sustainability

Article
Distributed Hydrological Model Based on Machine Learning
Algorithm: Assessment of Climate Change Impact on Floods
Zafar Iqbal 1 , Shamsuddin Shahid 1 , Tarmizi Ismail 1, * , Zulfaqar Sa’adi 1,2 , Aitazaz Farooque 3
and Zaher Mundher Yaseen 4,5,6

1 School of Civil Engineering, Faculty of Engineering, Universiti Teknologi Malaysia (UTM),

Johor Bahru 81310, Johor, Malaysia; [email protected] (Z.I.); [email protected] (S.S.);
[email protected] (Z.S.)
2 Centre for Environmental Sustainability and Water Security (IPASA), Research Institute for Sustainable
Environment (RISE), Universiti Teknologi Malaysia (UTM), Johor Bahru 81310, Johor, Malaysia
3 Faculty of Sustainable Design Engineering, University of Prince Edward Island,
Charlottetown, PE C1A4P3, Canada; [email protected]
4 Department of Earth Sciences and Environment, Faculty of Science and Technology, Universiti Kebangsaan
Malaysia, Bangi 43600, Selangor, Malaysia; [email protected]
5 USQ’s Advanced Data Analytics Research Group, School of Mathematics Physics and Computing, University
of Southern Queensland, Toowoomba, QLD 4350, Australia
6 New Era and Development in Civil Engineering Research Group, Scientific Research Center, Al-Ayen
University, Thi-Qar 64001, Iraq
* Correspondence: [email protected]

Abstract: Rapid population growth, economic development, land-use modifications, and climate
change are the major driving forces of growing hydrological disasters like floods and water stress.
Reliable flood modelling is challenging due to the spatiotemporal changes in precipitation intensity,
Citation: Iqbal, Z.; Shahid, S.; Ismail,
duration and frequency, heterogeneity in temperature rise and land-use changes. Reliable high-
T.; Sa’adi, Z.; Farooque, A.; Yaseen,
Z.M. Distributed Hydrological Model
resolution precipitation data and distributed hydrological model can solve the problem. This study
Based on Machine Learning aims to develop a distributed hydrological model using Machine Learning (ML) algorithms to
Algorithm: Assessment of Climate simulate streamflow extremes from satellite-based high-resolution climate data. Four widely used
Change Impact on Floods. bias correction methods were compared to select the best method for downscaling coupled model
Sustainability 2022, 14, 6620. intercomparison project (CMIP6) global climate model (GCMs) simulations. A novel ML-based
https://doi.org/10.3390/su14116620 distributed hydrological model was developed for modelling runoff from the corrected satellite
Academic Editors: Fadong Li, rainfall data. Finally, the model was used to project future changes in runoff and streamflow extremes
Gang Chen, Xinlin He and from the downscaled GCM projected climate. The Johor River Basin (JRB) in Malaysia was considered
Guang Yang as the case study area. The distributed hydrological model developed using ML showed Nash–
Sutcliffe efficiency (NSE) values of 0.96 and 0.78 and Root Mean Square Error (RMSE) of 4.01 and
Received: 9 April 2022
5.64 during calibration and validation. The simulated flow analysis using the model showed that
Accepted: 20 May 2022
the river discharge would increase in the near future (2020–2059) and the far future (2060–2099) for
Published: 28 May 2022
different Shared Socioeconomic Pathways (SSPs). The largest change in river discharge would be
Publisher’s Note: MDPI stays neutral for SSP-585. The extreme rainfall indices, such as Total Rainfall above 95th Percentile (R95TOT),
with regard to jurisdictional claims in
Total Rainfall above 99th Percentile (R99TOT), One day Max Rainfall (R × 1 day), Five-day Max
published maps and institutional affil-
Rainfall (R × 5 day), and Rainfall Intensity (RI), were projected to increase from 5% for SSP-119 to
iations.
37% for SSP-585 in the future compared to the base period. The results showed that climate change
and socio-economic development would cause an increase in the frequency of streamflow extremes,
causing larger flood events.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland. Keywords: satellite rainfall; distributed hydrological model; flood forecast; machine learning; rain-
This article is an open access article fall extremes
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).

Sustainability 2022, 14, 6620. https://doi.org/10.3390/su14116620 https://www.mdpi.com/journal/sustainability

Sustainability 2022, 14, 6620 2 of 30

1. Introduction
The hydrological system involves a complicated interaction between various compo-
nents [1]. The human interaction with some of these components has made it more intricate
over time [2,3]. Hence, the hydrological system is a dynamically complex system that
remained difficult to understand and a challenge to model due to its complexity [4]. Hy-
drological disasters like floods and water stress have become an every-year phenomenon
in many other countries across the globe [5]. Floods in a catchment are triggered when
precipitation becomes more than the storage and drainage capacity of the catchment [6,7].
Due to rapid population growth, economic development, land-use modifications, and cli-
mate change, many catchments across the world have become highly prone to hydrological
disasters [8,9]. This is particularly true for Malaysia, where land use and climate changes
are often mentioned as the responsible factors for the increased frequency and severity of
urban water scarcity and floods [10,11]. This has caused major concern among scientists
and policymakers in the context of global environmental changes.
The increase in atmospheric greenhouse gases (GHG) caused a significant rise in global
temperature [12]. The changes in precipitation patterns, including intensity, duration, and
frequency, have been recorded with the rise in temperature over the last few decades, re-
sulting in frequent hydrological extremes [13]. Water is the most important resource for the
survival of living beings [14]. Almost 80% of the world’s population lives under different
forms of water scarcity [15]. Increasing hydrological disasters may cause a quick depletion
of the available water resources [16]. The water management system needs to be advanced
with better management policy to attain sustainable development and management of
water resources to adapt to climate change [17]. This needs reliable information on climate
change projections and implications in catchment hydrological processes.
Rainfall–runoff models simulate the relationships between rainfall and the runoff
generated in a catchment [18]. Various methods and techniques have been developed to
simplify this complex relationship, ranging from a simple mathematical model to a complex
“black box” and physical models [19–21]. According to the methods used to develop
the relationship between rainfall and runoff, the models are categorized as empirical,
conceptual, and physical [22]. They are also categorized as lumped, semi-distributed, and
distributed models based on their ability to consider the spatial variability of catchment
properties. Devia et al. conducted a comparative study to compare various rainfall–
runoff models [22]. The study revealed that the empirical models require fewer input
data but are limited to a certain region or a boundary, whereas the conceptual models are
parametric. The parameters are catchment dependent, and thus, their derivations need
large hydrological and meteorological data [23]. The physical-based model establishes the
rainfall–runoff relationship based on the governing physical laws [24]. These models are
most accurate but suffer from scale-related issues and require extensive data [22]. Therefore,
they are considered the most complex rainfall–runoff models. The uncertainties associated
with extensive data and the parameters used to develop models are specific to the region,
making these models more time-consuming and site-specific [25].
In recent years, soft computing or machine learning (ML) methods, such as Artificial
Neural Network (ANN), Support Vector Regression (SVR), and Fuzzy Logic and Genetic
Algorithm (GA), have been employed to develop rainfall–runoff and other hydrological
applications [26–29]. However, these approaches cannot completely manage the dynam-
ics of hydrological processes because of their inherent limitations in the approaches [30].
Potential challenges also arise as these methods require long-term, continuous historical
records of hydrological and other variables [31,32]. Furthermore, many of these approaches
simplify the multi-factors and often make the nonlinear systems linear, reducing the simu-
lation accuracy [33,34]. The hybridization of ML and conventional physical or conceptual
model can improve the capability to model complex interactions. Such an approach also
can replicate the functional relationship between input and output by enhancing the origi-
nal methodologies by data processing, parameter estimation, and routing using machine
learning algorithms [35]. The application of such complex problem-solving methodologies
Sustainability 2022, 14, 6620 3 of 30

in hydrology and water resources can help to provide a technique for reliable simulation of
hydrological disasters, particularly water scarcity and floods, due to the changes in land
use driven by physical and socio-economic factors and climate. Incorporating quantitative
information on complex interactions of runoff with land use and climate can enhance the
model’s accuracy in simulating hydrological disasters [36].
The projection of water-related hazards in a catchment is very intricate due to the
complex relationship of climate and land use with various ecological and socio-economic
factors, including population growth, economic development, urbanization, and policy-
related factors, like water management strategies and legislation [37]. Therefore, reciprocat-
ing actual hydrological conditions using hydrological models is always challenging [38].
A hydrological model requires a lot of observed data and optimizing different parame-
ters [39]. The data availability or mismatch of any data leads to errors in simulation [40].
Therefore, the major challenge is finding the relationship among the water cycle compo-
nents that affect a system in various dimensions. Successful simulation of a hydrological
cycle using a dynamic approach can address hydrological modelling challenges. The
solution to this problem is extremely important for Malaysia, where rapid population
and economic growth along with climate and land-use changes have caused a significant
change in hydrological disasters. Consequently, a moderate dry spell often forces water
rationing, and moderate or extreme rainfall causes floods, especially in rapidly developing
urban catchments of Malaysia [13].
The influence of land-use changes, water consumption, temperature rise and ground-
water level causes changes in the hydrology of an area [12,41]. Deficiencies are found
in studying the impact of climate changes, which are (i) the effect of changes due to a
single component, (ii) statistical analysis of time series rather than assessing through a
hydrological model, and (iii) not using the updated data for the study. There is a need
to analyze the changes in hydrology with the combined effect of all such variables along
with the hypothetical climate scenarios based on long-term climate observation of the
specific region.
Modelling the dynamics of different factors individually and jointly can help under-
stand the complex nonlinear interrelations and interactions among different elements in
the complex physical, environmental, and behavioural systems [42]. The incorporation
of quantitative information on complex interactions of various factors can enhance the
prediction accuracy of the hydrological model to simulate hydrological disasters. It is
expected that the application of complex problem-solving methodologies in hydrology
and water resources will provide a reliable simulation of hydrological disasters, particu-
larly water scarcity and floods, due to the changes in land use, climate and other physical
and socio-economic aspects factors. Therefore, in this study, we develop an entity-based,
distributed hydrological model based on state-of-the-art machine learning algorithms to
incorporate various components of the environment to analyze the effect of climate and
land-use changes on the flood susceptibility in the Johor River basin, Malaysia. Further-
more, this study includes the projection of rainfall and flood extremes under various SSP
scenarios of CMIP6 GCM future projections.

2. Study Area and Data Description

2.1. Study Area
The study area opted in this research is Johor River Basin (JRB). It is situated in the
south-eastern part of the state of Johor. The total catchment area of JRB is approximately
1652 km2 . JRB is situated in Peninsular Malaysia (Figure 1), also known as West Malaysia.
West Malaysia covers 130,598 km2 and lies between latitudes of 1.20◦ –6.40◦ N and lon-
gitudes of 99.35◦ –104.20◦ E [43]. The JRB has undulating land with elevations ranging
up to 366 meters in height. The topography comprises forests and irregular mountains
sloping towards the South China Sea. The central and northern regions of JRB are covered
with swamps and natural forests; however, in the southern region, rubber and oil palm
plantations are the dominant landuse. Approximately 64% of the JRB have a slope angle
Sustainability 2022, 14, x FOR PEER REVIEW 4 of 31
Sustainability 2022, 14, 6620 4 of 30

plantations are the dominant landuse. Approximately 64% of the JRB have a slope angle
ranging ◦ [44]. The total length of JRB is 122.7 km with major tributaries of
ranging from
from 00 to
to 50
50° [44]. The total length of JRB is 122.7 km with major tributaries of
Penggeli
Penggeli River,
River, Linggiu
Linggiu River,
River,Sayong
SayongRiver,
River,Jengeli
JengeliRiver,
River,and
andBelitong
BelitongRiver
River[45].
[45].

Figure 1. Location of the study area.

Figure 1. Location of the study area.

