Environment, Development and Sustainability (2021) 23:18252–18277

https://doi.org/10.1007/s10668-021-01437-6

Remote sensing‑based water quality index estimation

using data‑driven approaches: a case study of the Kali River
in Uttar Pradesh, India

Saif Said1 · Shadab Ali Khan1

Received: 4 April 2020 / Accepted: 13 April 2021 / Published online: 21 April 2021
© The Author(s), under exclusive licence to Springer Nature B.V. 2021

Abstract
The present study evaluates the water quality status of 6-km-long Kali River stretch that
passes through the Aligarh district in Uttar Pradesh, India, by utilizing high-resolution IRS
P6 LISS IV imagery. In situ river water samples collected at 40 random locations were ana-
lyzed for seven physicochemical and four heavy metal concentrations, and the water qual-
ity index (WQI) was computed for each sampling location. A set of 11 spectral reflectance
band combinations were formulated to identify the most significant band combination that
is related to the observed WQI at each sampling location. Three approaches, namely multi-
ple linear regression (MLR), backpropagation neural network (BPNN) and gene expression
programming (GEP), were employed to relate WQI as a function of most significant band
combination. Comparative assessment among the three utilized approaches was performed
via quantitative indicators such as R2, RMSE and MAE. Results revealed that WQI esti-
mates ranged between 203.7 and 262.33 and rated as “very poor” status. Results further
indicated that GEP performed better than BPNN and MLR approaches and predicted WQI
estimates with high R2 values (i.e., 0.94 for calibration and 0.91 for validation data), low
RMSE and MAE values (i.e., 2.49 and 2.16 for calibration and 4.45 and 3.53 for valida-
tion data). Moreover, both GEP and BPNN depicted superiority over MLR approach that
yielded WQI with R2 ~ 0.81 and 0.67 for calibration and validation data, respectively. WQI
maps generated from the three approaches corroborate the existing pollution levels along
the river stretch. In order to examine the significant differences among WQI estimates from
the three approaches, one-way ANOVA test was performed, and the results in terms of
F-statistic (F = 0.01) and p-value (p = 0.994 > 0.05) revealed WQI estimates as “not sig-
nificant,” reasoned to the small water sample size (i.e., N = 40). The study therefore recom-
mends GEP as more rational and a better alternative for precise water quality monitoring of
surface water bodies by producing simplified mathematical expressions.

Keywords Kali River · WQI · Spectral reflectance · MLR · ANN · GEP

* Saif Said
[email protected]
Shadab Ali Khan
[email protected]
1
Civil Engineering Department, Aligarh Muslim University, (AMU), Aligarh, India

1 Vol:.(1234567890)
3
Remote sensing‑based water quality index estimation using… 18253

1 Introduction

İn recent years, water quality of major rivers, lakes and ponds in India has alarmingly dete-
riorated due to significant population increase leading to rapid urban development and
industrialization. Increased anthropogenic activities including direct discharge of untreated
industrial effluents, domestic sewage and agricultural waste have severely degraded the
quality of surface water bodies. In India, the management strategies for cleaning up of riv-
ers are often not optimally prioritized and therefore, spatiotemporal monitoring of pollu-
tion levels becomes essential to devise effective measures for reclamation of the degraded
urban water bodies (Farhad et al., 2013; Abba et al., 2015). In situ measurement and moni-
toring of water quality at point locations is exhaustive and time taking (Song et al., 2012).
Mathematical models integrated with geospatial techniques form a reliable time-saving
solution towards controlling and sustainably managing the surface water resources (Mon-
dal and Satpaty, 2020). Geospatial techniques offer uninterrupted scaled monitoring of
several water quality parameters (WQPs) over large water bodies at spatiotemporal scales
(Fulazzaky et al., 2010; Prabu et al., 2011).
In the last two decades, water quality monitoring of the urban water bodies has been
the focus of research for researchers across the globe. The qualitative assessment of river
water quality is carried out in terms of its physical, chemical and biological parameters
and involves the analysis of complicated data matrix with large number of water quality
attributes. Many studies concentrated on evaluating pollution levels in terms of individ-
ual WQPs, namely electrical conductivity (EC), turbidity, dissolved oxygen (DO), total
dissolved solids (TDS), biochemical oxygen demand (BOD), chemical oxygen demand
(COD), alkalinity, total suspended sediment (TSS), chlorophyll-a (Chl-a), and heavy met-
als such as Iron (Fe), magnesium (Mg), chromium (Cr) and lead (Pb) by utilizing remote
sensing data in geographical information system (GIS) framework (Milanović Pešić et al.,
2020; Nas et al., 2010; Sharma et al., 2018; Waxter, 2014; Yao et al., 2020). To reduce
the number of WQPs in the analysis, a lot of consideration has been given to the devel-
opment of single numerical indicators to ascertain the overall water quality trends with
respect to the threshold limits. The water quality index (WQI) is a numeric indicator of the
degree of severity in the quality of water for practical usage within the prescribed range
and is computed by considering several significant quality parameters (Bordalo et al.,
2006; Dunca, 2018; Markogianni et al., 2014; Mohamed et al., 2019; Said & Hussain,
2019; Sharaf, 2017; Sharma et al., 2018; Syahreza et al., 2012; Zhu, 2013). To classify the
degree of severity, WQI is grouped into broad classes, i.e., excellent, good, moderate, poor,
etc. For assessing the quality of any water body, numerous water quality indices have been
proposed. Most commonly utilized WQIs are weighted arithmetic index method (Brown
et al., 1970), national sanitation foundation water quality index (NSFWQI) (Hoseinzadeh
et al., 2014), overall index of pollution (OIP) (Sargaonkar & Deshpande, 2003), etc. The
OIP furnishes an in-depth understanding of the water quality status of the surface water
sources, especially under Indian conditions (Sargaonkar & Deshpande, 2003). Remote
sensing of water quality involves visible and infrared portion of the electromagnetic spec-
trum to explore the sensitivity of spectral band combinations by utilizing advanced com-
puting techniques. Several data-driven approaches have been implemented to quantify the
relationship between actual and modeled WQPs for qualitative modeling of water quality
and requires input data, model parameters, and other relevant information (Bordalo et al.,
2006). Many studies employed statistical approaches to explore linear correlations, such
as MLR, logarithmic relation and exponential relation, while others concentrated on more

