | − 0.155 | − 0.6 | − 1.08 | − 4.12 | − 3.93 | − 0.24 | − 0.45 |
| G16 | − 1.72 | − 0.62 | − 0.053 | 0.518 | − 0.467 | − 1.29 | 1.41 | 0.965 | − 0.083 | − 0.81 |
| G17 | 2.57 | 7.24 | 0.162 | 1.01 | − 0.091 | − 0.31 | 3.03 | 6.74 | 0 | − 0.19 |
| G18 | − 0.61 | 12.65 | − 0.038 | 1.76 | 0.0175 | − 0.22 | 1.255 | 11.48 | − 0.106 | − 0.30 |
| G19 | − 7.6 | − 6.51 | 0.2254 | 1.302 | 0.2254 | − 0.37 | − 6.30 | 6.39 | − 0.455 | − 1.05 |
| G20 | 2.96 | 2.33 | 0.6435 | 1.46 | − 0.22 | − 0.57 | 0.112 | 1.93 | − 0.05 | − 0.48 |
| G21 | 3.644 | 5.52 | − 0.09 | 0.90 | − 0.34 | − 0.82 | 5.31 | − 0.50 | − 0.014 | − 0.50 |
| G22 | 3.72 | 8.94 | − 0.172 | 0.831 | − 0.036 | − 0.32 | 4.647 | 8.62 | 0 | − 0.18 |
| Annual rainfall | Summer season | Winter season | Monsoon season | Post monsoon season | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Grid | MMK | ITA | MMK | ITA | MMK | ITA | MMK | ITA | MMK | ITA |
| G1 | − 8.98 | − 11.9 | 1.24 | 1.593 | 0.017 | − 0.36 | − 13.2 | − 12.6 | − 0.05 | − 0.54 |
| G2 | − 0.13 | − 4.71 | 1.2 | 1.464 | − 0.02 | − 0.34 | − 7.14 | − 5.8 | − 0.024 | − 0.03 |
| G3 | − 10.5 | − 4.98 | − 0.501 | 0.1199 | − 0.33 | − 0.87 | − 7.3 | − 3.87 | − 0.03 | − 0.35 |
| G4 | − 9.59 | − 9.12 | 1.46 | 1.614 | − 0.46 | − 1.30 | − 11.1 | − 7.89 | − 0.07 | − 1.54 |
| G5 | − 8.86 | − 10.3 | 1.372 | 1.484 | − 0.405 | − 1.2 | − 7.99 | − 9.35 | − 0.129 | − 1.23 |
| G6 | − 8.4 | − 7.11 | 0.446 | 0.476 | − 0.005 | − 0.50 | − 7.36 | − 7.21 | 0 | 0.129 |
| G7 | − 8.32 | − 9.94 | 0.254 | 0.719 | − 0.371 | − 0.70 | − 6.46 | − 8.55 | − 0.419 | − 1.4 |
| G8 | − 20.5 | − 28.2 | 0.405 | 0.586 | − 0.05 | − 1.4 | − 19.6 | − 25.1 | 0.0025 | − 2.17 |
| G9 | − 6.73 | − 14 | 0.925 | 1.18 | − 0.106 | − 1.21 | − 6.78 | − 12.1 | 0 | − 1.89 |
| G10 | − 14.5 | − 15.6 | − 0.469 | − 1.19 | − 0.487 | − 1.11 | − 10.8 | − 12.1 | − 0.382 | − 1.21 |
| G11 | − 0.80 | 0.047 | 0.933 | 1.73 | − 0.02 | 0.051 | − 0.80 | − 2.28 | − 0.016 | − 0.54 |
| G12 | − 12.3 | − 15.6 | − 0.111 | 0.315 | − 0.631 | − 1.01 | − 10.2 | − 13.0 | − 0.63 | − 1.85 |
| G13 | − 14.6 | − 21.2 | 0.6 | 1.101 | − 0.188 | − 1.11 | − 13.2 | − 19.5 | − 0.237 | − 1.68 |
| G14 | − 2.35 | − 6.64 | 1.55 | 1.88 | − 0.097 | − 1.1 | − 1.06 | − 6.35 | − 0.015 | − 1.11 |
| G15 | − 5.01 | − 5.63 | 0.235 | − 0.155 | − 0.6 | − 1.08 | − 4.12 | − 3.93 | − 0.24 | − 0.45 |
| G16 | − 1.72 | − 0.62 | − 0.053 | 0.518 | − 0.467 | − 1.29 | 1.41 | 0.965 | − 0.083 | − 0.81 |
| G17 | 2.57 | 7.24 | 0.162 | 1.01 | − 0.091 | − 0.31 | 3.03 | 6.74 | 0 | − 0.19 |
| G18 | − 0.61 | 12.65 | − 0.038 | 1.76 | 0.0175 | − 0.22 | 1.255 | 11.48 | − 0.106 | − 0.30 |
| G19 | − 7.6 | − 6.51 | 0.2254 | 1.302 | 0.2254 | − 0.37 | − 6.30 | 6.39 | − 0.455 | − 1.05 |
| G20 | 2.96 | 2.33 | 0.6435 | 1.46 | − 0.22 | − 0.57 | 0.112 | 1.93 | − 0.05 | − 0.48 |
| G21 | 3.644 | 5.52 | − 0.09 | 0.90 | − 0.34 | − 0.82 | 5.31 | − 0.50 | − 0.014 | − 0.50 |
| G22 | 3.72 | 8.94 | − 0.172 | 0.831 | − 0.036 | − 0.32 | 4.647 | 8.62 | 0 | − 0.18 |