Malaysia’s climate
Malaysia’s climate isis humid
humid and and hot
hot due
due to
to its
its proximity
proximity toto the
the equator.
equator. The
The region’s
region’s
rainforest climate is heavily influenced by Asian–Australian atmospheric
rainforest climate is heavily influenced by Asian–Australian atmospheric dynamics and dynamics and
land–sea interaction, varying topography, and monsoon winds [46]. The
land–sea interaction, varying topography, and monsoon winds [46]. The average daily tem- average daily
temperature
perature ranges
ranges between
between 21 and21 and 32 degrees
32 degrees Celsius,
Celsius, withwith an annual
an annual variation
variation of 3 de-
of 3 degrees
grees Celsius.
Celsius. The annual
The annual averageaverage
rainfall rainfall is approximately
is approximately 2000–40002000–4000
mm, withmm, with
150 to 200150 to
rainy
200 rainy days per year [47]. The regional precipitation distribution pattern
days per year [47]. The regional precipitation distribution pattern is determined by the is determined
by the combined
combined responseresponse
of localof local topography
topography and windandflow
wind flow direction.
direction.
Peninsular Malaysia experiences two seasons throughout throughout thethe year:
year: The Southwest
Southwest
Monsoon (SWM)
Monsoon (SWM)from
fromMayMaytotoAugust
Augustand andthe
theNortheast
NortheastMonsoon
Monsoon (NEM)
(NEM) from
from Novem-
November
berFebruary.
to to February. During
During NEM,NEM, extreme
extreme rainfall
rainfall events
events arecommon,
are common,but butthe
theweather
weather is is dry
during SWM.
SWM. Coastal
Coastalplaces
placesare areaffected
affectedbybythethe
NEM,
NEM, whilst higher
whilst altitude
higher areas
altitude are less
areas are
affected
less by the
affected by monsoon.
the monsoon. Peninsular
PeninsularMalaysia has humid
Malaysia has humidweather, withwith
weather, the highest pre-
the highest
cipitation recorded
precipitation recordedduring
duringthethe‘inter-monsoon
‘inter-monsoon period.’
period.’

2.2.
2.2. Data
Data Description
Description
River
River gauge
gaugedata
dataof
ofJRB
JRBwas
wascollected
collectedfrom the
from Department
the Departmentof of
Drainage andand
Drainage Irrigation
Irriga-
(DID)
tion (DID) Malaysia. Daily discharge data of the main tributary was used to calibrate and
Malaysia. Daily discharge data of the main tributary was used to calibrate and
validate
validate the
the model.
model. The
The details
details of
of the
the river
river gauge
gauge are
are given
given in
in Table
Table1.1.

Table 1. Description
Table 1. Description of
of River
River Flow
Flow data.
data.

Station ID Station Name River Basin Catchment

Catchment Area (km2 ) Analysis Period
Station ID Station Name River Basin Analysis Period
1130Area (km )
2
1737451 SG. JOHOR at RANTAU PANJANG Sg Johor 2007–2017
Sustainability 2022, 14, 6620 5 of 30

ERA-5-Land is a post-processed reanalysis of the European Commission, the European

Centre for Medium-Range Weather Forecasts (ECMWF). It has a higher resolution than the
previous products, such as the ERA-interim and ERA-5. This product provides complete
information on various climates and land variables over a longer period with higher
resolution. Physical laws govern the model output generated to produce a consistent set of
data by using observed data across the globe as an input. ERA-5 land contains the data of
50 variables which helps to study the energy and water cycles with a one-hour temporal
resolution spread globally at a 9km resolution spatial resolution [48]. ERA-5-Land data
contains over 50 climate variables. Downscaling the CMIP6 GCMs, long term continuous
higher-resolution data is required. Therefore rainfall and temperature data of ERA-5-Land
for the period 1981–2014 were used to downscale GCMs, whereas soil moisture was used for
the development of the distributed hydrological model. All these data are freely available
at https://cds.climate.copernicus.eu (accessed on 16 November 2021).
The CMIP6 GCMs were employed in this study for the simulation of the future runoff
in the basin. The new multi-model ensemble of CMIP6 was used in this study. The model
ensemble allows climate change evaluation and regional projections under various future
socio-economic scenarios. The fourth and fifth IPCC reports of the Erath System model
(ESM) and Atmosphere-Ocean General Circulation Models (AOGCMs) were coupled as
an input for CMIP6, which are known as General Circulation Model (GCM). The GCMs
were selected by Iqbal et al. [49] for (Mainland South East Asia) MSEA using a robust
selection method that uses the categorical and spatial indices. They selected three GCMs
for the region. These GCMs were downscaled in this study to use as an input to the
hydrological model to simulate future floods. Further details on the GCMs used can be
found in the article [49].
Integrated Multi-Satellite Retrievals for Global Precipitation Measurement (IMERG)
performed better among five other satellite products for precipitation over Peninsular
Malaysia. Iqbal et al. [50] bias corrected the IMERG data using rain gauge station data.
They developed a two-step bias correction method to improve the performance of IMERG
data up to 55% in RMSE. The method was also found better than the conventional bias
correction method such as Linear scaling and quantile regression [50]. The details of the
study and method can be found in Iqbal et al. [50]. This bias corrected IMERG dataset was
used in this study to develop the distributed hydrological model over Johor River Basin.
The details of various other datasets used in this modelling are given in Table 2.

Table 2. Details of various parameters used in this study.

Data Set Resolution Source

MODTBGA (MODIS/Terra Thermal Bands https://lpdaac.usgs.gov/
Land use 1 km
Daily L2G-Lite Global 1km SIN Grid V006 (accessed on 13 June 2021)
MOD16A2 (MODIS/Terra Net
https://lpdaac.usgs.gov/
Rainfall Evapotranspiration 8-Day L4 500 m
(accessed on 13 June 2021)
Global 500 m SIN Grid V006)
MOD11A1-MODIS/Terra Land Surface
https://lpdaac.usgs.gov/
Land Surface Temperature Temperature/Emissivity Daily L3 Global 1 km
(accessed on 14 June 2021)
1km SIN Grid
https://asf.alaska.edu/
Elevation ALOS/PALSAR DEM 12.5 m 12.5 m
(accessed on 19 July 2021)
https://soilgrids.org/
Soil Type SoilGrids250m version 2.0 250 m
(accessed on 22 July 2021)

3. Methodology
3.1. Procedure
The methodology adopted in this study consists of the following steps:
1. The catchment is divided into grids of 10 km each.
Sustainability 2022, 14, 6620 6 of 30

2. All the data sets are interpolated to 10 km to achieve a similar resolution.

3. The distributed hydrological model is developed using bias Corrected IMERG data
for the catchment.
4. The model is calibrated and validated with the observed river flow data (details given
in Table 1).
5. The selected GCMs are downscaled to 10 km resolution for the basin.
6. The downscaled GCMs data is used in the distributed model to simulate the future
flow condition under different SSP scenarios. The details of the methods used to
complete the analysis are given below.

3.2. K-Nearest Neighbour

The KNN is an efficient nonparametric classification algorithm that assigns data
to a class based on its nearest neighbours [51]. In the particular classification problem,
n oN
assuming that T = xn ∈ Rd indicates a training set comprises of N samples within
n =1
each M class in d-dimension; the sample xn is assigned the class mark “cn ”, the distance
between the unknown point x and xiNN is estimated using Euclidean distance method as
shown in Equation (1).
  q
T
d x, xiNN = x − xiNN x − xiNN


(1)

Next, the class name of the query point x is estimated based on the majority voting of
its neighbours, as shown in Equation (2).
 
ć = argmaxc ∑ δ c = ciNN (2)
( xiNN , ciNN )∈ T

where c is a class label and ciNN is the class label of i-th nearest neighbour. δ c = ciNN , an

indicator function, can have a value of one of the class ciNN of the neighbour xiNN . This
research used KNN to interpolate different data sets to a specific grid.

3.3. Downscaling of GCMs

GCM simulations were downscaled to a finer spatial scale for their use by end-users.
This study used the MOS approach to downscaling the selected GCMs into fine resolution.
In MOS, the statistical calibration between simulated and observed predictors is usually
done [52]. The advantage of using this method is that it improves reliability while keeping
the original accuracy. The MOS has been found more advantageous in studies related to
climate change [53].
The bias correction using MOS can be expressed as in Equations (3) and (4).

N
1
Bias =
N ∑ [ F(t) − O(t)] (3)
t =1

where F = Forecast; O = Observations and N = days in training sample.

F 0 = F (t) − Bias (4)

where F’= Corrected Value.

3.3.1. Gamma Quantile Mapping

Gamma Quantile mapping (GammaQM) was introduced by Piani et al. [54]. This
method assumes that a gamma distribution approximates the observed and simulated
intensity distribution well. GammaQM builds a model variable Pm using probability
Sustainability 2022, 14, 6620 7 of 30

integral transform to make the new build distribution equal to the distribution of the
observed variable Po . The mathematical expression of this method is given in Equation (5)

Po = F−1 o (Fm (Pm )) (5)

where Fm = Cumulative function of Pm ; and F−1 o = inverse cumulative function of Po .

x
e(− θ ) x (k−1)
pd f ( x ) = (6)
T (k)θ k

where in Equation (6), k signifies the form parameter, x denotes the Normalized daily
precipitation, while θ denotes the scaling parameter.
GammaQM could not be applied if the k value is less than 1 or 0; therefore, the value
is presumed greater than 1. GammaQM deliberates mean and extreme values, making it an
effective bias-correction method [54–56]. GammaQM is only valid for precipitation data.

3.3.2. Power Transformation

Power Transformation (PowerTr) considers the bias in the mean and the differences in
the variance for the correction of data [57]. In power transformation, a nonlinear correction
in the exponential form such as aPb is used for the adjustment of variance. According to
this method daily precipitation of P was transformed into a corrected amount of P* using
Equation (7).
P∗ = aPb (7)
A distribution-free approach can calculate the parameter, b. It is first identified by
matching the coefficient of variation (CV) corrected daily GCM precipitation (Pb ) with
the CV of observed daily precipitation for each month of training. The value of b was
determined iteratively. Data grouping was done every five days to reduce sampling
variability [58]. The value of b was used to calculate the transformed precipitation by
Equation (8):
P∗ = Pb (8)
A parameter is subjected to the observed and the transformed mean values. It is a
dependent parameter to the value of b parameter and subsequently b is a dependent to the
magnitude of CV Both a and b are differed for every block-annual of 5 days.

3.3.3. Generalized Quantile Mapping

Generalized quantile mapping (GenQM) is a kind of parametric quantile mapping. Its
main differ is the implementation of gamma distribution and generalized Pareto distribu-
tion (GPD). Mathmatically, GenQM can be expressed as follows:

Po = Fo−1 ( Fm( Pm)) (9)

The pdf is chnaged with the value of GPD and gamma distribution. The value of the
GPD is tailed the extreme distribution [59], as expressed in Equation(10).
  
1 − 1 + ξ x , i f ξ 6= 0
Pr( X − u ≤ x | X > u) =  eσ  (10)
 1 − exp − ξ x
σ ,
e if ξ = 0

Here, the u value is the 95th% threshold, e σ = σ + ξ( u − µ ), e

σ is the scale parameter. ξ
is presented the shape parameter. In this equation, gamma distribution was employed on a
smaller threshold. In addition, the GPD was employed on values larger than this threshold,
as given in Equation (11):
−1
(

Fobs,gamma FCCLM,gamma , if x < 95 thpercentile
y= −1 (11)
Fobs,GPD ( FCCLM,GPD ), if x ≥ 95 thpercentile
Sustainability 2022, 14, 6620 8 of 30

3.3.4. Linear Scaling

The Linear Scaling (LS) approach was introduced by Lenderink et al. [60]. This bias
correction method utilizes the monthly correction values “the difference modeled and
observed daily dataset”. The monthly scaling factor is then applied to uncorrected daily
data. The daily precipitation P was corrected by the following Equations (12) and (13).

P∗ = αP (12)

While the temperature, T, is corrected using the following equation,

T ∗ = αT (13)

whereas α is the monthly scaling factor for precipitation is calculated by Equation (14),

Po
α= (14)
Ps

Po is the observed monthly mean and Ps is the simulated monthly mean.

For the bias correction of temperature, the scaling factor is calculated by Equation (15).

α = To − Ts (15)

To is the observed temperature mean whereas, Ts is the monthly mean simulated tempera-
ture. The LS method is simple and requires less information, such as only monthly data is
required to calculate the scaling factor [61].

3.4. Hydrological Model Development

Hydrological interactions, such as transpiration, evaporation, streamflow, rainfall,
groundwater flow, and infiltration constitute a hydrological system. The interaction among
the hydrological system’s components is complex and variable in space and time. However,
four major components mostly govern the hydrological cycle: precipitation, infiltration,
runoff, and evapotranspiration. Various methods are adopted to develop the relationships
between these major hydrological components and understand the hydrology of any region.
The major interacting components of a hydrological cycle are shown in Figure 2. The study
is divided into several grid boxes to model the distributed nature of hydrological processes.
The divisions of the study area into grid boxes are also shown in Figure 2.