13
18254 S. Said, S. A. Khan

efficient, nonlinear analytical methods, viz. artificial neural network (ANN), genetic pro-
gramming (GP), group method of data handling (GMDH), GEP, etc., in conjunction with
geospatial techniques (Akbal et al., 2011; Avdan et al., 2019; Boyacioglu, 2010; Chapagain
et al., 2010; Hussain et al., 2008; Lotfinasabasl et al., 2018).
In recent years, ANN modeling has been widely utilized to quantify the severity of
water quality issues due to its fast training process and ability to solve linear and nonlinear
complex problems (Bonansea et al., 2015; Nasri, 2010; Nathan et al., 2017). Many studies
utilized the BPNN and radial basis function (RBF) neural network for evaluating water
quality and provided favorable outcomes through modeling complex nonlinear response
functions, such as spectral reflectance values and WQP estimates (Ekercin, 2007; Gür-
soy & Atun, 2019; Marquez et al., 2018; Zhang et al., 2003; Zhao et al., 2014). In river
management programs, ANNs have effectively been used to evaluate the WQI levels to
simulate wetland processes (Reynolds & Maberly, 2002; Kuo et al., 2007; Li et al., 2009;
Song et al., 2012; Wang et al., 2012). Chu et al. (2013) developed ANN model that could
effectively predict the quality of the surface water bodies and introduced the factor analy-
sis technique to identify significant water quality parameters. In another study conducted
by Hafeez et al. (2018), four machine learning approaches, namely artificial neural net-
work (ANN), random forest (RF), cubist regression (CB) and support vector regression
(SVR), were compared for retrieval of water quality indicators (i.e., Chl-a, SS and turbid-
ity) over the coastal waters of Hong Kong by employing water reflectance values acquired
from hand-held spectroradiometer and satellite data. Results revealed ANN as the best per-
former than other three approaches. More recent studies conducted by Wang et al. (2019,
2020) inferred deep learning process as a promising tool for formulating environmental
property prediction models for screening of green solvents. Several studies successfully
applied GEP, along with GP, to a variety of water resources issues (Azamathulla & Ghani,
2011; Ghavidel & Montaseri, 2014; Liu & Wang, 2019; Zakaria et al., 2010). Furthermore,
these techniques have been considered as substantial tools in solving complex environmen-
tal and river engineering problems (Aras et al., 2007; Chen et al., 2008; Mohammadpour
et al., 2015). Ni et al. (2012) effectively evaluated the water fluctuations in the wetlands
by utilizing the GP approach. Xu and Qin (2013) measured the agricultural water quality
through the combined application of GA and fuzzy simulation. In a significant study by
Martí et al. (2013), comparison of three approaches, namely ANN, GEP and MLR for esti-
mation of outlet dissolved oxygen in micro-irrigation, was carried out, and the outcomes
revealed GEP as the most effective approach. In a recent study carried out by Li and Wang
(2019), a reliable turbidity model was developed to predict reservoir turbidity based on
Landsat-8 satellite imagery by utilizing an MLR and GEP approach. Results revealed GEP
to be more rational and accurate for turbidity simulation. Quantification of pollution levels
in water bodies during the lockdown period worldwide forms a crucial aspect for research-
ers to interpret the short and long-term effect of the coronavirus disease 2019 (COVID-19)
on the river dynamics. It has been reported in few recent studies that the pollution level has
exceedingly reduced and most water bodies have completely been restored (Clifford, 2020;
Häder et al., 2020; Stone, 2020).
Kali River, a major source of irrigation in western Uttar Pradesh, India, has completely
deteriorated due to ever increasing disposal of municipal and industrial waste from adjoin-
ing cities. Some earlier studies suggested the river water quality as safe for irrigation pur-
poses, whereas later studies revealed river water to be severely polluted with heavy metal
concentrations exceeding far beyond the permissible limits (Mishra et al., 2015; Maurya
& Malik, 2016). The Kali River has been identified as the most critically contaminated
after Markanda River (in Haryana State) in terms of BOD levels (CPCB, 2012). Spatial

13
Remote sensing‑based water quality index estimation using… 18255

monitoring of the water quality of Kali River by employing reliable data-driven approaches
is a prerequisite to conserve and manage the river restoration process. Therefore, the main
objective of the study is to evaluate and map WQI estimates along a 6-km-long stretch
of the Kali River passing through the Aligarh district in Uttar Pradesh, India, by utiliz-
ing high-resolution IRS P6 LISS IV imagery. Eleven spectral reflectance band combina-
tions were formulated to identify the most significant band combination associated with
the observed WQI at the sampling locations. Three approaches, namely MLR, BPNN and
GEP, were employed to relate WQI as a function of most significant band combination.
The performance of three approaches was assessed by via quantitative indicators such as
coefficient of determination (R2), root mean square error (RMSE) and mean absolute error
(MAE). A one-way ANOVA (analysis of variance) test was also performed to assess signif-
icant differences among WQI estimates from the three approaches at a confidence level of
0.05. Maps depicting spatial variation of WQI levels in the river stretch were generated in
GIS framework. The present study configures the basis for policy makers and environmen-
talists to devise effective and sustainable strategies and policies to reclaim the completely
degraded river ecosystems.