Slope of ITA and MMK test: (a) ITA of annual rainfall, (b) MMK of annual rainfall, (c) ITA of monsoon rainfall, (d) MMK of monsoon rainfall, I ITA of post monsoon rainfall, (f) MMK of post monsoon rainfall, (g) ITA of summer rainfall, (h) MMK of summer rainfall, (i) ITA of winter rainfall, (j) MMK of winter rainfall, (k) ITA of annual maximum rainfall and (l) MMK of annual maximum rainfall
Slope of ITA and MMK test: (a) ITA of annual rainfall, (b) MMK of annual rainfall, (c) ITA of monsoon rainfall, (d) MMK of monsoon rainfall, I ITA of post monsoon rainfall, (f) MMK of post monsoon rainfall, (g) ITA of summer rainfall, (h) MMK of summer rainfall, (i) ITA of winter rainfall, (j) MMK of winter rainfall, (k) ITA of annual maximum rainfall and (l) MMK of annual maximum rainfall
Parajuli et al. (2021) have analysed monthly and annual precipitation trends in Ganga-Brahmaputra Basin and it was found that annual precipitation was declining at the rate of 5.8 mm/year. Praveen et al. (2020) have analysed rainfall trends for Indian States and it was found for east Uttar Pradesh that annual and seasonal rainfall is declining while in west Uttar Pradesh, summer rainfall has shown an increasing trend. Bera (2017) conducted MK test on 236 districts of Ganga Basin and found that 39 districts were showing a significant decreasing trend. All the seasonal along with annual precipitation has shown a decreasing trend. Due to the change in climate, the basin is receiving less rainfall and hence annual rainfall is declining at the rate of −28.16 mm/year in some parts of the basin. Annual and monsoon rainfall is showing a decreasing trend indicating a decrease in the number of rainy days in the monsoon season. In contrast, post monsoon and summer season are showing increasing trends but overall rainfall in the basin is decreasing.
Analysis of extreme precipitation indices
Analysis of precipitation indices: (a) average CDD, (b) average CWD, (c) average PRPTOT, (d) average R × 1, (e) average R × 5, (f) average ×10, (g) average R × 20 average R ≥ 1 and (i) average SDII
Analysis of precipitation indices: (a) average CDD, (b) average CWD, (c) average PRPTOT, (d) average R × 1, (e) average R × 5, (f) average ×10, (g) average R × 20 average R ≥ 1 and (i) average SDII
Different timescales are analysed like annual rainfall, annual max, winter rainfall, summer rainfall, monsoon and post monsoon rainfall. The variation pattern is different for different timescales due to climatic variability. Different parts of the basin receive different amounts of rainfall due to the vast area covered by the basin. Rapti River receives maximum rainfall in the monsoon season; a difference in magnitude is observed for different seasons, i.e., different timescales. Due to change in weather conditions, high temperature is observed in summer and thus a difference in rainfall pattern is observed. Hence, a variation pattern is observed in different timescales.