3.4.1. Concept of the Distributed Model

A multiple bucket modelling approach was used in this research to account for the
spatial variability of the land and climate variables in the catchment. The study area
was divided into grids of 10 km each to calculate the cumulative flow in the catchment,
considering the variability of soil and climate inputs at a coarser scale. Each bucket
was subdivided based on the major hydrological processes. The unsaturated region,
evapotranspiration, and surface runoff simulate the flow generation. The model was
developed for JRB on a daily time scale using the bias-corrected satellite data. A simple
water balance equation was used for each bucket, including rainfall, evaporation, and
saturated excess flow. A nonlinear storage discharge relationship was established to
generate surface runoff using near-real-time RS data for creating an early flood hazard
system to minimize the flood causalities.
Sustainability 2022,
Sustainability 2022, 14,
14, 6620
x FOR PEER REVIEW 99 of 31
of 30

Figure 2. Distribution of the study area into grids.

Figure 2. Distribution of the study area into grids.
The main water balance equation used to calculate the discharge from each bucket is
given in Equation
3.4.1. Concept (16).
of the Distributed Model
A multiple bucket modelling approach was used in this research to account for the
ds (t)
spatial variability of the land and R(t) − variables
=climate f ss (t) − f sein(tthe
) − catchment.
et(t) The study area was (16)
dt
divided into grids of 10 km each to calculate the cumulative flow in the catchment, con-
sidering
where R(the
t) isvariability
the amount of of
soil and climate
rainfall inputstimestep
at a certain at a coarser
t. qssscale. Each
(t) and qse (bucket was
t) are the sub-
runoff
divided
at basedand
subsurface on the major hydrological
a surface scale, whereas processes.
et representsThe unsaturated region, evapotran-
the evapotranspiration for the
spiration,bucket.
pertinent and surface runoff simulate the flow generation. The model was developed for
JRB on a daily time scale using the bias-corrected satellite data. A simple water balance
3.4.2. Excess
equation wasSaturation
used for eachRunoff Rate including rainfall, evaporation, and saturated excess
bucket,
flow.The
A nonlinear storage
overland flow dischargeexcess
or saturated relationship
flow of was established
a bucket to generate
was calculated surface run-
by Equation (17),
off using
given the near-real-time RS datawere
following conditions for creating
fulfilled.an early flood hazard system to minimize the
flood causalities.
The main water balance V − Vb
f se = equation i fused
V >toVbcalculate
f se = 0 the i f discharge
V < Vb from each bucket is
(17)
given in Equation (16). ∆t
where V is the volume of soil𝑑water
(𝑡) storage and Vb is the soil moisture storage capacity. Vb
depends on the average soil depth =(L)
𝑅(𝑡) 𝑓 (𝑡)
and−the − 𝑓 (𝑡)
average soil−porosity
𝑒𝑡(𝑡) (16)
𝑑 Φ.
Sustainability 2022, 14, 6620 10 of 30

3.4.3. Subsurface Runoff

The subsurface runoff is the function of soil storage and catchment response time. The
function to calculate the subsurface runoff depends on the values of soil water storage
capacity and the soil water storage at the pertinent grid, as given in Equation (18) below.

V − Vf
f ss = i f V > Vf f ss = 0 i f V < Vf (18)
tc

where V represents the soil water storage and Vf is the threshold storage assumed to be
equal to Vf = f c L, the product of soil field capacity and the average soil depth. Darcy’s Law
is used to calculate the catchment response time considering the hydraulic gradient equal
to the hillslope of the ground calculated using the DEM. The equation for the calculation of
catchment response time is given in Equation (19),

LΦ
tc = (19)
2 Ks tanβ

where L is the hillslope length, tanβ is the average ground surface slope, and Ks is the
average saturated hydraulic conductivity.

3.4.4. Evapotranspiration
The evapotranspiration in the water balance model is calculated using an empirical
relationship that uses minimum parameters. The FAO Blaney–Criddle was used in this
study to find the evapotranspiration using the precipitation and temperature at a specific
grid [62], using Equation (20) below.
 
eto(i) = p(i) 0.46 Tmean(i) + 8.13 (20)

where, p(i) is the average precipitation and Tmean(i) is the mean temperature of the grid (i).

3.4.5. Flow Routing

This distributed hydrological model’s flow routing relies on the “Eight Direction Pour
point model”. It calculates the direction of flow of a single grid based on the difference in
elevation of the surrounding eight grids. Furthermore, the flow direction of each cell is
determined using the “Direction of steepest descent” method. The steps to calculate the
flow routing are as follows:
1. The average elevation of each grid is calculated for all the cells.
2. The flow direction of each cell is calculated using the Eight Direction Pour point model.
3. The flow accumulation in each cell is calculated using the bucket model developed in
Section 3.3.1.
4. Flow accumulation is calculated by adding the cumulated flow of the grids flowing
into the particular grid
5. The flow route is calculated by connecting the low water accumulated cells with high
water accumulated cells.
A machine learning model was used to generate runoff from rainfall in each grid cell.
R-packages “r.watershed” tool and “rdwplus” were used to estimate the routing of the
generated runoff from the cell to the catchment outlet.

3.4.6. Projections of Climate Change Impacts on Hydrological Extremes

The framework to analyze the impact of climate changes on hydrological extremes
is shown in Figure 3. The extremes in the flow at Ratu Panjang station were simulated
for historical and future scenarios. To simulate the flow using the spatially distributed
hydrological model developed in this study, the downscaled GCMs data was used as an
input. The model generated the flow for historical and future scenarios for various SSPs.
 shown in Figure 3. The extremes in the flow at Ratu Panjang station were simulated for
historical and future scenarios. To simulate the flow using the spatially distributed hydro-
logical model developed in this study, the downscaled GCMs data was used as an input.
The model generated the flow for historical and future scenarios for various SSPs. The
Sustainability 2022, 14, 6620 output was used to calculate various flow quantiles, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8,
11and
of 30
0.9. The changes in each quantile were calculated by taking the difference between simu-
lated future flow to the historical flow by each GCM.
The output was used to calculate various flow quantiles, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8,
and 0.9. The changes in each quantile were calculated by taking the difference between
simulated future flow to the historical flow by each GCM.

Figure 3. Framework to analyze the impact of climate change on hydrological extremes.

The downscaled GCMs data was used to analyze various rainfall extremes, as given
in Table 3. These indices were calculated for historical and each SSP. The difference in these
indices was calculated with the historical period as the reference. The details of each index
are shown in Table 3.
Figure 3. Framework to analyze the impact of climate change on hydrological extremes.
Table 3. WMO hydrological extreme indices used in this study.
The downscaled GCMs data was used to analyze various rainfall extremes, as given
Indices inSymbol Description
Table 3. These indices were calculated for historical and each Formula
SSP. The difference in
Total rainfall above these indices wasAnnual
calculated with the historical
total rainfall when period as the
W reference. The details of each
R95pTOT R95p = ∑ RR ∗ where RR > RR 95
95th Percentile index are shown in Table 3.
rainfall > 95p wj wj wn
w =1
Total Rainfall above Annual total rainfall when W
Table 3. WMO hydrological extreme indices used in this
R99pTOT R99p study.
= ∑ RR∗wj where RRwj > RRwn 99
99th Percentile rainfall > 99p
w =1
Indices Symbol
Annual maximum
Description Formula
One day Max Rainfall × 1 day
RTotal rainfall Rx1day j = max( RRij∗∗ )
1-day rainfall
Annual total rainfall when 𝑅95𝑝 = ∑ 𝑅𝑅 ∗ where
above 95th R95pTOT
Five-day Max Rainfall R× 5 day
Annual maximumrainfall > 95p 𝑅𝑅
Rx5day j = max(>RR𝑅𝑅
∗∗ ) 95
Percentile 5-day rainfall ij

Average rainfall on the w

∑w=1 RRwj
Rainfall Intensity RI RI =
rainy days W ∗∗∗

* Daily Rainfall amount on wet days (Rainfall > 0). ** Daily rainfall amount on the day, i, in period j. *** Number
of wet days (Rainfall > 0).

4. Application Results
4.1. Downscaling of GCMs
4.1.1. Downscaling of Precipitation
ERA-5-Land data was used as the reference data to downscale the precipitation and
temperature of EC-Earth, EC-Earth-Veg, and MRI-ESM2. LS, GammaQM, PowerTr, and
GenQM were used to downscale the historical GCMs. Index of agreement (d), Normalized
Root Mean Square Error (NRMSE), Percentage Bias (Pbias), and Skill Score (SS) of the
downscaled precipitation are shown in Figure 4. The downscaled result of one out of three
GCMS are shown below. The results of the remaining GCMs are provided as supplementary
material for further reference (Figures S1, S2, S5, S6, S9 and S10). The results showed that
LS performs better than the other bias correction methods. Compared to the other methods,
LS has improved the d values by up to 20% for each GCM.
 Sustainability2022,
Sustainability 2022,14,
14,6620
x FOR PEER REVIEW 13 of30
12 of 31

Figure4.4.The
Figure Thestatistical
statisticalperformance
performanceofofthe
thedownscaled
downscaledand
andraw
rawGCMs.
GCMs.

The error
Taylor in downscaling
diagram was usedwas compared
to compare theusing
degreeNRMSE %. Plots between
of correspondence betweenERA-5 and
the bias-
GCM with different bias correction methods are shown in Figure 4. The results
corrected data, as shown in Figure 5. The figure shows the bias-corrected outputs for the show
that the NRMSE
EC-Earth duringofcalibration
raw GCMand ranges betweenperiods.
validation 120 andThe
130%, whereas
results showedthe that,
LS reduced
in termstheof
NRMSE by 100–110%. GammaQM, PowerTr, and GenQM showed poor performance
three statistical matrices (Standard Deviation, Correlation and RMSE), the LS method per- in
reducing the NRMSE.
formed better to reduce the bias in both the calibration and validation periods. The LS
The PBIAS
showed in the coefficient
a correlation bias-corrected outputs
higher is shown
than 0.4 duringin Figure 4. and
calibration Thevalidation,
results showed
while
that the raw
all other GCM showed
methods biases range
less than −40
from0.4. The −45% compared to
to root-mean-square ERA-5
error of LSdata. The most
corrected data
suitable model which reduced the bias very close to zero was the LS and PowerTr. The SS
was less than the other models, while the standard deviation was nearer to the observed
 Sustainability 2022, 14, 6620 13 of 30

of the raw and the bias-corrected GCMs were also compared to show the model’s accuracy.
The best SS was found for the LS. The raw GCM of EC-Earth showed a mean SS of 0.42,
whereas it improved to 0.63 using LS, 0.48 using GammaQM and 0.53 using PowerTr, and
it reduced to 0.25 for GenQM, as shown in Figure 4. Similar improvements were observed
for EC-Earth-Veg and MRI-ESM2 using the LS method.
Taylor diagram was used to compare the degree of correspondence between the
bias-corrected data, as shown in Figure 5. The figure shows the bias-corrected outputs
for the EC-Earth during calibration and validation periods. The results showed that,
in terms of three statistical matrices (Standard Deviation, Correlation and RMSE), the
LS method performed better to reduce the bias in both the calibration and validation
Sustainability 2022, 14, x FOR PEER REVIEW 14 of 31
periods. The LS showed a correlation coefficient higher than 0.4 during calibration and
validation, while all other methods showed less than 0.4. The root-mean-square error of
LS corrected data was less than the other models, while the standard deviation was nearer
one as it is radially nearer to the observation (hollow circle on the x-axis). The Taylor dia-
to the observed one as it is radially nearer to the observation (hollow circle on the x-axis).
grams of the remaining GCMs are provided as supplementary materials (Figures S3, S4,
The Taylor diagrams of the remaining GCMs are provided as supplementary materials
S7, S8, S11 and S12).
(Figures S3, S4, S7, S8, S11 and S12).