2 Materials and methods

2.1 Study area

The study area, illustrated in Fig. 1, covers 6-km-long stretch of Kali River (meaning
“black” in the local language) that passes through the Aligarh district in Uttar Pradesh,
India. Study area is confined within latitude 28.11°N to 28.15°N and longitude 78.14°E
to 78.18°E at an elevation of 213 m above the mean sea level. The river had been a major
source of water for domestic as well as irrigation requirements in the past two decades.
The Kali River originates from the village of Antwada, in the Muzaffarnagar district,
Uttar Pradesh, passes through many important cities and joins the Ganges River at the city
of Kannauj in the Farrukhabad district. The river covers a total span of almost 300 km.
Large cities, including Meerut, Hapur and Bulandshahr, accommodate numerous small-
and large-scale industries along the river banks, such as sugar mills, paper mills, textile
industries, slaughterhouses and distilleries. The current status of the river justifies its name,
owing to the excessive discharge of domestic sewage and untreated industrial effluents into
the river thus, conveying more than 60 per cent of the pollution load (CPCB, 2012). Over
the years, the river has completely transformed into a highly toxic flow of chemicals, harm-
ful for human consumption, and offers a restricted use for irrigation or any other purpose.
Toxic water from the Kali River is widely consumed for fulfilling the irrigation require-
ments of surrounding areas. The present condition of the river is pity and demands imme-
diate attention for its reclamation.

2.2 Data collection and analysis

River water samples were collected from the midstream at a depth of 0.5 m on April 27,
2018, concurrent to the date of satellite overpass. Grab sampling procedure was adopted
for the analysis of various WQPs as recommended by the standard methods of analysis
(APHA, 1998). Water samples from the Kali River were analyzed in the laboratory of
Environmental Engineering, Civil Engineering Department, AMU, Aligarh, and the WQI

13
18256 S. Said, S. A. Khan

Fig. 1  Location map of the study area (map not to scale)

for each sampling location was estimated from 11 physicochemical parameters and heavy
metals, namely pH, EC, DO, TDS, BOD, COD, alkalinity, Fe, Mn, Cr and Pb. The heavy
metal concentration was measured by adopting American Society for Testing and Materi-
als (ASTM, 2000) procedure involving the digestion of water samples with concentrated
HNO3 and employing an atomic absorption spectrophotometer (AAS).
WQI values for 40 water samples were computed by following a three-step procedure
(Water programme, 2007). The first step assigns weight (wi) to all the WQPs ranging from
1 to 5 in accordance with their relative significance towards the overall quality grading
of the water for irrigation purposes. The relative significance among WQPs was decided

13
Remote sensing‑based water quality index estimation using… 18257

on the basis of collective expert opinions taken from different published studies (Ram-
akrishnaiah et al., 2009; Nabizadeh et al., 2013; Suneetha et al., 2015). The highest weight
value, i.e., 5, was assigned to two heavy metals, i.e., Pb and Cr, on account of their promi-
nence towards rendering severity to the water quality. Lower rank of 1 was assigned to pH,
and 2 was assigned to COD and BOD. Ranks 3 and 4 were appropriately assigned to alka-
linity, TDS, DO, EC, Fe and Mn on the basis of their relative severity (Srinivasamoorthy
et al., 2008). The second step computes the relative weight (Wi) as per the equation below.
�
Wi = wi ∑n w , (1)
i
i=1

where Wi is the relative weight, wi is the individual parameter weight, and n is the number
of parameters. In the third step, a quality rating scale (qi) for each parameter was evaluated
by dividing its concentration levels for every water sample by its corresponding standard
concentration, as per the Bureau of Indian Standards (BIS, 1986).
/
qi = Ci S × 100, (2)
i

where qi is the quality rating in percent, Ci is the concentration of each chemical param-
eter in each water sample in mg/L, and Si is the irrigation water quality standard for each
chemical parameter in mg/L. Finally, the WQI for each sampling location was computed as
per Brown et al. (1970) expressed as Eq. 3, where SLi is the product of Wi and qi.
n
∑
WQI = SLi (3)
i=1

The WQI values corresponding to the sampling locations were evaluated by following
the above procedure and scaled for quality rating in accordance with BIS (1986) specifica-
tions, provided in Table 1.

2.3 Remote sensing data used

Image from IRS P6 Resourcesat-2 LISS IV sensor of April 27, 2018, was utilized in
the present study for evaluating and mapping the water quality of Kali River in terms
of WQI measures. The study area was delineated, and a subset image was created using
the Erdas Imagine software, shown in Fig. 2. IRS LISS IV sensor produces a high-
resolution multispectral image in three bands (i.e., green, red and near Infrared) with
5.8 m spatial resolution in the multispectral mode at nadir. Corresponding to the sam-
pling locations, pixel values with reference to digital numbers (DN) from three spectral
bands were extracted and converted into physical quantities (e.g., radiance) and then

Table 1  WQI and corresponding S no WQI Status Possible usages

water quality rating as per the
BIS (1986) specifications
1 0–50 Excellent Drinking, irrigation and industrial
2 50–100 Good Domestic, irrigation and industrial
3 100–200 Poor Irrigation
4 200–300 Very poor Restricted use for all purposes
5 > 300 Severe Proper treatment required before use

13
18258 S. Said, S. A. Khan

Fig. 2  Subset image of study area with sampling locations along the river stretch

into spectral reflectance. The process takes into account the terrain and atmospheric
corrections. The conversion involved the utilization of the radiometric “gain and offset”
extracted from the image metadata and employed Eqs. 4 and 5 for radiance and reflec-
tance, respectively, proposed by Chander and Markham (2003)
L𝜆 = Gain𝜆 × DN𝜆 + offset𝜆 , (4)
where λ is the specific spectral band of the image; Lλ is the spectral radiance for band λ at
the sensor’s aperture (mW/cm2/µm/str); ­gainλ is the radiometric calibration gain (mW/cm2/
µm/str/DN) for band λ from product metadata (gain values for three bands were consid-
ered: G = 52, R = 47 and NIR = 31); ­DNλ is digital number value for band λ of the image;
and ­offsetλ is the radiometric calibration (mW/cm2/µm/str) for band λ from product meta-
data, which is zero for the three bands