Results of trend of rainfall indices
Trend analysis of rainfall indices
| CDD | CWD | SDII | R1 | R nn | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Grid | Z | SS | Z | SS | Z | SS | Z | SS | Z | SS |
| G1 | − 0.48 | − 0.33 | − 1.22 | − 0.08 | − 1.91 | − 0.18 | − 0.89 | − 0.76 | − 0.32 | − 0.10 |
| G2 | − 0.89 | − 0.80 | 0.10 | 0.00 | − 2.92 | − 0.23 | − 1.81 | − 0.97 | 1.55 | 0.43 |
| G3 | 1.08 | 1.00 | − 1.23 | − 0.07 | − 1.08 | − 0.07 | − 1.58 | − 1.07 | − 0.52 | − 0.17 |
| G4 | − 0.71 | − 0.42 | − 0.51 | 0.00 | − 1.78 | − 0.16 | − 0.21 | − 0.11 | 0.78 | 0.20 |
| G5 | 1.07 | 0.64 | 0.00 | 0.00 | − 0.87 | − 0.07 | 0.66 | 0.53 | − 0.57 | − 0.09 |
| G6 | 0.146 | 0.25 | − 0.08 | 0.00 | − 0.15 | − 0.02 | − 0.06 | − 0.07 | − 0.55 | − 0.22 |
| G7 | 1.15 | 1.00 | 0.03 | 0.00 | − 1.24 | − 0.10 | 0.51 | 0.29 | 0.00 | 0.00 |
| G8 | 0.064 | 0.06 | 0.00 | 0.00 | − 3.01 | − 0.30 | − 2.04 | − 1.74 | 0.45 | 0.16 |
| G9 | 1.45 | 1.02 | − 0.73 | 0.00 | − 0.97 | − 0.07 | 0.71 | 0.32 | − 1.02 | − 0.29 |
| G10 | 1.86 | 1.63 | − 0.83 | − 0.04 | − 1.33 | − 0.07 | − 1.81 | − 0.77 | − 2.09 | − 0.66 |
| G11 | 0.73 | 0.40 | − 1.07 | − 0.07 | 0.75 | 0.05 | − 0.02 | − 0.07 | − 1.22 | − 0.30 |
| G12 | 0.567 | 0.58 | 0.07 | 0.00 | − 1.42 | − 0.14 | − 0.28 | − 0.26 | − 0.65 | − 0.14 |
| G13 | 0.616 | 0.49 | − 0.92 | − 0.06 | − 2.06 | − 0.14 | − 0.47 | − 0.40 | − 0.37 | − 0.09 |
| G14 | 0.427 | 0.66 | − 0.91 | − 0.02 | 0.96 | 0.06 | 0.86 | 0.60 | − 1.82 | − 0.40 |
| G15 | 2.23 | 1.91 | − 0.81 | − 0.06 | 0.34 | 0.03 | − 0.47 | − 0.52 | − 1.78 | − 0.59 |
| G16 | 0.113 | 0.06 | − 0.13 | 0.00 | 0.15 | 0.01 | 0.31 | 0.21 | − 1.33 | − 0.21 |
| G17 | 0.908 | 0.70 | − 0.49 | 0.00 | 1.20 | 0.07 | 0.63 | 0.36 | − 1.21 | − 0.30 |
| G18 | 0.468 | 0.41 | − 0.03 | 0.00 | 0.39 | 0.05 | 0.70 | 0.51 | − 0.33 | − 0.05 |
| G19 | 1.31 | 1.17 | 0.10 | 0.00 | − 1.13 | − 0.09 | − 0.11 | − 0.10 | − 1.02 | − 0.22 |
| G20 | 1.396 | 0.64 | − 1.54 | − 0.07 | 2.89 | 0.25 | 1.05 | 0.60 | − 2.74 | − 0.79 |
| G21 | 0.194 | 0.05 | − 1.66 | − 0.13 | 2.52 | 0.13 | 1.07 | 0.65 | − 1.62 | − 0.43 |
| G22 | 0.503 | 0.32 | − 0.63 | − 0.04 | 2.01 | 0.12 | 1.29 | 0.70 | − 1.47 | − 0.31 |
| CDD | CWD | SDII | R1 | R nn | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Grid | Z | SS | Z | SS | Z | SS | Z | SS | Z | SS |
| G1 | − 0.48 | − 0.33 | − 1.22 | − 0.08 | − 1.91 | − 0.18 | − 0.89 | − 0.76 | − 0.32 | − 0.10 |
| G2 | − 0.89 | − 0.80 | 0.10 | 0.00 | − 2.92 | − 0.23 | − 1.81 | − 0.97 | 1.55 | 0.43 |
| G3 | 1.08 | 1.00 | − 1.23 | − 0.07 | − 1.08 | − 0.07 | − 1.58 | − 1.07 | − 0.52 | − 0.17 |
| G4 | − 0.71 | − 0.42 | − 0.51 | 0.00 | − 1.78 | − 0.16 | − 0.21 | − 0.11 | 0.78 | 0.20 |
| G5 | 1.07 | 0.64 | 0.00 | 0.00 | − 0.87 | − 0.07 | 0.66 | 0.53 | − 0.57 | − 0.09 |