Figure
Figure 5. 5. Taylordiagram
Taylor diagramshows
showsbias
biascorrection
correction methods’
methods’ performance
performance to
to correct
correct EC-Earth
EC-Earthduring
during(a)
(a)calibration;
calibration;(b)
(b)validation
validationperiods.
periods.
4.1.2. Downscaling of Maximum Temperature
4.1.2. Downscaling of Maximum Temperature
The comparison of d in Figure 6 showed that the PowerTr downscaling method
The comparison of d in Figure 6 showed that the PowerTr downscaling method im-
improved the d value from 0.48 to 0.56, a comparison of NRMSE % values showed that
proved the d value from 0.48 to 0.56, a comparison of NRMSE % values showed that the
the GammaQM and GenQM methods failed to downscale the GCM because the NRMSE
GammaQM and GenQM methods failed to downscale the GCM because the NRMSE %
% values increased for these two models. However, LS and PowerTr showed a slight
values increasedinfor
improvement thethese
NRMSEtwo%models. However, LS and PowerTr showed a slight im-
by 5–10%.
provement in the NRMSE % by 5–10%.
The output of the downscaling models is compared in Figure 6. The average percent-
age biases in downscaled data compared to the ERA-5 ranged from 0.1 to 0.4. The results
showed that the PowerTr model reduced the biases in the three GCM by an average of
20−30%. Similar results were demonstrated by the LS method, whereas the GenQM and
GammaQM showed unsatisfactory performance in the bias correction. The SS of the mod-
els was also compared to the ERA-5 data, as shown in Figure 6. The SS of raw GCM was
0.995, which was further improved up to 0.999 by the PowerTr method in most cases.
However, the SS was reduced in the case of GammaQM and GenQM. Therefore, in terms
of improving these indices, the PowerTr downscaling method proved to be a better model
than the others.
 Sustainability 2022, 14, x FOR PEER REVIEW 15 of 31
Sustainability 2022, 14, 6620 14 of 30

Figure6.6.The
Figure Thestatistical
statisticalperformance
performanceof
ofthe
thedownscaled
downscaledand
andraw
rawGCMs
GCMsfor
formaximum
maximumtemperature.
tempera-
ture.
The output of the downscaling models is compared in Figure 6. The average per-
centage biases in downscaled data compared to the ERA-5 ranged from 0.1 to 0.4. The
results showed that the PowerTr model reduced the biases in the three GCM by an average
Sustainability 2022, 14, 6620 15 of 30

of 20–30%. Similar results were demonstrated by the LS method, whereas the GenQM
and GammaQM showed unsatisfactory performance in the bias correction. The SS of the
models was also compared to the ERA-5 data, as shown in Figure 6. The SS of raw GCM
Sustainability 2022, 14,0.995,
was which
x FOR PEER was further improved up to 0.999 by the PowerTr method in most cases.
REVIEW
However, the SS was reduced in the case of GammaQM and GenQM. Therefore, in terms
of improving these indices, the PowerTr downscaling method proved to be a better model
than the others.
Figure 7 shows theFigure 7 shows
Taylor diagramthefor
Taylor diagram
EC-Earth for The
GCM. EC-Earth GCM.and
calibration Thevalidation
calibration and va
period results showed that the best model to downscale
period results showed that the best model to downscale the EC-Earth is the PowerTrthe EC-Earth is the Powe
reduced the RMSE and increased the correlation. The results
as it reduced the RMSE and increased the correlation. The results of EC-Earth-Veg and of EC-Earth-Veg an
ESM2 are given
MRI-ESM2 are given in appendices. in appendices.

Figure 7. Taylor diagram shows bias correction methods’ performance to correct EC-Earth during (a)
Figure 7. Taylor diagram shows bias correction methods’ performance to correct EC-Eart
calibration; (b) validation period.
(a) calibration; (b) validation period.
4.1.3. Downscaling of Minimum Temperature
4.1.3. Downscaling of Minimum Temperature
The results for downscaling minimum temperature are presented in this section.
The results
Figure 8 shows the efficacy of the for downscaling
downscaling minimum
models temperature
in terms of the d. TheareEC-Earth
presented in this sect
raw
values showed aure 8 shows
d value thewhereas
of 0.5, efficacythe
of the downscaling
downscaling models
models in terms
showed of the d. The EC-Ea
improvement,
specially PowerTr values showed
increased the da values
d valueupof to
0.5, whereas
0.57. Figurethe downscaling
8 shows the NRMSE models showed impro
% values
of GCM and other specially PowerTr
downscaled outputincreased
comparedthetod values up to
the ERA-5 0.57. Figure
historical data.8The
shows the NRMSE %
NRMSE
of GCM
% values of the GCM wereand other downscaled
observed output
between 0 and 50%compared
for all thetothree
the ERA-5
GCMs.historical
However, data. The
% values of the GCM were observed between 0
the improvement in MRI-ESM2 using PowerTr and LS was up to 10% for NRMSE %. and 50% for all the three GCMs. H
the
The biases in improvement
the downscalinginGCMs MRI-ESM2 using PowerTr
are compared in Figureand8.LS was
The up to 10%
average for NRMSE
biases
in these models comparedThe biases
to theinERA-5
the downscaling
were in theGCMsrangeareof −compared
0.2 to 0.7%.in Figure 8. The averag
The results
showed that thein these models
PowerTr model compared
reduced the tobiases
the ERA-5
in thewere
threeinGCM
the range
by an of −0.2 toof0.7%. The
average
20–30%. Similarshowed
results that
werethe PowerTr
shown model
by the reduced whereas
LS method, the biasesthe in GammaQM
the three GCM andby an av
GenQM showed20−30%.very high Similar
errorsresults werecorrection.
in the bias shown byThe the SSLSofmethod,
the models whereas
were the
alsoGammaQ
compared to theGenQM
ERA-5 data.
showed Thevery
SS ofhigh
raw errors
GCM was 0.997,
in the biaswhich was improved
correction. The SS of up thetomodels w
0.999 by the PowerTr method
compared in most
to the ERA-5cases.
data.However,
The SS ofthe SSGCM
raw was noticed to decrease
was 0.997, which wasfor improve
GammaQM and 0.999 GenQM. by the PowerTr method in most cases. However, the SS was noticed to decr
GammaQM and GenQM.
Sustainability 2022, 14, x FOR PEER REVIEW 17 of 31
Sustainability 2022, 14, 6620 16 of 30

Figure 8. The statistical performance of the downscaled and raw GCMs for minimum temperature.
Figure 8. The statistical performance of the downscaled and raw GCMs for minimum tempera-
ture.
 Sustainability 2022, 14, x FOR PEER REVIEW
Sustainability 2022, 14, 6620 17 of 30

Figure 9 shows the Taylor diagram for EC-Earth minimum temperature. The
Figure 9 shows the Taylor diagram for EC-Earth minimum temperature. The cali-
tion and validation period show that the best model to downscale the EC-Earth
bration and validation period show that the best model to downscale the EC-Earth is the
PowerTr as it has reduced the RMSE and increased the correlation.
PowerTr as it has reduced the RMSE and increased the correlation.

Figure 9. Taylor Figure

diagram showsdiagram
9. Taylor bias correction methods’
shows bias performance
correction to downscaletoEC-Earth
methods’ performance downscale EC-Ea
minimum temperature during (a) calibration; (b) validation periods.
imum temperature during (a) calibration; (b) validation periods.

4.2. Calibration and Validation of Hydrological Model

4.2. Calibration and Validation of Hydrological Model
The integrated hydrological model was developed for each grid using the BIMERG
The integrated hydrological model was developed for each grid using the B
data. Model calibration and validation were performed using river flow data at Ratu
data. Model calibration and validation were performed using river flow data at Ra
Panjang from 2007 to 2017. Seven years’ data starting from 2007 was used for calibration,
jang from 2007 to 2017. Seven years’ data starting from 2007 was used for calib
whereas the remaining four years of river flow data were used for validation. The model
whereas the remaining four years of river flow data were used for validation. The
was developed using RF by tuning the parameters using repeated cross-validation for
was developed using RF by tuning the parameters using repeated cross-validat
random sampling. The ‘repeatedcv’ package in R was used to split the data into ten parts.
random sampling. The ‘repeatedcv’ package in R was used to split the data into te
Nine parts were used to train and the reaming part for validation in each of the ten iterations
Nine parts were used to train and the reaming part for validation in each of the te
used. The performance during each iteration was measured using evaluation metrics. The
tions used. The performance during each iteration was measured using evaluatio
average performance of ten folds with ten repetitions was calculated to summarize the
rics. The average performance of ten folds with ten repetitions was calculated to s
performance. The calibration and validation results are shown in Figure 10. The model
rize the performance.
showed good performance The calibration
for the validation and validation
period, giving the NSE, results
d, KGE,are shown
RMSE, andin Figure
Pbias of 0.96, 0.99, 0.92, 4.01, and −0.2, respectively. The NSE value of 0.96 was much better NSE, d
model showed good performance for the validation period, giving the
than the reported RMSE,
NSE and Pbias
values forof 0.96, 0.99,
similar other0.92, 4.01, and
models, such−0.2, respectively.
as SWAT and HSPF,The APEX,
NSE value of 0
and SAC-SMA [63]. Conventionally, the NSE value greater than 0.65 is considered goodsuch
much better than the reported NSE values for similar other models, for as SWA
model evaluating criteria [64]. The model showed satisfactory values for the coefficient ofthan 0.65
HSPF, APEX, and SAC-SMA [63]. Conventionally, the NSE value greater
determination (d) sidered
during good for model
calibration evaluating
(0.99) criteria [64].
and validation The(0.94).
period modelThe showed satisfactory
results also va
the coefficient of determination (d) during calibration (0.99)
showed very less bias in calibration and validation periods. Pbias showed overestimation and validation period
or underestimationThe ofresults also showed
the measured flow.very less
The bias in calibration
acceptable and validation
range of Pbias periods.
is less than 10%, Pbias s
whereas in this case, the Pbias was −0.2% for calibration and −7.2% for the validationrange of
overestimation or underestimation of the measured flow. The acceptable
period. The RMSE lessvalues
than 10%,
range whereas
from 4.01in this case,
to 5.64 forthe Pbias was
calibration and−0.2% for calibration
validation. The KGE and −7.2%
values were alsovalidation period. The
in the acceptable range RMSE values
of 0.92 range fromand
for calibration 4.01 to 5.64
0.86 for calibration and vali
for validation.
The KGE values were also in the acceptable range of 0.92 for calibration and 0.86 f
dation.
The model was evaluated using other statistical indices, such as MAR, d, md
RMSE, and Pbias. The results of these statistical indices are given in Table 4. The c
tion terms, such as d, md, and R2 were in the acceptable range, showing a good
mance in simulating the runoff during the calibration and validation period. Th
Sustainability 2022, 14, x FOR PEER REVIEW 19 of 31

Sustainability 2022, 14, 6620

terms showed very negligible values, indicating the model’s good performance in 18 of 30
simu-
lating observed flow.

Figure 10.
Figure 10. Observed and modelled
Observed and modelled river
river flow
flowduring
duringcalibration
calibrationand
andvalidation
validationperiod
periodofof2007–2017.
2007–
2017.
The model was evaluated using other statistical indices, such as MAR, d, md, KGE,
RMSE,A and
boxplot of The
Pbias. simulated
resultsand observed
of these flowindices
statistical for all the months
are given in during
Table 4.the
Theanalysis pe-
correlation
riod (2007−2017)
terms, was and
such as d, md, R2 were
plotted to analyze the seasonal
in the acceptable streamflow
range, showingvariations. Figure 11
a good performance
shows
in that thethe
simulating model simulates
runoff duringthe theseasonal variation
calibration well. The mean
and validation andThe
period. quantile
errorranges
terms
of the mean
showed verymonthly rainfall
negligible of each
values, monththe
indicating depicted
model’s a very
goodgood range of in
performance values. It can
simulating
be observed
observed in Figure 11 that the extreme values of the streamflow during January, May,
flow.
July, October, November, and December were also well simulated by the model. The sim-
Table Performances
ulated4. mean values of
of the model
each month during
weretheapproximately
calibration and validation periods.
equal to the observed flow. The
result indicates that the model can simulate the seasonal variation and2 the seasonal ex-
MAE RMSE NRMSE% Pbias NSE d md R KGE
tremes. Therefore, it can be used for climate change impact on river flow in the basin.
Caliberation 2.24 4.01 20.2 −0.2 0.96 0.99 0.91 0.97 0.92
Validation 3.8 5.64 50.2 −0.7 0.75 0.94 0.76 0.78 0.86

A boxplot of simulated and observed flow for all the months during the analysis
period (2007–2017) was plotted to analyze the seasonal streamflow variations. Figure 11
shows that the model simulates the seasonal variation well. The mean and quantile ranges
of the mean monthly rainfall of each month depicted a very good range of values. It can be
observed in Figure 11 that the extreme values of the streamflow during January, May, July,
October, November, and December were also well simulated by the model. The simulated
mean values of each month were approximately equal to the observed flow. The result
indicates that the model can simulate the seasonal variation and the seasonal extremes.
Therefore, it can be used for climate change impact on river flow in the basin.
Sustainability 2022,
Sustainability 2022, 14,
14, 6620
x FOR PEER REVIEW 20
19 of 31
of 30

Figure 11. Seasonal variations in mean monthly stream

stream flow.
flow.