𝜋 × L𝜆 × d2
𝜌P = , (5)
ESUN × Cos𝜃S

where ρP is the dimensionless planetary reflectance, d is the Earth–Sun distance (astro-

nomical units, 1 − (0.01674 cos (0.9856 (JD-4)))2, where JD is Julian Day), ESUNλ is the
average solar exo-atmospheric spectral irradiances (mW/cm2/µm) at 1 astronomical unit
(AU) distance between the Earth and Sun, θs is the Sun’s zenith angle (~ 67.337461° from
product metadata), and Lλ is the spectral radiance for band λ at the sensor’s aperture (mW/
cm2/µm/str).

13
Remote sensing‑based water quality index estimation using… 18259

2.4 Modeling approaches

2.4.1 Multiple linear regression (MLR)

MLR analysis predicts the unknown variable from two or more known variables that are
termed as the predictors. In other words, a multiple regression analysis aids in predict-
ing the Y value for given X1, X2, …, Xk values. The multiple regression equation of Y
with known X1, X2, …, Xk is given by
Y = b0 + b1 X1 + b2 X2 + ⋯ + bk Xk , (6)
where b0 is the intercept and b1, b2, b3, …, bk are the regression coefficients that correspond
to the slope in a linear regression equation. An MLR was employed to examine the most
appropriate formulated spectral reflectance band combination, producing WQI estimates
with high R2 values and low RMSE and MAE values.

2.4.2 Artificial neural network (ANN)

The feed forward backpropagation neural network (FF-BPNN) algorithm looks for the
least error function in weight space by employing the gradient descent method. The
learning process resolves the complexity of the problem through randomly assigning
weights that produce the least error function. The entire process is executed in two
phases. In the first phase, assigned weights to the network architecture are initialized
randomly to propagate forward, along with input data, to compute the target value. In
the second phase, the error between the actual and estimated targets is compared and
the error value that is higher than the threshold value is rolled backward through the
network. The weight values are recalculated, and the process is continued until the mini-
mum error is attained. During the training process, the errors for both training and test-
ing data decrease with number of iterations until a constant minimum error value is
attained. Training is stopped at a point when, the least difference between training and
testing data errors is observed so as to avoid overtraining of the network (Said et al.,
2008). The most general neural network architecture consists of three layers, i.e., input,
hidden and output layers, as illustrated in Fig. 3.
Every unit in a layer is connected with units in the adjoining layer with a unique weight
value. Variables in the input layer, along with connected weights, propagate to every unit
of the next hidden layer. The end product of every unit forming an output is compounded
with weights of preceding connecting units and is advanced to the successive layer before
finally being subjected to the sigmoid activation function. The output value from the jth unit
of layer m is represented as
( )
6Om (7)
m
j = S lj ,

where S is the sigmoid activation function, as proposed by Rumelhart et al. (1986), and
1
f (x) =
1 + exp(−x)
. (8)

The function f(x) acquires values from zero to unity for the entire range of inputs; x is
the input value, viz. ljm obtained for of layer m, as

13
18260 S. Said, S. A. Khan

Fig. 3  Neural network architecture with input variables as bands/band combinations and WQI as target var-
iable

∑
ljm = Om−1
i wij + bm
j , (9)

where bm j
is the threshold value of the jth unit of layer m. Om−1
j
and wij are the outputs of the
ith unit of layer m − 1 and the weight of the connection between ith and jth units of layers
m − 1 and m, respectively. The error function is expressed as
∑( )2
E = 0.5 Tk − Ok , (10)
k

where Tk is the desired target value and Ok is the corresponding output value determined
for k training samples.

2.4.3 Gene expression programming (GEP)

GEP, proposed by Ferreira (2001), is an evolutionary technique that has the advantage of
solving complex nonlinear problems based on the GP approach developed by Koza (1999).
GEP is an improved version of GA and GP that overcomes premature convergence and a
100 times higher evolution rate. GEP undergoes a continuous evolution process with the
random propagation of an initial population comprising of individual chromosomes of pre-
defined length containing one, or more than one, gene. The structure of genes comprises a
head and a tail. The head consists of both functions and terminals, whereas the tail holds
only terminals. For reaching an optimal solution to the defined problem, the head length h
is selected; further, the tail length t is related to h, and the function is evaluated by using
Eq. 11 below:
t = h × (n − 1) + 1, (11)
where n is the number of arguments of the function. Ferreira (2001) represented the
encoded genetic information in the gene in the form of an expression tree (ET). With the
help of the unequivocal Karva language, the gene composition of a given ET can be gen-
eralized on the basis of simple rules of top–down and right–left (Li and Wang, 2019). An

13
Remote sensing‑based water quality index estimation using… 18261

example of a gene is shown in the form of an ET in Fig. 4, for which an equivalent math-
ematical expression is encoded as [(b × a) × (b + a)] + [(a/b) × (b − a)].
The fitness of every chromosome i in the initial population is computed by utilizing the
fitness function fi expressed as Eq. 12, proposed by Ferreira (2001).
Ci ( )
∑ | |
fi = M − |Cij − Tj |
| | (12)
j=1