| G6 | 0.146 | 0.25 | − 0.08 | 0.00 | − 0.15 | − 0.02 | − 0.06 | − 0.07 | − 0.55 | − 0.22 |
| G7 | 1.15 | 1.00 | 0.03 | 0.00 | − 1.24 | − 0.10 | 0.51 | 0.29 | 0.00 | 0.00 |
| G8 | 0.064 | 0.06 | 0.00 | 0.00 | − 3.01 | − 0.30 | − 2.04 | − 1.74 | 0.45 | 0.16 |
| G9 | 1.45 | 1.02 | − 0.73 | 0.00 | − 0.97 | − 0.07 | 0.71 | 0.32 | − 1.02 | − 0.29 |
| G10 | 1.86 | 1.63 | − 0.83 | − 0.04 | − 1.33 | − 0.07 | − 1.81 | − 0.77 | − 2.09 | − 0.66 |
| G11 | 0.73 | 0.40 | − 1.07 | − 0.07 | 0.75 | 0.05 | − 0.02 | − 0.07 | − 1.22 | − 0.30 |
| G12 | 0.567 | 0.58 | 0.07 | 0.00 | − 1.42 | − 0.14 | − 0.28 | − 0.26 | − 0.65 | − 0.14 |
| G13 | 0.616 | 0.49 | − 0.92 | − 0.06 | − 2.06 | − 0.14 | − 0.47 | − 0.40 | − 0.37 | − 0.09 |
| G14 | 0.427 | 0.66 | − 0.91 | − 0.02 | 0.96 | 0.06 | 0.86 | 0.60 | − 1.82 | − 0.40 |
| G15 | 2.23 | 1.91 | − 0.81 | − 0.06 | 0.34 | 0.03 | − 0.47 | − 0.52 | − 1.78 | − 0.59 |
| G16 | 0.113 | 0.06 | − 0.13 | 0.00 | 0.15 | 0.01 | 0.31 | 0.21 | − 1.33 | − 0.21 |
| G17 | 0.908 | 0.70 | − 0.49 | 0.00 | 1.20 | 0.07 | 0.63 | 0.36 | − 1.21 | − 0.30 |
| G18 | 0.468 | 0.41 | − 0.03 | 0.00 | 0.39 | 0.05 | 0.70 | 0.51 | − 0.33 | − 0.05 |
| G19 | 1.31 | 1.17 | 0.10 | 0.00 | − 1.13 | − 0.09 | − 0.11 | − 0.10 | − 1.02 | − 0.22 |
| G20 | 1.396 | 0.64 | − 1.54 | − 0.07 | 2.89 | 0.25 | 1.05 | 0.60 | − 2.74 | − 0.79 |
| G21 | 0.194 | 0.05 | − 1.66 | − 0.13 | 2.52 | 0.13 | 1.07 | 0.65 | − 1.62 | − 0.43 |
| G22 | 0.503 | 0.32 | − 0.63 | − 0.04 | 2.01 | 0.12 | 1.29 | 0.70 | − 1.47 | − 0.31 |
Trend analysis of rainfall indices
| R5 | R10 | R20 | PRCPTOT | |||||
|---|---|---|---|---|---|---|---|---|
| Grid | Z | SS | Z | SS | Z | SS | Z | SS |
| G1 | − 1.28 | − 2.51 | − 1.42 | − 0.26 | − 1.36 | − 0.13 | − 1.44 | − 9.06 |
| G2 | − 2.18 | − 1.81 | − 0.91 | − 0.14 | − 1.19 | − 0.15 | − 0.98 | − 6.63 |
| G3 | − 0.94 | − 1.70 | − 0.56 | − 0.10 | − 1.12 | − 0.11 | − 1.31 | − 10.5 |
| G4 | − 0.66 | − 1.11 | − 0.91 | − 0.17 | − 1.28 | − 0.15 | − 1.31 | − 9.21 |
| G5 | − 0.11 | − 0.13 | − 1.26 | − 0.22 | − 1.39 | − 0.17 | − 1.37 | − 9.12 |
| G6 | 0.15 | 0.48 | − 1.21 | − 0.26 | − 0.80 | − 0.13 | − 0.76 | − 7.39 |
| G7 | 0.21 | 0.28 | − 0.75 | − 0.15 | − 1.71 | − 0.20 | − 0.99 | − 8.35 |
| G8 | − 1.76 | − 2.62 | − 2.19 | − 0.40 | − 2.73 | − 0.41 | − 2.57 | − 20.2 |
| G9 | − 0.08 | − 0.20 | − 1.15 | − 0.20 | − 1.13 | − 0.17 | − 1.24 | − 6.42 |
| G10 | − 1.39 | − 1.86 | − 1.62 | − 0.28 | − 1.86 | − 0.22 | − 1.96 | − 14.7 |
| G11 | − 0.28 | − 0.54 | − 0.40 | − 0.07 | − 0.50 | − 0.07 | − 0.11 | − 0.65 |
| G12 | − 0.21 | − 0.45 | − 1.99 | − 0.27 | − 1.48 | − 0.14 | − 1.60 | − 12.4 |
| G13 | − 1.05 | − 1.56 | − 1.91 | − 0.29 | − 2.60 | − 0.33 | − 2.44 | − 14.6 |