4.3. Hydrological Changes under Future Scenarios

Table 4. Performances of the model during the calibration and validation periods.
4.3.1. Projected Rainfall Extremes
This sectionMAE RMSE
evaluated NRMSE%
five extreme Pbias
rainfall indicesNSE
defineddby WMO md for R2fourKGE
SSPs
Caliberation
and three GCMs.2.24 4.01 include
The indices 20.2
R95pTOT,−0.2R99pTOT,0.96Rx1day,
0.99 Rx5day,
0.91 0.97
and RI.0.92
The
Validation
historical 3.8 precipitation
and future 5.64 50.2
simulations of−0.7
the most0.75 0.94GCMs.
suitable 0.76 (As
0.78 0.86
discussed
in our previous paper [49], EC-Earth, EC-Earth-Veg, and MRI-ESM2 were used to assess
4.3. changes
the Hydrological Changes underextremes.
in precipitation Future Scenarios
The simulated extremes were analyzed for each
GCM individually
4.3.1. Projected to cover
Rainfall the maximum uncertainty range in the near (2020–2059) and
Extremes
far (2060–2099) future. The changes in these indices compared to the historical period are
This section evaluated five extreme rainfall indices defined by WMO for four SSPs
discussed in the following sections).
and three GCMs. The indices include R95pTOT, R99pTOT, Rx1day, Rx5day, and RI. The
historical
Total andabove
Rainfall future95th
precipitation
Percentilesimulations
(R95pTOT) of the most suitable GCMs. (As discussed
in our previous paper [49], EC-Earth, EC-Earth-Veg, and MRI-ESM2 were used to assess
Figure 12a shows the changes in R95TOT for different SSPs using three GCMs com-
the changes in precipitation extremes. The simulated extremes were analyzed for each
pared to the historical period. The results for EC-Earth showed the highest increase in
GCM individually to cover the maximum uncertainty range in the near (2020–2059) and
R95pTOT for SSP-119. The increase was 13 in the northern part of JRB. For SSP-370, a
far (2060–2099) future. The changes in these indices compared to the historical period are
moderate increase over the whole basin ranging from 2 to 10 mm was observed. However,
discussed in the following sections.)
SSP-585 projected a decrease in the northern part but an increase in total annual rainfall in
the southern region up to 10 mm.
Total Rainfall above 95th Percentile (R95pTOT)
Figure 12a shows the changes in R95TOT for different SSPs using three GCMs com-
pared to the historical period. The results for EC-Earth showed the highest increase in
R95pTOT for SSP-119. The increase was 13 in the northern part of JRB. For SSP-370, a
moderate increase over the whole basin ranging from 2 to 10 mm was observed. However,
SSP-585 projected a decrease in the northern part but an increase in total annual rainfall
in the southern region up to 10 mm.
The EC-Earth-Veg GCMs showed a slightly less increase than the EC-Earth for all the
scenarios. Under SSP-119, the R95pTOT value range from 5 mm in the northeastern region
of JRB to −2 mm in the southern region. Similarly, SSP-245 showed a small change of 1−2
mm over the entire basin, whereas SSP-370 showed an increase of 5 mm in the south and
a decrease of −6 mm in the northern parts. For the fossil fuel development scenarios (SSP-
 119 to 21 mm for SSP-585 for EC-Earth GCM. EC-Earth-Veg showed a similar pattern of
increase in R95pTOT for SSP-119 and SSP-585. A decrease in R95pTOT was observed for
SSP-199, whereas an increase by 5 and 21 mm was observed for SSP-245 and SSP-585,
respectively. MRI-ESM2 showed a reverse pattern for R95pTOT under SSP-119, whereas
Sustainability 2022, 14, 6620
a decrease in R95pTOT was projected for SSP-585. SSP-370 showed no changes in 20 of 30

R95pTOT in the far future.

Figure 12. Change Figure

in annual total rainfall
12. Change above
in annual 95 rainfall
total percentile (a) 95
above in percentile
the near future; (b)near
(a) in the far future.
future; (b) far future.

The EC-Earth-Veg
Total Rainfall above 99th GCMs showed a slightly less increase than the EC-Earth for all
Percentile (R99pTOT)
the scenarios. Under SSP-119, the R95pTOT value range from 5 mm in the northeastern
The percentage changes in R99pTOT are shown in Figure 13a. For SSP-119, the EC-
region of JRB to −2 mm in the southern region. Similarly, SSP-245 showed a small change
Earth showed an increase of 3 mm in the north and 1 mm in the southern part. EC-Earth-
of 1–2 mm over the entire basin, whereas SSP-370 showed an increase of 5 mm in the south
Veg showed a similar increase, whereas MRI-ESM2 showed a higher increase up to 9 mm
and a decrease of −6 mm in the northern parts. For the fossil fuel development scenarios
in the entire JRB(SSP-585),
for rainfall thewhich
R95pTOTexceeded
showedthe 99 percentiles
an increase of 8 mm ofindaily rain during
the southern the
region and a decrease
near future. For SSP-245, the EC-Earth showed
of −9 mm in the northern part. an increase of 3 mm. Similarly, EC-Earth-
Veg showed an increase of 3−5 mm,
MRI-ESM2 showed whereas the MRI-ESM2
a moderate increase inshowed an future
the near increase forofSSP-119,
5−7 whereas
mm in the basin.the percentage change in the R95pTOT showed a reduction for SSP-370 and SSP-585.
The EC-Earth Theand EC-Earth
rainfall Vegan
showed showed a similar
increase up to 5–9change
mm for SSP-370
in the like the previ-
sustainability scenario (SSP-119),
ous scenarios, but MRI-ESM2
whereas for theshowed
middle ofa slight increase
the road in the
scenario, the increase
R99pTOT (up to 1−2
in R95pTOT mm).
was minimal (2 mm).
Furthermore,showed
EC-Earth and EC-Earth-Veg the reduction of −change
a positive 2 to −13atmmthe in the R95pTOT
majority was points
of the grid noticed during the
regional
for SSP-585, whereas rivalry (SSP-370)
MRI-ESM2 showed and fossildecrease
a slight fuel development
of -1 mm(SSP-585)
in most of scenarios.
the grids
The
in JRB during 2020−2059. change in R95pTOT of the far future (2060–2099) compared to the base period
is shown in Figure 12b. The maps show a gradual increase
The changes in R99pTOT for the far future period are shown in Figure 13b. For SSP- in R95pTOT from 5 mm for
SSP-119 to 21 mm for SSP-585 for EC-Earth GCM. EC-Earth-Veg
119, the highest increase of 7 mm was observed for MRI-ESM2, whereas EC-Earth-Veg showed a similar pattern
of increase in R95pTOT for SSP-119 and SSP-585. A decrease in R95pTOT was observed
and EC-Earth showed an increase of 3−5 mm. An average increase of 4–5 mm was ob-
for SSP-199, whereas an increase by 5 and 21 mm was observed for SSP-245 and SSP-585,
served for all the GCMs for SSP-245 and SSP-375, except MRI-ESM2, which showed a neg-
respectively. MRI-ESM2 showed a reverse pattern for R95pTOT under SSP-119, whereas a
ative change of −1 to −2 mm at some grid points. Under fossil fuel development scenarios
decrease in R95pTOT was projected for SSP-585. SSP-370 showed no changes in R95pTOT
(SSP-585), the EC-Earth
in the farand EC-Earth-Veg showed an increase in R99pTOT up to 5 mm,
future.
whereas MRI-ESM2 showed a decrease by −1 to −2 mm.
Total Rainfall above 99th Percentile (R99pTOT)
The percentage changes in R99pTOT are shown in Figure 13a. For SSP-119, the EC-
Earth showed an increase of 3 mm in the north and 1 mm in the southern part. EC-Earth-Veg
showed a similar increase, whereas MRI-ESM2 showed a higher increase up to 9 mm in the
entire JRB for rainfall which exceeded the 99 percentiles of daily rain during the near future.
For SSP-245, the EC-Earth showed an increase of 3 mm. Similarly, EC-Earth-Veg showed an
increase of 3–5 mm, whereas the MRI-ESM2 showed an increase of 5–7 mm in the basin.
Sustainability 2022,
Sustainability 14,14,
2022, x FOR
6620 PEER REVIEW 2122 of 31
of 30

Figure
Figure13.
13.Change
Changein
inannual
annual total rainfallabove
total rainfall above99
99percentile
percentile(a)(a)
inin the
the near
near future;
future; (b)(b)
far far future.
future.

Thein
Changes EC-Earth
One Dayand EC-Earth
Max Veg(R×1day)
Rainfall showed a similar change for SSP-370 like the previous
scenarios, but MRI-ESM2 showed a slight increase in the R99pTOT (up to 1–2 mm). EC-
Figure 14a shows the changes in maximum one-day rainfall for 2020−2059. All GCMs
Earth and EC-Earth-Veg showed a positive change at the majority of the grid points for
showed an increase in R×1day at the majority of the grids for all the scenarios. For SSP-
SSP-585, whereas MRI-ESM2 showed a slight decrease of -1 mm in most of the grids in JRB
119, the 2020–2059.
during least change of 1–4 mm was observed for EC-Earth-Veg, whereas EC-Earth
showed Theanchanges
increase
inof up to 10for
R99pTOT mm theand
far MRI-ESM2
future periodbyare
19 shown
mm in intheFigure
near future.
13b. For SSP-
All GCMs projected a lower increase for SSP-245 than SSP-119. The
119, the highest increase of 7 mm was observed for MRI-ESM2, whereas EC-Earth-Veg maximum change
and
ofEC-Earth
10 mm followed
showed anbyincrease
an average increase
of 3–5 mm. Anofaverage
4−7 mm and 1−4
increase of mm waswas
4–5 mm observed for for
observed MRI-
ESM2,
all theEC-Earth, and EC-Earth-Veg,
GCMs for SSP-245 and SSP-375,respectively. For SSP-370,
except MRI-ESM2, the R×1day
which showed showed
a negative an in-
change
crease
of −1byto −4−7 mmatfor
2 mm EC-Earth,
some whereas
grid points. Undera slight
fossil decrease in the northern
fuel development part
scenarios by −2 mm
(SSP-585),
for
theEC-Earth-Veg
EC-Earth and and an overallshowed
EC-Earth-Veg increaseanofincrease
7−10 mm for MRI-ESM2.
in R99pTOT up to 5 For
mm,SSP-585,
whereasthe
MRI-ESM2 showed a decrease by − 1 to − 2 mm.
Rx1day showed an increase of 4−7 mm for EC-Earth. EC-Earth-Veg showed a slight de-
crease in the northern region, whereas an increase of 4−7 mm in the southern part. MRI-
Changes in One Day Max Rainfall (R × 1 day)
ESM2 showed a slight decrease at the majority of the grid points with a value ranging
Figure
from −2 to −514a
mm.shows the changes in maximum one-day rainfall for 2020–2059. All GCMs
showed an increase
The changes in the × 1future
in Rfar day atare
theshown
majority
in of the grids
Figure 14b. for
Forall the scenarios.
SSP-119, For
a decrease
Sustainability 2022, 14, x FOR PEER REVIEW 23 of in
31
SSP-119, the least change of 1–4 mm was observed for EC-Earth-Veg, whereas EC-Earth
Rx1day was observed for EC-Earth and EC-Earth-Veg by -3 to -9 mm. MRI-ESM2 showed
showed an increase of up to 10 mm and MRI-ESM2 by 19 mm in the near future.
an increase in the northern region up to 15 mm and an increase of 9 mm in the southern
part. The change in Rx1day was in the range of 9 to 21 mm for EC-Earth and EC-Earth-
Veg, whereas MRI-ESM2 showed an increase of 3 to 9 mm for SSP-245. SSP-370 showed a
further increase of Rx1day, EC-Earth (15–21mm), EC-Earth-Veg (15−27 mm), and MRI-
ESM2 (15−21 mm). Rx1day was projected to increase by 34 to 40 mm for the fossil fuel
development scenario. EC-Earth-Veg showed a similar increase of 21−34 mm, whereas
MRI-ESM2 decreased to −3 mm in the entire basin.

Figure
Figure14.
14.Change
Changein
inone-day
one-day max rainfall
rainfall (a)
(a)in
inthe
thenear
nearfuture;
future;(b)(b)
farfar future.
future.