where M is the selection range, Ci,j is the value recalled by the ith chromosome for the jth
fitness case, and Tj is the target value for the jth fitness case. It is to be noted that, for a
perfect fit, Ci,j = Tj and fi = fmax = Ct × M. Fitness function resolves the selection of the opti-
mal chromosomes for the next generation level through modifications achieved by genetic
operators such as mutation, inversion, transposition and recombination.
Mutation is the most effective genetic operator that represents the probability of a func-
tion or a variable (symbol) to get mutated in each generation. Any symbol in the gene
heads can be replaced by a terminal function; however, in the gene tails, terminals can be
replaced by variables only, since there is no function in the tail. Inversion chooses a ran-
dom starting as well as ending symbol in a gene, which is then reversed in order. Transpo-
sition involves actuating a sequence of symbols from one position to another within a gene
or from one gene to another gene in the same chromosome. In a recombination stage, two
new chromosomes are developed by the exchange of genetic information through random
selection. The process is analogous to the breeding of two biological species that produces
a new offspring sharing genetic material from both parents. Figure 5 illustrates the general-
ized process of GEP model building in the form of a flowchart.
Table 2 depicts 11 spectral reflectance bands/band combinations (including three inher-
ent single bands, i.e., green, red and infrared) formulated to explore the most significant
band combination related to the observed WQI estimates. As described in the preceding
sections, WQI estimates as a function of most sensitive spectral band combination were
examined via three approaches and the performance were compared using R2, RMSE and
MAE (quantitative indicators). Out of 40 data samples in total, 80% were used for training
and testing or calibration and the remaining (20%) were used for validation. Neural net-
work architectures were developed in accordance with the band combinations, i.e., 2, 3, 4,
5 and 6 spectral bands/combinations as input variables. The same band combinations were
analyzed for MLR and GEP approaches, keeping WQI as target variable. For BPNN and

Fig. 4  An example of gene ET

13
18262 S. Said, S. A. Khan

No
Proceed?

Yes

Fig. 5  Flowchart illustrating the process of GEP model building

Table 2  Formulated band combinations with details of BPNN architectures

Band combi- Input/independent variables Target variable Network archi- Learning rate
nation cases tecture I–H–Oa
No Description

1 2 G, R WQI 2–2-1 0.058

2 2 G, NIR 2–2-1 0.072
3 2 R, NIR 2–3-1 0.069
4 3 G, R, NIR 3–4-1 0.087
5 4 G, R, NIR, G/R 4–5-1 0.055
6 4 G, R, NIR, G/NIR 4–3-1 0.047
7 4 G, R, NIR, R/NIR 4–6-1 0.025
8 5 G, R, NIR, G/R, G/NIR 5–4-1 0.065
9 5 G, R, NIR, G/R, R/NIR 5–6-1 0.046
10 5 G, R, NIR, G/NIR, R/NIR 5–7-1 0.095
11 6 G, R, NIR, G/R, G/NIR, R/NIR 6–8-1 0.075
a
I–H–O: input–hidden layer neurons–output

13
Remote sensing‑based water quality index estimation using… 18263

GEP analysis, the entire data set was normalized to lie within 0 to 1 range by using Eq. 13
below (Rajurkar et al., 2004), to ensure that data are logically structured and proportionally
scaled.
( )
Xi
Xnorm = 0.1 + 0.8 × , (13)
Xmax

where Xnorm is the normalized, unitless variable; Xi is the observed variable; and Xmax is
the maximum value in the data range. The optimal count of neurons in the hidden layer
was ascertained by a hit-and-trial procedure. The learning rate for BPNN was gradually
varied within the defined range of 0.01 to 0.5. The final values of the learning rate and the
optimum count of neurons in the hidden layer obtained by the trial process are provided in
Table 2.
Further, for building the optimal GEP model, the number of chromosomes or popula-
tion size after many trials was selected as 50, the gene head length was selected as 14,
and the number of genes per chromosome was selected as 8. Seven necessary function
operators, i.e., + , − , × , ÷ , 1/a, − a, a2, were adopted for building the simplified GEP model
with a reduced iteration process as well as nonconvergence occurrences. Furthermore, sub-
gene ETs were linked by an addition function. The parameters adopted for the optimal GEP
model for precise evaluation of WQI levels are illustrated in Table 3.

3 Results

In situ water samples collected were analyzed for 11 WQPs in the laboratory, and the basic
descriptive statistics of the samples are summarized in Table 4. The physicochemical and
heavy metal concentrations ranged far beyond the permissible limits prescribed under BIS

Table 3  Parameters adopted for Parameter Value

the optimal GEP model
Population size 50
Genes per chromosomes 8
Gene head length 14
Maximum generations 5000
Fitness function R2
Precision (hit tolerance) 0.01
Mutation rate 0.054
Inversion rate 0.1
Computational functions + , − , × ,/,1/a,a2, − a
Addition, subtraction, multiplica-
tion, division, inverse, square,
negation
Linking function addition
IS transposition rate 0.1
RIS transposition rate 0.1
Gene transposition rate 0.1
Recombination one-point rate 0.3
Recombination two-point rate 0.3

13
 18264

13
Table 4  Descriptive statistics of the measured WQPs
S. no Water quality param- Range min–max Mean Standard devia- Population vari- Sample variance ( σ2) SEMa BIS stand-
eters (mg/l) tion ( σ) ance (s2) ard (BIS,
1986)