| G14 | 0.00 | 0.00 | − 0.31 | 0.00 | − 0.80 | − 0.10 | − 0.31 | − 2.22 |
| G15 | − 0.60 | − 0.99 | 0.13 | 0.00 | − 0.39 | 0.00 | − 0.73 | − 5.04 |
| G16 | 0.89 | 1.44 | − 0.26 | 0.00 | 0.55 | 0.07 | − 0.15 | − 1.61 |
| G17 | 0.79 | 1.48 | 0.37 | 0.07 | 1.04 | 0.14 | 0.41 | 2.60 |
| G18 | 0.08 | 0.25 | − 0.20 | 0.00 | − 0.22 | 0.00 | − 0.02 | − 0.67 |
| G19 | − 0.60 | − 0.98 | − 1.47 | − 0.24 | − 0.68 | − 0.09 | − 0.98 | − 7.32 |
| G20 | 0.73 | 1.20 | − 0.85 | − 0.09 | 1.17 | 0.11 | 0.79 | 3.25 |
| G21 | 1.07 | 0.99 | 0.00 | 0.00 | 1.34 | 0.11 | 0.92 | 3.33 |
| G22 | 1.37 | 1.54 | 0.94 | 0.12 | 0.44 | 0.06 | 0.57 | 3.88 |
| R5 | R10 | R20 | PRCPTOT | |||||
|---|---|---|---|---|---|---|---|---|
| Grid | Z | SS | Z | SS | Z | SS | Z | SS |
| G1 | − 1.28 | − 2.51 | − 1.42 | − 0.26 | − 1.36 | − 0.13 | − 1.44 | − 9.06 |
| G2 | − 2.18 | − 1.81 | − 0.91 | − 0.14 | − 1.19 | − 0.15 | − 0.98 | − 6.63 |
| G3 | − 0.94 | − 1.70 | − 0.56 | − 0.10 | − 1.12 | − 0.11 | − 1.31 | − 10.5 |
| G4 | − 0.66 | − 1.11 | − 0.91 | − 0.17 | − 1.28 | − 0.15 | − 1.31 | − 9.21 |
| G5 | − 0.11 | − 0.13 | − 1.26 | − 0.22 | − 1.39 | − 0.17 | − 1.37 | − 9.12 |
| G6 | 0.15 | 0.48 | − 1.21 | − 0.26 | − 0.80 | − 0.13 | − 0.76 | − 7.39 |
| G7 | 0.21 | 0.28 | − 0.75 | − 0.15 | − 1.71 | − 0.20 | − 0.99 | − 8.35 |
| G8 | − 1.76 | − 2.62 | − 2.19 | − 0.40 | − 2.73 | − 0.41 | − 2.57 | − 20.2 |
| G9 | − 0.08 | − 0.20 | − 1.15 | − 0.20 | − 1.13 | − 0.17 | − 1.24 | − 6.42 |
| G10 | − 1.39 | − 1.86 | − 1.62 | − 0.28 | − 1.86 | − 0.22 | − 1.96 | − 14.7 |
| G11 | − 0.28 | − 0.54 | − 0.40 | − 0.07 | − 0.50 | − 0.07 | − 0.11 | − 0.65 |
| G12 | − 0.21 | − 0.45 | − 1.99 | − 0.27 | − 1.48 | − 0.14 | − 1.60 | − 12.4 |
| G13 | − 1.05 | − 1.56 | − 1.91 | − 0.29 | − 2.60 | − 0.33 | − 2.44 | − 14.6 |
| G14 | 0.00 | 0.00 | − 0.31 | 0.00 | − 0.80 | − 0.10 | − 0.31 | − 2.22 |
| G15 | − 0.60 | − 0.99 | 0.13 | 0.00 | − 0.39 | 0.00 | − 0.73 | − 5.04 |
| G16 | 0.89 | 1.44 | − 0.26 | 0.00 | 0.55 | 0.07 | − 0.15 | − 1.61 |
| G17 | 0.79 | 1.48 | 0.37 | 0.07 | 1.04 | 0.14 | 0.41 | 2.60 |
| G18 | 0.08 | 0.25 | − 0.20 | 0.00 | − 0.22 | 0.00 | − 0.02 | − 0.67 |
| G19 | − 0.60 | − 0.98 | − 1.47 | − 0.24 | − 0.68 | − 0.09 | − 0.98 | − 7.32 |
| G20 | 0.73 | 1.20 | − 0.85 | − 0.09 | 1.17 | 0.11 | 0.79 | 3.25 |
| G21 | 1.07 | 0.99 | 0.00 | 0.00 | 1.34 | 0.11 | 0.92 | 3.33 |
| G22 | 1.37 | 1.54 | 0.94 | 0.12 | 0.44 | 0.06 | 0.57 | 3.88 |
Variability of trend in rainfall indices: (a) slope obtained in CDD, (b) slope obtained in CWD, (c) slope obtained in PRCPTOT, (d) slope obtained in R × 1, (e) slope obtained in R × 5, (f) slope obtained in R × 10, (g) slope obtained in R × 20, (h) slope obtained in R ≥ 1 and (i) slope obtained in SDII