Changes in 5-Day Max Rainfall (R×5day)

Figure 15a shows the changes in R×5day for 2020−2059. EC-Earth showed the greatest
changes of up to 44 mm for all SSPs, whereas EC-Earth-Veg showed a reduction. For SSP-
 Sustainability 2022, 14, 6620 22 of 30

All GCMs projected a lower increase for SSP-245 than SSP-119. The maximum change
of 10 mm followed by an average increase of 4–7 mm and 1–4 mm was observed for
MRI-ESM2, EC-Earth, and EC-Earth-Veg, respectively. For SSP-370, the R × 1 day showed
an increase by 4–7 mm for EC-Earth, whereas a slight decrease in the northern part by
−2 mm for EC-Earth-Veg and an overall increase of 7–10 mm for MRI-ESM2. For SSP-585,
the Rx1day showed an increase of 4–7 mm for EC-Earth. EC-Earth-Veg showed a slight
decrease in the northern region, whereas an increase of 4–7 mm in the southern part. MRI-
ESM2 showed a slight decrease at the majority of the grid points with a value ranging from
−2 to −5 mm.
The changes in the far future are shown in Figure 14b. For SSP-119, a decrease in
Rx1day was observed for EC-Earth and EC-Earth-Veg by -3 to -9 mm. MRI-ESM2 showed
an increase in the northern region up to 15 mm and an increase of 9 mm in the southern
part. The change in Rx1day was in the range of 9 to 21 mm for EC-Earth and EC-Earth-Veg,
whereas MRI-ESM2 showed an increase of 3 to 9 mm for SSP-245. SSP-370 showed a
further increase of Rx1day, EC-Earth (15–21mm), EC-Earth-Veg (15–27 mm), and MRI-
ESM2 (15–21 mm). Rx1day was projected to increase by 34 to 40 mm for the fossil fuel
development scenario. EC-Earth-Veg showed a similar increase of 21–34 mm, whereas
MRI-ESM2 decreased to −3 mm in the entire basin.

Changes in 5-Day Max Rainfall (R × 5 day)

Figure 15a shows the changes in R × 5 day for 2020–2059. EC-Earth showed the
greatest changes of up to 44 mm for all SSPs, whereas EC-Earth-Veg showed a reduction.
Sustainability 2022, 14, x FOR PEER REVIEW 24 of 31
For SSP-119, the change ranges from 12 to 25 mm for EC-Earth, −7 to 6 mm for EC-Earth-
Veg, and 0 to 6 mm for MRI-ESM2.

Figure
Figure15.
15.Change
Changein
infive-day
five-day max rainfall
rainfall (a)
(a)in
inthe
thenear
nearfuture;
future;(b)(b)
farfar future.
future.

ChangesEC-Earth showed
in Rainfall a −5 to
Intensity 12 mm change at different grids for SSP-245. EC-Earth
(RI)
showed a decrease by 0 to −7 mm. MRI-ESM2 also showed a decrease like EC-Earth-Veg.
The changes in RI under various climate change scenarios for the near and far future
For SSP-370, the southern region showed an increase of up to 44 mm for EC-Earth GCM,
are shownMRI-ESM2
whereas in Figure 16a,and respectively.
EC-Earth-VegThe Figure
showed an 16a shows
increase of the
2 todecrease
12 mm. in RI for SSP-
MRI-ESM2
119. EC-Earth projected a decrease by −1 to −3 mm, EC-Earth-Veg showed
showed an increase up to 38 mm in the southern part of JRB for SSP-585. In contrast, a small increase
ofEC-Earth-Veg
up to 1 mm,showedand MRI-ESM2 showed
a slight increase anR increase
in the × 5 day inofthe 6 to 8 mm.part
southern TheandRI afor SSP-245
decrease
showed
up to −no
3 mmmajor variations
in the northern for GCMs.
region. It showed
All grids underan increase
EC-Earth for SSP-370
showed in theinrange
an increase the R of
3–8 mm for EC-Earth, 3–10 mm for EC-Earth-Veg
× 5 day index ranging from 12–25 mm in the near future. and 6−8 mm for MRI-ESM2. The highest
increase
Theinchanges
rainfallin intensity
R × 5 day was
forfor
theSSP-585.
far futureIt are
wasshown
projected to increase
in Figure 15b. Theupplot
to 6−10
shows mm
anEC-Earth,
for overall increase
3–10 mm of up
for to a maximum of
EC-Earth-Veg, 108–10
and mm mm under forSSP-119
MRI-ESM2 for all GCMs.
over mostThere
parts of
was a further increase for EC-Earth up to 19 mm, whereas a decrease up to -19 mm in the
JRB.
northern
Change region
in RIforfor
EC-Earth-Veg
the far futureandisMRI-ESM2
shown in for SSP-245.
Figure 16b. AnTheoverall
resultsincrease
show aindecrease
the R
in RI for SSP-119 for all GCMs. However, the RI showed an increase all over JRB for SSP-
370 and 585.
Sustainability 2022, 14, 6620 23 of 30

Figure 15. Change

× 5 day in five-day
was observed max rainfall
for SSP-370. (a) in theshowed
MRI-ESM2 near future; (b) far future.
the maximum increase of 39 mm,
followed by EC-Earth (30 mm) and EC-Earth-Veg (19 mm). For SSP-585, EC-Earth R × 5
Changes in Rainfall
day showed Intensity
an increase of 40(RI)
mm in the southern region and 10–19 mm in the northern
region
The changes in RI under variousa slight
of JRB. EC-Earth-Veg showed decrease
climate the R × 5for
changeinscenarios daythe
in near
the northern
and far part
future
range of − 9 to −
are shown in Figure 16a, respectively. The Figure 16a shows the decrease observed
in the 19 mm. However, positive changes up to 10 mm were in RI for in
SSP-
the southern grids. MRI-ESM2 showed the highest increase of 48 mm in the JRB.
119. EC-Earth projected a decrease by −1 to −3 mm, EC-Earth-Veg showed a small increase
ofChanges
up to 1inmm, and Intensity
Rainfall MRI-ESM2 (RI) showed an increase of 6 to 8 mm. The RI for SSP-245
showed no major variations for GCMs. It showed an increase for SSP-370 in the range of
The changes in RI under various climate change scenarios for the near and far future
3–8
aremm
shownfor EC-Earth, 3–10
in Figure 16a, mm for EC-Earth-Veg
respectively. and
The Figure 16a 6−8 mm
shows for MRI-ESM2.
the decrease in RI forThe highest
SSP-119.
increase in rainfall intensity was for SSP-585. It was projected to increase
EC-Earth projected a decrease by −1 to −3 mm, EC-Earth-Veg showed a small increase of up to 6−10 mm
for
upEC-Earth,
to 1 mm, and3–10MRI-ESM2
mm for EC-Earth-Veg, and 8–10
showed an increase of 6 mm for MRI-ESM2
to 8 mm. over most
The RI for SSP-245 parts of
showed
JRB.
no major variations for GCMs. It showed an increase for SSP-370 in the range of 3–8 mm for
Change
EC-Earth, in RI
3–10 mm for
forthe far future isand
EC-Earth-Veg shown in Figure
6–8 mm 16b. TheThe
for MRI-ESM2. results show
highest a decrease
increase in
inrainfall
RI for intensity
SSP-119 for
wasallforGCMs. However,
SSP-585. the RI showed
It was projected an up
to increase increase
to 6–10all
mm over
forJRB for SSP-
EC-Earth,
3–10
370 andmm for EC-Earth-Veg, and 8–10 mm for MRI-ESM2 over most parts of JRB.
585.

Figure
Figure16.
16.Change
Changein
inrainfall
rainfall intensity (a) in
intensity (a) in the
thenear
nearfuture;
future;(b)
(b)farfarfuture.
future.

Change ininRIRiver
4.3.2. Changes for the far future is shown in Figure 16b. The results show a decrease in
Flow
RI for SSP-119 for all GCMs. However, the RI showed an increase all over JRB for SSP-370
The distributed hydrological model developed using RF was used to simulate the
and 585.
historical and future river flows using the downscaled data of CMIP6 GCMs. The histori-
cal andChanges
4.3.2. future inprecipitation
River Flow and temperature data of the most suitable GCMs (As
The distributed hydrological model developed using RF was used to simulate the
historical and future river flows using the downscaled data of CMIP6 GCMs. The historical
and future precipitation and temperature data of the most suitable GCMs (As discussed
in our previous paper [49], EC-Earth, EC-Earth-Veg, and MRI-ESM2 were used as input
for the model. The simulated flow for these projected data sets was analyzed for each
GCM individually to cover the maximum uncertainty range in the near (2020–2059) and
far (2060–2099) future. A comparison of quantiles for various SSPs for EC-Earth is shown
in Figure 17. Changes for low quantile flow were observed to decrease in the near future.
The maximum reduction (−14%) was for SSP-245, whereas a −8% reduction was for
0.1 quantiles compared to the historical flow period. For mid and higher quantiles, the
change in river flow was projected to increase up to 28%. The projected highest increase
Sustainability 2022, 14, 6620 24 of 30

Sustainability 2022, 14, x FOR PEER REVIEW 26 of 31

was in the near future period for the SSP-375. However, the higher quantiles, such as 0.90,
the minimum change were recorded for SSP-275 for the near future.

Figure 17. Changes in simulated flow for GCM EC-Earth.

Figure 17. Changes in simulated flow for GCM EC-Earth.
The changes in quantiles for EC-Earth show that the river flow reduces in all scenarios
5. Discussion
for lower quantiles, while the maximum increase of 32% was observed for higher quartiles
5.1. Reliability
for SSP-585, of the
followed byNewly
22% forDeveloped
SSP-119.Model
The model simulation using EC-Earth-Veg
Estimating river flow is an intricate process,showedespecially
a reductionin in river
data flowcatchments.
scares for lower A
quantiles in the near future, as shown in Figure 17. However, an increase in
huge set of data with a long temporal resolution is required for parameter estimation river flow was and
noted for higher quantiles,
optimization. The recent indicating
use of MLan in
increase in extreme
hydrological flows inisfuture
modelling gaining periods. The
more attention
maximum change ofcommunity.
in the scientific 25% was observed in higher
Integrated streamflow
hydrological modelquantiles in the with
development near the
future
help of
for SSP-245,
ML provedwhereas
to bethe lowestaschanges
efficient compared in the higher
to the extremes were
conventional for SSP-585.
modelling In the
methods. far
However,
future period,
there a similar
remains a gapflowforpattern for lowerbyquantiles
improvement optimizingand its
an internal
increase parameters.
in higher quantiles
Therefore,
werethis
recorded with a maximum increase of 32% for SSP-119 and 27%
study developed a distributed hydrological model using RF for parameter for SSP-245. estimation.
The calibration and validation results are provided in Section 4.2. The statistical indices
used to show the efficiency of the model output shows the model’s good capability to
simulate the river flow in JRB. The model showed good performance for the calibration
Sustainability 2022, 14, 6620 25 of 30

Figure 17 shows the changes in river flow for MRI-ESM2. The figure shows the greatest
change in the higher quantiles for SSP-119 (80%) and the lowest change in lower quantiles
(<0.3) for SSP-585. The percentage reduction for SSP-585 was −23% in the near future.
The far future showed an increase in river flow extremes, whereas lower flows showed a
reduction. The maximum change of 68% was observed for SSP-119 for higher quantiles in
the far future. The lowest reduction in the flow was −24% for SSP-585 in the near future.

5. Discussion
5.1. Reliability of the Newly Developed Model
Estimating river flow is an intricate process, especially in data scares catchments. A
huge set of data with a long temporal resolution is required for parameter estimation and
optimization. The recent use of ML in hydrological modelling is gaining more attention
in the scientific community. Integrated hydrological model development with the help of
ML proved to be efficient as compared to the conventional modelling methods. However,
there remains a gap for improvement by optimizing its internal parameters. Therefore, this
study developed a distributed hydrological model using RF for parameter estimation. The
calibration and validation results are provided in Section 4.2. The statistical indices used to
show the efficiency of the model output shows the model’s good capability to simulate the
river flow in JRB. The model showed good performance for the calibration period, giving
the NSE, d, KGE, RMSE, and Pbias of 0.96,0.99,0.92 4.01, and −0.2, respectively. The NSE
value of 0.96 is much better than the reported NSE values for the calibration period for
similar other models, such as SWAT and HSPF, APEX and SAC-SMA [63]. Conventionally,
the NSE value greater than 0.65 is considered good for model evaluating criteria [64].
The model showed satisfactory values for d during the calibration (0.99) and validation
period (0.94). The simulation of the model train with machine learning algorithms showed
very less bias in calibration and validation periods. Pbias revealed the overestimation or
underestimation of the simulated flow compared to the measured flow. From the literature,
the acceptable range of Pbias for model simulation is less than 10%, whereas in this case,
the Pbias is −0.2% for calibration and −7.2% for the validation period. The error terms
such as RMSE values range from 4.01 to 5.64 for calibration and validation. The KGE values
are also in the acceptable range of 0.92 for calibration and 0.86 for validation. Tan et al. [45]
validated the re-known SWAT model in JRB. The statistical results of the SWAT model for
calibration and validation are NSE 0.66 and 0.62, respectively, whereas the model developed
in this study showed NSE values of 0.96–0.75. The model calibration and validation result
shows that this model can be used to simulate river flows in JRB for other datasets.