1 pH 6.93–7.56 7.33 0.12 0.0147 0.0141 0.0242 6.5–8.5

2 EC (μs/cm) 1083–1852 1644 190.61 36,334 34,881 38.12 300
3 TDS 754–851 800.32 20.06 402.56 386.53 4.013 500
4 Alkalinity 540–680 616.68 40.39 1631.56 1566.3 8.078 200
5 COD 72.15–91.20 81.55 6.10 57.38 49.32 1.515 250
6 DO 2.08–6.74 5.85 1.04 1.089 0.882 0.211 5
7 BOD 23.20–38.50 29.61 4.02 16.124 15.48 0.803 5
8 Cr 0 0 0 0 0 0 0.05
9 Pb 0.19–0.24 0.21 2.21 1.4 × ­10–4 1.1 × ­10–4 0.002 0.1
10 Fe 0.01–0.03 0.02 0.01 8.7 × ­10–5 8.4 × ­10–5 0.0018 0.3
11 Mn 0 0 0 0 0 0 0.1
a
SEM standard error of means
S. Said, S. A. Khan
Remote sensing‑based water quality index estimation using… 18265

specifications, although there were no traces of Cr and Mn in all the measured samples.
The WQI values computed from nine WQPs (excluding Cr and Mn) for 40 water samples
collected along the Kali River stretch ranged between 203.7 and 262.33, and rated under
“very poor” category on the basis of BIS criteria provided in Table 1. The WQI range indi-
cates restricted use of river water almost for all purposes including irrigation. The results
of the WQI estimates from the three employed approaches, i.e., MLR, BPNN and GEP, are
illustrated in Table 5.
Results from the MLR analysis indicate that, out of 11 band combination cases ana-
lyzed, a combination of 4 bands, i.e., G, R, NIR and G/R (band combination case no. 5),
exhibited strong correlation with the observed WQI yielding R2 ~ 0.81 and low RMSE and
MAE values (i.e., 4.36 and 4.64, respectively) for calibration data. However, the same band
combination yielded WQI estimates with R2 ~ 0.6, and relatively high RMSE and MAE
values (i.e., 6.3 and 4.64) for validation data. Regression coefficients for the most signifi-
cant band combination are provided in Table 6, and the formulated regression equation
is expressed as Eq. 14. Scatter plot between the observed and estimated WQI for calibra-
tion and validation data is illustrated in Fig. 7(a), depicting estimated values of the WQI
within ± 20% error lines. The regression equation formulated for the most significant band
combination was utilized in the generation of spatially distributed WQI map of the river
segment.
( )
GREEN
WQI = −183.98 + (2309.744 × GREEN) + (297.18 × RED) + (200.93 × NIR) + 35.84 ×
RED (14)

Neural network architectures for all band combinations were trained using the
TRAINGD function and FF-BPNN algorithm. Optimal architectures were obtained during
the training process by adopting the number of neurons in the hidden layer from 2 to 10
and varying the learning rate in the defined range of 0.001 to 0.5. It was observed that neu-
ral network architectures trained with 3, 4 and 6 neurons in the hidden layer yielded much
better WQI estimates in terms of R2, RMSE and MAE values (Table 6).
Results further reveal that neural network architecture trained with 3 input bands, i.e.,
G, R and NIR, and 4 neurons in the hidden layer (i.e., 3-4-1) produced WQI estimates
with highest accuracy than the rest of combinations, yielding R2 ~ 0.95 and 0.87, RMSE
as 2.36 and 4.48, and MAE as 2.15 and 3.61 for calibration and validation data, respec-
tively. Scatter plot between the observed and estimated WQIs as shown in Fig. 7b depicted
WQI estimates within ± 10% error lines. It was also observed that almost all neural network
architectures with different band combinations conceded WQI estimates with considera-
ble accuracies for calibration data, i.e., R2 ranging from 0.92 to 0.79, respectively. Table 7
illustrates the final weight matrix for the most optimal neural network architecture (i.e.,
3-4-1) producing highest WQI retrieval accuracies.
The optimal GEP model was achieved through many trials (Table 6), comprising a
chromosomal architecture with 50 chromosomes, head length at 14 and number of genes
at 8, and 4 spectral bands as input, viz. G, R, NIR and G/R (band combination case no.
5). The optimized GEP model produced WQI estimates with considerably high accura-
cies, yielding R2 ~ 0.94 and 0.91, RMSE as 2.49 and 4.45, and MAE as 2.16 and 3.53 for
calibration and validation data, respectively. As observed from the results, GEP model
performs substantially well with validation data as compared with BPNN and MLR
models, thus indicating significant rationality in the optimized GEP model. The optimal
GEP model constitutes four subordinate expression trees (i.e., sub-ET1, sub-ET2, sub-
ET3 and sub-ET4), developed in accordance with the selection of the number of input

13
 18266

Table 5  Coefficient of determination (R2), RMSE and MAE between the observed and estimated WQIs from three approaches

13
Band combination cases 1 2 3 4 5 6 7 8 9 10 11
a
Inputs/variables X 2 2 2 3 4 4 4 5 5 5 6
b
Iterations I 35,234 28,731 32,186 33,174 36,871 44,528 47,351 51,277 49,545 55,285 57,368
2
MLR Cal (80%) R 0.46 0.66 0.65 0.72 0.81 0.77 0.56 0.59 0.51 0.32 0.27
RMSE 18.51 15.67 9.45 6.85 4.36 7.31 10.52 12.35 11.68 17.77 21.45
MAE 16.77 15.21 8.63 5.37 3.00 6.82 8.86 10.73 11.24 15.52 18.96
Val (20%) R2 0.34 0.52 0.47 0.51 0.60 0.55 0.57 0.49 0.43 0.15 0.09
RMSE 16.52 8.96 13.76 9.63 6.30 7.55 7.37 11.52 13.93 28.79 36.21
MAE 15.23 8.44 10.73 8.68 4.64 7.15 6.97 9.38 12.51 23.48 31.92
BPNN Cal Tr ­Tec R2 0.84 0.88 0.85 0.95 0.92 0.89 0.92 0.87 0.79 0.87 0.93
RMSE 3.78 3.64 3.32 2.36 2.58 3.38 3.76 4.81 5.24 4.95 3.17
MAE 3.34 3.46 2.98 2.15 2.42 3.15 3.28 4.47 4.89 4.33 2.87
Val (20%) R2 0.76 0.79 0.84 0.87 0.85 0.81 0.83 0.69 0.73 0.79 0.83
RMSE 5.21 6.87 5.91 4.48 5.85 6.76 6.62 7.23 7.17 6.95 6.34
MAE 4.85 5.54 5.73 3.61 4.97 6.16 5.65 7.12 6.78 6.23 5.92
GEP Cal (80%) R2 0.78 0.84 0.91 0.88 0.94 0.92 0.89 0.83 0.91 0.79 0.82
RMSE 6.89 6.34 5.21 5.96 2.49 3.16 3.67 7.78 3.41 7.39 6.57
MAE 5.74 5.22 4.78 5.17 2.16 2.87 2.83 6.41 2.75 5.53 5.27
Val (20%) R2 0.63 0.79 0.82 0.86 0.91 0.87 0.81 0.76 0.82 0.71 0.79
RMSE 8.87 7.21 6.87 6.34 4.45 5.92 6.54 7.11 7.08 7.87 8.29
MAE 7.34 6.88 6.12 5.43 3.53 4.58 5.71 6.51 5.24 6.19 7.33
a
X no. of input/independent variables, bI total no. of iterations for BPNN, cTr training data (60%), Te testing data (20%), Cal calibration, Val validation
‘Bold’ indicates the optimal WQI model achieved
S. Said, S. A. Khan
 Table 6  MLR coefficients for calibration data for the most appropriate band combination
Multiple regression equation
Y = bo + b1 × X1 + b2 × X2 + b3 × X3 + ⋯ + e
where Y = WQI
BCa case Nn Xb Parameter distribution Regression coefficients R2 ­(Calc) RMSE MAE R2 ­(Vald) RMSE MAE