Variability of trend in rainfall indices: (a) slope obtained in CDD, (b) slope obtained in CWD, (c) slope obtained in PRCPTOT, (d) slope obtained in R × 1, (e) slope obtained in R × 5, (f) slope obtained in R × 10, (g) slope obtained in R × 20, (h) slope obtained in R ≥ 1 and (i) slope obtained in SDII
Flood event analysis
From the Pettitt test, the change point of the basin is obtained at 2008. Flooding in the basin has increased after 2008. The threshold value specified by IMD is 64.4 mm for heavy rainfall days. A day on which more than 64.4 mm of rainfall has occurred is considered as heavy rainfall day. Hence, analysis has been carried out to identify flood events after the change point and heavy rainfall day determination before and after the change point. Average heavy rainfall days are decreased in the basin from 2.77 to 2.24 days in a year after the change point. Flood events of 2014 and 2017 in Rapti Basin are analysed and number of heavy rainfall days which led to flooding in the basin are identified. The aim of carrying out this analysis is to validate the threshold value of heavy rainfall for Rapti Basin. This will help in forecasting future floods in the basin which will help in preventing damages and loss of life in the Rapti Basin.
Flood of 2014
Flood hydrograph of (a) 2014 and (b) 2017.
Flood of 2017
The 2017 floods in the Rapti River basin were caused by the heavy rainfall during 12–14 August due to the break-in-monsoon system. The highest rainfall of 303.6 mm was recorded at Kakarahi G/D site on 14 August. Overall, the high rainfall was recorded at all the G/D sites of the basin on 14 August. Due to the high water level in Rapti, Burhi Rapti, Gaura and Rohini River, breaches in embankments occurred during the 2017 floods. Heavy rainfall led to flooding in the catchment. In the watershed, the flood duration spanned from 3 to 18 days from upstream to downstream respectively. Due to the decrease in slope and tributary flows, the period of the flood unveiled an increased flooding time duration from upstream to downstream (Kumar 2018). The extreme downstream Birdghat G/D site recorded the highest flood duration of 18 days from 15 August to 1 September 2017. Figure 6(b) shows the flood event of 2017 where discharge at Birdghat G.S. is plotted with average gridded rainfall of the basin during that period. Daily rainfall and discharge from 12 August 2017 to 31 August 2017 is plotted. There was heavy rainfall on 13 and 14 August, with maximum rainfall on 13 August of 164.2 mm at grid 14 and, on 14 August 202 mm rainfall was observed at grid 15. Grid 14 and 15 correspond to Maharajganj district of the Rapti Basin. On 14 August 2017, 17 out of 19 grid points received rainfall greater than 64.4 mm.