5.2. Changes in Precipitation Flood Frequency under Future Scenario

Variation in intensity and frequency of various climate and weather extremes has been
found in the literature. The climate extremes are mostly found increasing in many parts
of the world [65,66]. Over MSEA, substantial changes in daily rainfall intensity and the
number of heavy rainfall days (R20 mm) have been observed for various future scenarios
of CIMP6 [67]. Therefore, in this study, CMIP6 model outputs of three selected models
were used for the historic forcing and future scenarios, such as SSP-119, SSP-245, SSP-370
and SSP-585, to study the changes in precipitation extremes in JRB. These data of CIMP6
GCMs were also used in the hydrological model to determine the change in flow extremes
in JRB under various SSP scenarios. The results are shown in Section 4.3.2. Under different
SSP scenarios, the flow at Ratu Panjang station was observed to be increasing for higher
quantiles. However, the changes in lower quantiles compared to the historical flow were
decreasing. Similarly, the precipitation extremes such as five-day max rainfall and rainfall
intensity were observed to be increasing in the latter part of the century. The changes in
these rainfall extremes and the substantial increase in some other extremes such as R × 1
day, R95pTOT, and R99pTOT justify the increasing river flow for higher quantiles than the
historical flow at the river station.
Sustainability 2022, 14, 6620 26 of 30

Similar changes in climate extremes have also been reported by Kharin et al. [68].
They used the transient non-stationary GEV to study the global scale frequency change in
climate extremes and risk ratio. The risk ratio determined in the study showed an increase
from 0.65 to 1.22, while the global temperature increased from the preindustrial level under
scenarios of CMIP5. Our study also strengthens the hypothesis that “The contrast in relative
frequency changes between more extreme and weaker events is projected to become larger as climate
warms”. Li et al. [69] analyzed 20 GCM from CMIP6 to study the change in temperature and
precipitation extremes over the globe. The study found that most of the model increases
the intensity and frequency of precipitation extremes, especially over tropical regions. The
maximum one- and five-day rainfall events R × 1 day and R × 5 day increased up to 7.2%
compared to the historical extremes. In the majority region of the world, the temperature
and precipitation extremes were following the “intense gets intenser” tendency. However,
comparable results of flow quantiles were observed in our study. The lower flow quantiles
were found to decrease in most scenarios, whereas the higher quantiles were increasing for
all models and scenarios. Therefore, it can be remarked that, in JRB, the precipitation and
river flow extremes at Ratu Panjang gauge station will be increasing in the future.

5.3. Significance of the Study

Hydrological disasters like floods and water stress have become a common phe-
nomenon in many countries globally. Consequently, a moderate dry spell often forces
water rationing and moderate or extreme rainfall causing floods, especially in rapidly
developing urban catchments [70]. The changing pattern of hydrological disasters due to
environmental changes is a major concern for scientists and policymakers all over the globe.
Numerous hydrological models have been developed to estimate runoff from rainfall to
predict hydrological disasters [19]. Distributed hydrological models have been found to be
most reliable for runoff prediction. However, they need extensive data and parameters in
space and time for reliable runoff estimation. The outputs of such models are also prone
to uncertainties due to the simple approximation of many hydrological processes. This
study attempt to introduce a machine learning algorithm to improve the performance of
the distributed hydrological model. The results showed that the hybridization of ML and
conventional physical or conceptual model improved the capability to model complex
interactions and runoff prediction. The model can provide a more accurate estimation of
streamflow extremes. Therefore, the model can be used for reliable simulation of hydro-
logical disasters, particularly water scarcity and floods, due to the changes in land use
driven by physical and socio-economic factors and climate. This is particularly important
for developing countries where rapid landuse changes have significantly affected local
hydrology. This study also showed the suitability of the model in reliable projections of
hydrological changes due to climate change. Therefore, the model’s output can be used for
climate change adaptation and mitigation planning.

6. Conclusions
The distributed hydrological model for JRB was developed by using ML algorithms.
RF was used to estimate the parameters to calculate the simulated flows. The model was
developed using the bias-corrected IMERG data with an approximate resolution of 10 km.
The soil properties and the topographical characteristics were included in calculating the
model output. The model showed a varying flow simulation at Ratu Panjang compared to
the observed flow. The efficiency of the model was assessed by calculating statistical indices.
These indices values, such as RMSE, NSE, and R2 , proved that the distributed hydrological
model can simulate the flow of any catchment. The calibration and validation results
and the processing time prove that the ML-based models are good in flood simulation in
any catchment with insufficient historical data. The model developed in this study can
efficiently simulate the hydrological behaviour like the physical models, and also it can
be applied to generate the long-term simulation. The model provided a near real-time
Sustainability 2022, 14, 6620 27 of 30

flood simulation using the bias-corrected IMERG data and used it to indicate the flood
susceptibly of any region.
The study found that the river flow under the change climate scenarios increases
with the higher carbon concentration pathways. The results also revealed that the rainfall
extremes are also getting worse in intensity and frequency. The reduction of flow up
to −14% at lower quantiles and an increase of 28% at mid and higher quantiles were
recorded in this analysis. Similarly, the sustainability pathway (SSP1) showed a reduction
in projected river flow extremes, whereas the middle of the road (SSP2) showed a balance
increase in the higher flow quantiles. Contrary to these, the regional rivalry (SSP3) and
fossil fuel development (SSP5) showed a higher increase in streamflow extremes by up
to 68% at the end of the century. The framework developed in this study can simulate
the historical and future surface runoff very effectively with very few parameters. The
model’s efficacy is improved due to the use of RF in parameters estimation and GCM data,
enabling it to simulate the effect of climate change on the river discharge in the region. The
simulation takes less time, showing that the model can also be considered for NRT flood
simulation in any region.

Supplementary Materials: The following supporting information can be downloaded at: https:
//www.mdpi.com/article/10.3390/su14116620/s1, Supplemetry materials are available with this
paper as appendices.
Author Contributions: Conceptualization, Z.I. and S.S.; Methodology, Z.I.; Software, Z.I. and S.S.;
Validation, S.S.; Formal analysis, Z.I. and S.S.; Resources, T.I.; data curation, Z.S.; writing—original
draft preparation, Z.I., Z.M.Y. and S.S.; writing—review and editing, A.F., T.I. and Z.M.Y.; funding
acquisition, T.I. and A.F. All authors have read and agreed to the published version of the manuscript.
Funding: Ministry of Higher Education Malaysia (FRGS) (No. R.J130000.78515F092).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The observed streamflow data is not available to be shared with a third
party as per instruction from the Department of Irrigation and Drainage Malaysia. However, the
GCM and Satellite data sets are freely available on the website/references given in the article.
Acknowledgments: The authors would like to acknowledge Higher Education Commission Pakistan
(HEC) and Ministry of Higher Education Malaysia (FRGS) (No. R.J130000.78515F092) for providing
financial support to conduct this research. We also acknowledge the Department of Irrigation and
Drainage Malaysia for providing the rainfall data of entire Peninsular Malaysia.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Yaseen, Z.M.; Sulaiman, S.O.; Deo, R.C.; Chau, K.-W. An enhanced extreme learning machine model for river flow forecasting:
State-of-the-art, practical applications in water resource engineering area and future research direction. J. Hydrol. 2019, 569,
387–408. [CrossRef]
2. Vogel, R.M.; Lall, U.; Cai, X.; Rajagopalan, B.; Weiskel, P.K.; Hooper, R.P.; Matalas, N.C. Hydrology: The interdisciplinary science
of water. Water Resour. Res. 2015, 51, 4409–4430. [CrossRef]
3. Halder, B.; Haghbin, M.; Farooque, A.A. An Assessment of Urban Expansion Impacts on Land Transformation of Rajpur-Sonarpur
Municipality. Knowl.-Based Eng. Sci. 2021, 2, 34–53. [CrossRef]
4. Ahmad, S.; Simonovic, S.P. Spatial System Dynamics: New Approach for Simulation of Water Resources Systems. J. Comput. Civ.
Eng. 2004, 18, 331–340. [CrossRef]
5. Sa’adi, Z.; Shiru, M.S.; Shahid, S.; Ismail, T. Selection of general circulation models for the projections of spatio-temporal changes
in temperature of Borneo Island based on CMIP5. Theor. Appl. Climatol. 2020, 139, 351–371. [CrossRef]
6. Ehteram, M.; Othman, F.B.; Yaseen, Z.M.; Afan, H.A.; Allawi, M.F.; Malek, M.B.A.; Ahmed, A.N.; Shahid, S.; Singh, V.P.;
El-Shafie, A. Improving the Muskingum flood routing method using a hybrid of particle swarm optimization and bat algorithm.
Water 2018, 10, 807. [CrossRef]
7. Sharafati, A.; Khazaei, M.R.; Nashwan, M.S.; Al-Ansari, N.; Yaseen, Z.M.; Shahid, S. Assessing the uncertainty associated with
flood features due to variability of rainfall and hydrological parameters. Adv. Civ. Eng. 2020, 2020, 7948902. [CrossRef]
Sustainability 2022, 14, 6620 28 of 30

8. Mango, L.M.; Melesse, A.M.; McClain, M.E.; Gann, D.; Setegn, S.G. Land use and climate change impacts on the hydrology of the
upper Mara River Basin, Kenya: Results of a modeling study to support better resource management. Hydrol. Earth Syst. Sci.
2011, 15, 2245–2258. [CrossRef]
9. Halder, B.; Ameen, A.M.S.; Bandyopadhyay, J.; Khedher, K.M.; Yaseen, Z.M. The impact of climate change on land degradation
along with shoreline migration in Ghoramara Island, India. Phys. Chem. Earth Parts A/B/C 2022, 103135. [CrossRef]
10. Saudi, A.S.M.; Juahir, H.; Azid, A.; Azaman, F. Flood risk index assessment in Johor River Basin. Malays. J. Anal. Sci. 2015, 19,
991–1000.
11. Muzamil, S.A.H.B.S.; Zainun, N.Y.; Ajman, N.N.; Sulaiman, N.; Khahro, S.H.; Rohani, M.M.; Mohd, S.M.B.; Ahmad, H. Proposed
Framework for the Flood Disaster Management Cycle in Malaysia. Sustainability 2022, 14, 4088. [CrossRef]
12. Shahid, S.; Alamgir, M.; Wang, X.; Eslamian, S. Climate Change Impacts on and Adaptation to Groundwater. Handb. Drought
Water Scarcity Environ. Impacts Anal. Drought Water Scarcity 2017, 2, 107–124.
13. Ziarh, G.F.; Asaduzzaman, M.; Dewan, A.; Nashwan, M.S.; Shahid, S. Integration of catastrophe and entropy theories for flood
risk mapping in peninsular Malaysia. J. Flood Risk Manag. 2021, 14, e12686. [CrossRef]
14. Connor, R. The United Nations World Water Development Report 2015: Water for a Sustainable World; UNESCO Publishing: Bonn,
Germany, 2015; Volume 1, ISBN 9231000713.
15. Chemicals, U. Standardized Toolkit for Identification and Quantification of Dioxin and Furan Releases; United Nations Environment
Programme: Geneva, Switzerland, 2003; Volume 194.
16. Iqbal, Z.; Shahid, S.; Ahmed, K.; Ismail, T.; Nawaz, N. Spatial distribution of the trends in precipitation and precipitation extremes
in the sub-Himalayan region of Pakistan. Theor. Appl. Climatol. 2019, 137, 2755–2769. [CrossRef]
17. Ahmad, S.; Simonovic, S.P. System Dynamics Modeling of Reservoir Operations for Flood Management. J. Comput. Civ. Eng.
2000, 14, 190–198. [CrossRef]
18. Sitterson, J.; Sinnathamby, S.; Parmar, R.; Koblich, J.; Wolfe, K.; Knightes, C.D. Demonstration of an online web services tool
incorporating automatic retrieval and comparison of precipitation data. Environ. Model. Softw. 2020, 123, 104570. [CrossRef]
19. Young, P.C. Advances in real–time flood forecasting. Philos. Trans. R. Soc. London. Ser. A Math. Phys. Eng. Sci. 2002, 360, 1433–1450.
[CrossRef]
20. Fahimi, F.; Yaseen, Z.M.; El-shafie, A. Application of soft computing based hybrid models in hydrological variables modeling:
A comprehensive review. Theor. Appl. Climatol. 2017, 128, 875–903. [CrossRef]
21. Naganna, S.R.; Beyaztas, B.H.; Bokde, N.; Armanuos, A.M. On the evaluation of the gradient tree boosting model for groundwater
level forecasting. Knowl.-Based Eng. Sci. 2020, 1, 48–57. [CrossRef]
22. Devia, G.K.; Ganasri, B.P.; Dwarakish, G.S. A Review on Hydrological Models. Aquat. Procedia 2015, 4, 1001–1007. [CrossRef]
23. Perrin, C.; Michel, C.; Andréassian, V. Does a large number of parameters enhance model performance? Comparative assessment
of common catchment model structures on 429 catchments. J. Hydrol. 2001, 242, 275–301. [CrossRef]
24. Agrawal, N.; Desmukh, T.S. Rainfall Runoff Modeling using MIKE 11 Nam—A Review. Int. J. Innov. Sci. Eng. Technol. 2016, 3,
659–667.
25. Yaseen, Z.M.; Ebtehaj, I.; Kim, S.; Sanikhani, H.; Asadi, H.; Ghareb, M.I.; Bonakdari, H.; Wan Mohtar, W.H.M.; Al-Ansari, N.;
Shahid, S. Novel hybrid data-intelligence model for forecasting monthly rainfall with uncertainty analysis. Water 2019, 11, 502.
[CrossRef]
26. Khosravi, K.; Golkarian, A.; Booij, M.J.; Barzegar, R.; Sun, W.; Yaseen, Z.M.; Mosavi, A. Improving daily stochastic streamflow
prediction: Comparison of novel hybrid data-mining algorithms. Hydrol. Sci. J. 2021, 66, 1457–1474. [CrossRef]
27. Johari, A.; Javadi, A.A.; Habibagahi, G. Modelling the mechanical behaviour of unsaturated soils using a genetic algorithm-based
neural network. Comput. Geotech. 2011, 38, 2–13. [CrossRef]
28. Omeje, O.E.; Maccido, H.S.; Badamasi, Y.A.; Abba, S.I. Performance of Hybrid Neuro-Fuzzy Model for Solar Radiation Simulation
at Abuja, Nigeria: A Correlation Based Input Selection Technique. Knowl.-Based Eng. Sci. 2021, 2, 54–66.
29. Khan, N.; Shahid, S.; Juneng, L.; Ahmed, K.; Ismail, T.; Nawaz, N. Prediction of heat waves in Pakistan using quantile regression
forests. Atmos. Res. 2019, 221, 1–11. [CrossRef]
30. Wang, X.; Zhang, J.; He, R.; Amgad, E.; Sondoss, E.; Shang, M. A strategy to deal with water crisis under climate change for
mainstream in the middle reaches of Yellow River. Mitig. Adapt. Strateg. Glob. Chang. 2010, 16, 555–566. [CrossRef]
31. Qin, H.-P.; Su, Q.; Khu, S.-T. An integrated model for water management in a rapidly urbanizing catchment. Environ. Model.
Softw. 2011, 26, 1502–1514. [CrossRef]
32. Tidwell, V.C.; Passell, H.D.; Conrad, S.H.; Thomas, R.P. System dynamics modeling for community-based water planning:
Application to the Middle Rio Grande. Aquat. Sci. 2004, 66, 357–372. [CrossRef]
33. Ropero, R.F.; Rumí, R.; Aguilera, P.A. Modelling uncertainty in social–natural interactions. Environ. Model. Softw. 2016, 75,
362–372. [CrossRef]
34. Yaseen, Z.M.; Shahid, S. Drought Index Prediction Using Data Intelligent Analytic Models: A Review. In Intelligent Data Analytics
for Decision-Support Systems in Hazard Mitigation; Springer: Cham, Switzerland, 2020; pp. 1–27.
35. Chandwani, V.; Vyas, S.K.; Agrawal, V.; Sharma, G. Soft computing approach for rainfall-runoff modelling: A review. Aquat.
Procedia 2015, 4, 1054–1061. [CrossRef]
36. Koch, J.; Demirel, M.C.; Stisen, S. The SPAtial EFficiency metric (SPAEF): Multiple-component evaluation of spatial patterns for
optimization of hydrological models. Geosci. Model Dev. 2018, 11, 1873–1886. [CrossRef]
Sustainability 2022, 14, 6620 29 of 30