5 04: G, R, NIR, G/R X1 = G b0 =  − 183.98 0.81 4.36 3.0 0.60 6.3 4.64
Remote sensing‑based water quality index estimation using…

X2 = R b1 = 2309.74
X3 = NIR
X4 = G/R b2 = 297.18
b3 = 200.93
b4 = 35.84
a
BC band combination,bX no. of input/independent variables, cCal calibration, dVal validation
18267

13
18268 S. Said, S. A. Khan

Table 7  Final weight matrix of Predictor variables BPNN prediction model (hidden layer)
the trained BPNN model with 3
input variables Connecting weights of 4 neurons
Input layer N1 N2 N3 N4

G − 0.234 0.046 0.824 0.147

R − 0.463 − 0.042 0.568 − 0.221
NIR 0.173 − 0.386 − 0.034 − 0.579
Bias 0.251 − 0.366 − 0.721 0.632
Target variable 0.437 − 0.254 − 0.022 0.771
(WQI) output layer

N1 neuron 1, N2 neuron 2, N3 neuron 3, N4 neuron 4

variables and function operators during model-building process. Sub-ETs were linked
together by an addition function to finally form the mathematical expression that was
further simplified to obtain more generalized form for estimating the WQI, expressed as
Eq. 15. The developed subgene ETs are shown as in Fig. 6, and a scatter plot between
the observed and estimated WQI is shown in Fig. 7c, depicting estimated WQI values
within ± 10% error lines.

Fig. 6  Expression trees for the optimal GEP model with 4 spectral bands

13
Remote sensing‑based water quality index estimation using… 18269

Fig. 7  Scatter plots between

observed and estimated WQIs
from a MLR approach for band
combination 5; 4 inputs, b BPNN
approach for band combination
4; 3 inputs and c GEP approach
for band combination 5; 4 inputs

(a)

(b)

(c)

13
18270 S. Said, S. A. Khan

( )
1.86 × 10−3
WQI = Sub ET - 1
1.31 × 10−1 − GREEN
[ ( )]
GREEN
RED 477.4 − + Sub ET - 2
RED (15)
(426.2NIR − GREEN) + Sub ET - 3
[( ) ]
GREEN
130.5 − 27.78 Sub ET - 4
RED

4 Discussion

The severe contamination of River Kali stretch assessed through the laboratory analysis
of 11 physicochemical parameters and heavy metals as well as WQI estimates is mainly
attributed to the unrestricted toxic waste disposal from numerous small- and large-scale
industries. Although several studies on Kali River water quality have predicted the simi-
lar outcomes (CPCB, 2012; Mishra et al., 2015; Singh et al., 2020; Sirohi et al., 2014), a
comprehensive monitoring of WQI levels by formulating spectral band combinations has
been lacking. Results from the three approaches further reveal that GEP outperforms the
other two approaches in terms of WQI estimates for validation data (i.e., R2 ~ 0.91, 0.87
and 0.60; RMSE ~ 4.45, 4.48 and 6.30 for GEP, ANN and MLR, respectively), suggesting
a higher measure of explanatory power possessed by this approach. Moreover, the GEP
approach is simple and produces reliable WQI measures and reduces substantial time and
effort by optimizing the computations to generate simplified prediction expressions. This
technique is highly recommended by many researchers (Hashmi et al., 2011; Mohammad-
pour et al., 2016; Liu & Wang, 2019) for the water quality evaluation of wetlands and other
surface water bodies. In addition, the ANN approach is relatively time-consuming and does
not furnish any governing equations of the optimized models, which is considered as one
of its major disadvantages. The WQI estimates predicted by MLR model were of insuffi-
cient accuracy when tested with validation data, since this approach utilizes the method of
least squares and is linear in nature. However, MLR is still practicable for its fast predicting
ability. Figure 8 depicts comparative line plots of WQI estimates for calibration and valida-
tion data, along with the observed WQI measures.
The contamination levels throughout the Kali River stretch exhibited consistency which
lead to similar spectral distribution of remotely-sensed signal above the water surface.
Therefore, WQI maps created in the GIS framework (Fig. 9) from the three approaches
corroborate to the actual severity in WQI levels, exhibited by the darker spectral tones
covering the entire length of the river segment. This severe contamination in the river is
majorly attributed to the addition of industrial effluents, agricultural runoff, natural matter
and nutrients in the water body (Jindal & Sharma, 2011).
A one-way ANOVA test for means and variance was applied to further ascertain the spatial
variability of WQI estimates from the three approaches. The null hypothesis “H0” stated “no
significant difference between means of WQI estimates from the three approaches,” whereas
alternate hypothesis “Ha” stated “significant difference between means of WQI estimates
from three approaches.” The test results unveiled F-statistic (i.e., F = 0.01 and p-value, i.e.,
p = 0.994) as exceedingly higher than the significance level α = 0.05 (Table 8), implying that
there were no critical differences in the mean values and variances of WQI estimates. The