The present study discussed rainfall characteristics of the Rapti River Basin for different seasons, annual rainfall and also for rainfall indices. Different trends are observed for different seasons.
CONCLUSIONS
This study was carried out to determine the spatial and temporal variation in the rainfall pattern of Rapti River. It becomes crucial to comprehend the impact of climate change as rainfall is the main cause of flooding in the basin. Hence, trend analysis was carried out in the present study to understand the variability. MK test, MMK test and ITA were carried out to understand the trend in the rainfall. To comprehend the extreme rainfall characteristics, rainfall indices were calculated given by ETCCDI. Trend of rainfall indices was determined to understand change patterns in extreme rainfall characteristics. Based on the results of this study, the following conclusions are drawn:
The change point for most of the grids varies between 2005 and 2009 for rainfall in the basin between 1990 and 2020. Ten out of 22 grids have change point between 2005 and 2009 indicating the change in pattern of rainfall over the Rapti Basin. Highest frequency was observed for year 2008 which is observed over six grids. Therefore, 2008 is considered as a change point for the whole basin.
Flooding in the basin has increased after occurrence of the change point. Before 2008, major floods occurred in 1992, 1998 and 2000. After 2008, flooding occurred in 2014, 2017, 2018, 2019 and 2020.
The flooding in the basin has increased after the year 2008. Flood events of 2014 and 2017 are analysed and it is found that flooding is caused by the rainfall occurring in the basin.
Peak discharge in the 2014 flood was 2,657.51 m3/s and on 14 August 2014, 9 out of 19 grids received rainfall greater than 64.4 mm. Peak discharge in the flood of 2017 was observed as 3,360 m3/s and 17 out of 19 grids received rainfall greater than 64.4 mm on 14 August 2017.
Annual and monsoon rainfall has shown decreasing trend over the basin which was determined using ITA and MMK test. Increasing trends are observed at Shrawasti district while maximum decreasing trend is observed in some parts of the Balrampur district. Summer and winter seasons showed decreasing trends over the basin. Post monsoon has shown a decreasing trend in some parts of Balrampur district and Basti district in the basin.
Gorakhpur district has the maximum number of CDD while all the other indices have the lowest values at Gorakhpur district. Gorakhpur receives least rainfall in the basin. While Sant Kabir Nagar receives maximum rainfall in the basin. Gorakhpur is the most and worst affected region of the Rapti Basin as it lies in the downstream part of the basin and due to flat terrain, flood duration in the basin increases from upstream to downstream.
Out of all precipitation indices, PRCPTOT and SDII have shown the most decreasing trend in the basin, while in some parts of the basin SDII has shown significant increasing trend. Additionally, the Maharajganj district has seen a significant rise in CDD values. Rainfall during the monsoon has been trending down.
Overall, this study provides a comprehensive indication of the rainfall that occurs across the Rapti Basin. The basin's rainfall has been trending downward. Monsoon season rainfall and annual rainfall have shown significant decreasing trends as compared to other seasons. Rainfall greater than 64.4 mm leads to flooding in the basin. The outcomes of this research will aid in the decision-makers' comprehension of rainfall variability of the catchment.
CODE AVAILABILITY
The Fortran and C codes (open-source) to read the gridded rainfall data are available from IMD (see Data availability statement, below).
AUTHOR CONTRIBUTIONS
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Urvashi Malani. The first draft of the manuscript was written by Urvashi Malani and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
DATA AVAILABILITY STATEMENT
All relevant data are available from an online repository or repositories: (https://www.imdpune.gov.in/cmpg/Griddata/Rainfall_25_NetCDF.html).
CONFLICT OF INTEREST
The authors declare there is no conflict.
%20cropped.png?versionId=7149)