37. Guo, H.C.; Liu, L.; Huang, G.H.; Fuller, G.A.; Zou, R.; Yin, Y.Y. A system dynamics approach for regional environmental planning
and management: A study for the Lake Erhai Basin. J. Environ. Manag. 2001, 61, 93–111. [CrossRef]
38. Sood, A. Integrated Watershed Management as an Effective Tool for Sustainable Development: Using Distributed Hydrological
Models in Policy Making. Ph.D. Thesis, University of Delaware, Newark, NJ, USA, 2009.
39. Minville, M.; Cartier, D.; Guay, C.; Leclaire, L.-A.; Audet, C.; Le Digabel, S.; Merleau, J. Improving process representation in
conceptual hydrological model calibration using climate simulations. Water Resour. Res. 2014, 50, 5044–5073. [CrossRef]
40. Bárdossy, A.; Singh, S.K. Robust estimation of hydrological model parameters. Hydrol. Earth Syst. Sci. 2008, 12, 1273–1283.
[CrossRef]
41. Zhang, L.; Nan, Z.; Xu, Y.; Li, S. Hydrological Impacts of Land Use Change and Climate Variability in the Headwater Region of
the Heihe River Basin, Northwest China. PLoS ONE 2016, 11, e0158394. [CrossRef]
42. Zomorodian, M.; Lai, S.H.; Homayounfar, M.; Ibrahim, S.; Fatemi, S.E.; El-Shafie, A. The state-of-the-art system dynamics
application in integrated water resources modeling. J. Environ. Manag. 2018, 227, 294–304. [CrossRef]
43. Ratnayeke, S.; van Manen, F.T.; Clements, G.R.; Kulaimi, N.A.M.; Sharp, S.P. Carnivore hotspots in Peninsular Malaysia and their
landscape attributes. PLoS ONE 2018, 13, e0194217. [CrossRef]
44. Kia, M.B.; Pirasteh, S.; Pradhan, B.; Mahmud, A.R.; Sulaiman, W.N.A.; Moradi, A. An artificial neural network model for flood
simulation using GIS: Johor River Basin, Malaysia. Environ. Earth Sci. 2012, 67, 251–264. [CrossRef]
45. Tan, M.L.; Ficklin, D.L.; Ibrahim, A.L.; Yusop, Z. Impacts and uncertainties of climate change on streamflow of the Johor River
Basin, Malaysia using a CMIP5 General Circulation Model ensemble. J. Water Clim. Chang. 2014, 5, 676–695. [CrossRef]
46. Webster, P.J.; Magaña, V.O.; Palmer, T.N.; Shukla, J.; Tomas, R.A.; Yanai, M.; Yasunari, T. Monsoons: Processes, predictability, and
the prospects for prediction. J. Geophys. Res. Ocean. 1998, 103, 14451–14510. [CrossRef]
47. Noor, M.; Ismail, T.B.; Shahid, S.; Ahmed, K.; Chung, E.-S.; Nawaz, N. Selection of CMIP5 multi-model ensemble for the projection
of spatial and temporal variability of rainfall in peninsular Malaysia. Theor. Appl. Climatol. 2019, 138, 999–1012. [CrossRef]
48. Zhang, W.; Villarini, G.; Scoccimarro, E.; Napolitano, F. Examining the precipitation associated with medicanes in the high-
resolution ERA-5 reanalysis data. Int. J. Climatol. 2020, 41, E126–E132. [CrossRef]
49. Iqbal, Z.; Shahid, S.; Ahmed, K.; Ismail, T.; Ziarh, G.F.; Chung, E.-S.; Wang, X. Evaluation of CMIP6 GCM rainfall in mainland
Southeast Asia. Atmos. Res. 2021, 254, 105525. [CrossRef]
50. Iqbal, Z.; Shahid, S.; Ahmed, K.; Wang, X.; Ismail, T.; Gabriel, H.F. Bias correction method of high-resolution satellite-based
precipitation product for Peninsular Malaysia. Theor. Appl. Climatol. 2022, 148, 1429–1446. [CrossRef]
51. Huang, M.; Lin, R.; Huang, S.; Xing, T. A novel approach for precipitation forecast via improved K-nearest neighbor algorithm.
Adv. Eng. Inform. 2017, 33, 89–95. [CrossRef]
52. Maraun, D.; Wetterhall, F.; Ireson, A.M.; Chandler, R.E.; Kendon, E.J.; Widmann, M.; Brienen, S.; Rust, H.W.; Sauter, T.; Themel,
M.; et al. Precipitation downscaling under climate change: Recent developments to bridge the gap between dynamical models
and the end user. Rev. Geophys. 2010, 48, 2009RG000314. [CrossRef]
53. Eden, J.M.; Widmann, M. Downscaling of GCM-Simulated Precipitation Using Model Output Statistics. J. Clim. 2014, 27, 312–324.
[CrossRef]
54. Piani, C.; Haerter, J.O.; Coppola, E. Statistical bias correction for daily precipitation in regional climate models over Europe. Theor.
Appl. Climatol. 2010, 99, 187–192. [CrossRef]
55. Wilcke, R.A.I.; Mendlik, T.; Gobiet, A. Multi-variable error correction of regional climate models. Clim. Change 2013, 120, 871–887.
[CrossRef]
56. Amengual, A.; Homar, V.; Romero, R.; Alonso, S.; Ramis, C. A statistical adjustment of regional climate model outputs to local
scales: Application to Platja de Palma, Spain. J. Clim. 2012, 25, 939–957. [CrossRef]
57. Leander, R.; Buishand, T.A. Resampling of regional climate model output for the simulation of extreme river flows. J. Hydrol.
2007, 332, 487–496. [CrossRef]
58. Terink, W.; Hurkmans, R.; Torfs, P.; Uijlenhoet, R. Bias correction of temperature and precipitation data for regional climate model
application to the Rhine basin. Hydrol. Earth Syst. Sci. Discuss. 2009, 6, 5377–5413.
59. Coles, S.; Bawa, J.; Trenner, L.; Dorazio, P. An Introduction to Statistical Modeling of Extreme Values; Springer: Cham, Switzerland,
2001; Volume 208.
60. Lenderink, G.; Buishand, A.; Van Deursen, W. Estimates of future discharges of the river Rhine using two scenario methodologies:
Direct versus delta approach. Hydrol. Earth Syst. Sci. 2007, 11, 1145–1159. [CrossRef]
61. Lafon, T.; Dadson, S.; Buys, G.; Prudhomme, C. Bias correction of daily precipitation simulated by a regional climate model:
A comparison of methods. Int. J. Climatol. 2013, 33, 1367–1381. [CrossRef]
62. Allen, R.G.; Pruitt, W.O. FAO-24 Reference Evapotranspiration Factors. J. Irrig. Drain. Eng. 1991, 117, 758–773. [CrossRef]
63. Moriasi, D.N.; Arnold, J.G.; Liew, M.W.V.a.n.; Bingner, R.L.; Harmel, R.D.; Veith, T.L.; Van Liew, M.; Bingner, R.L.; Harmel, R.D.;
Veith, T.L. Model Evaluation Guidelines for Systematic Quantification of Accuracy in Watershed Simulations. Trans. ASABE 2007,
50, 885–900. [CrossRef]
64. Jeong, J.; Kannan, N.; Arnold, J.; Glick, R.; Gosselink, L.; Srinivasan, R. Development and Integration of Sub-hourly Rainfall–
Runoff Modeling Capability Within a Watershed Model. Water Resour. Manag. 2010, 24, 4505–4527. [CrossRef]
65. Ahmed, K.; Shahid, S.; Chung, E.S.; Ismail, T.; Wang, X.J. Spatial distribution of secular trends in annual and seasonal precipitation
over Pakistan. Clim. Res. 2017, 74, 95–107. [CrossRef]
Sustainability 2022, 14, 6620 30 of 30

66. Nashwan, M.S.; Shahid, S. Spatial distribution of unidirectional trends in climate and weather extremes in Nile river basin. Theor.
Appl. Climatol. 2019, 137, 1181–1199. [CrossRef]
67. Ge, F.; Zhu, S.; Luo, H.; Zhi, X.; Wang, H. Future changes in precipitation extremes over Southeast Asia: Insights from CMIP6
multi-model ensemble. Environ. Res. Lett. 2021, 16, 24013. [CrossRef]
68. Kharin, V.V.; Flato, G.M.; Zhang, X.; Gillett, N.P.; Zwiers, F.; Anderson, K.J. Risks from Climate Extremes Change Differently from
1.5 ◦ C to 2.0 ◦ C Depending on Rarity. Earth’s Future 2018, 6, 704–715. [CrossRef]
69. Li, C.; Zwiers, F.; Zhang, X.; Li, G.; Sun, Y.; Wehner, M. Changes in Annual Extremes of Daily Temperature and Precipitation in
CMIP6 Models. J. Clim. 2021, 34, 3441–3460. [CrossRef]
70. Pereira, L.S.; Cordery, I.; Iacovides, I. Coping with Water Scarcity: Addressing the Challenges; Springer Science & Business Media:
New York, NY, USA, 2009; ISBN 978-1-4020-9578-8.