13
Remote sensing‑based water quality index estimation using… 18271

CALIBRATION DATA VALIDATION DATA

270
260
250
240
230
WQI

220
210
200
1 3 5 7 9 11 13
15 17 19
21 23 25
27 29 31
Data points 33 35 37 39
Observed WQI
GEP modelled WQI
BPNN modelled WQI
MLR modelled WQI

Fig. 8  Comparative line plot of observed WQI and estimated WQI from the three employed approaches

ANOVA test results therefore fail to reject the null hypothesis inferring that the WQI estimates
from the three approaches are statistically “not significant.” The data set may, however, be con-
sistent with the differences of practical importance. Moreover, failing to reject the null hypoth-
esis does not necessarily imply that no potential difference in the data set exists, rather; an
increased sample size could bring out the difference. Thus, larger sample sizes allow hypoth-
esis tests to detect effects that are statistically significant. Further, to visually summarize
and compare the results, box plot of WQI estimates shown in Fig. 10, were analyzed. İt was
observed that, the respective medians of each box plot laid at the same level (i.e., 233.23 for
GEP, 232.94 for ANN and 231.96 for MLR) suggesting no likely difference between the three
estimated WQI groups. The median line of the three box plots further indicates symmetric
data representation with no right or left skewness within each of the three WQI groups. Upon
comparing the interquartile ranges, the relatively longer box corresponding to MLR revealed
slight dispersion in WQI estimates.
Overall comparison of the results indicate that GEP is much superior to MLR and ANN
approaches. Furthermore, despite the restrictive spectral resolution of IRS P6 LISS IV sensor
(i.e., comprising three bands), a combination of 4 bands (i.e., G, R, NIR, G/R) is identified as
the most effective for modeling WQI levels through GEP approach. The methodology adopted
and the WQI maps generated can be of immense help in the decision making to impose cor-
rective conservation measures for improvement in the Kali River water quality so that the river
may regain its historical importance. Moreover, the methodology can be implemented to other
contaminated surface water bodies to generalize the GEP model prediction ability.

5 Conclusions

The present study evaluates WQI levels along 6-km-long Kali River segment from three
approaches, namely MLR, BPNN and GEP, by utilizing spectral reflectance values from
high-resolution IRS P6 LISS IV image. The water samples were collected from 40 random

13
18272 S. Said, S. A. Khan

Fig. 9  WQI maps of the river

stretch generated from a MLR, b
BPNN and c GEP analysis

13
Remote sensing‑based water quality index estimation using… 18273

Table 8  One-way ANOVA test for WQI estimates from the three employed approaches
Square summation Degree of Mean of square F P-value F crit
freedom

Between groups 0.9329 2 0.4665 0.01 0.994 3.08

Within groups 9830.53 117 84.021
Total 9831.47 119

As the P-value is more than 0.05, it is not significant at 2.5% level

Fig. 10  Boxplot of WQI estimates from the three employed approaches

locations along the river stretch and analyzed for seven physicochemical and four heavy
metal concentrations (i.e., 11 WQPs in total). All measured WQP concentrations ranged
beyond the permissible limits as per BIS specifications, except for Cr and Mn, that were
found to be absent in the water samples. Further, the WQI values computed from nine
WQPs were found to range between 203.7 and 262.33, thus, designating the river con-
dition as unfit for all purposes. Eleven spectral reflectance band combinations (including
three inherent single bands) were considered to explore the sensitivity of the most signifi-
cant band/band combination with the observed WQI. The analyses of the results revealed
that GEP approach outperformed both BPNN and MLR approaches with considerably high
WQI retrieval accuracies, yielding R2 ~ 0.94 and 0.91, RMSE as 2.49 and 4.45 and MAE
as 2.16 and 3.53 for calibration and validation data, respectively. Results further revealed
that both GEP and MLR approaches identified the combination of 4 spectral bands (i.e., G,
R, NIR, G/R) as the most significant band combination for estimating WQI levels, whereas
BPNN recognized 3 band combination (i.e., G, R, NIR) as the most significant. The results
are also suggestive of the fact that machine learning approaches, viz. ANN and GEP, yield
promising potential for water quality monitoring by utilizing spectral band combinations,
wherein GEP proved to be superior. The ANOVA test revealed statistically insignifi-
cant difference among WQI estimates from the three approaches at a confidence level of
0.05, attributed to small river water sample size. The spatial distribution maps of WQI
levels exhibited uniform spectral tones in the entire river stretch, signifying the severity

13
18274 S. Said, S. A. Khan

of pollution concentrations in the river water. The study showcases the river condition as
extremely critical, requiring immediate attention of the decision makers involved in the
task of its reclamation. Future research can be focused on using hyperspectral satellite data
along with integrated approaches such as fuzzy optimal model, GP, support vector machine
(SVM) and RBF along with an increased water sample size.

Acknowledgements Authors acknowledge anonymous reviewers for their constructive comments and sug-
gestions that have substantially improved the quality of the manuscript.

References
Abba, S. I., Said, Y. S., & Bashir, A. (2015). Assessment of water quality changes at two location of Yamuna
River using the National Sanitation Foundation of water quality. Journal of Civil Engineering and
Environmental Technology, 2(8), 730–733
Akbal, F., Gürel, L., Bahadır, T., Güler, İ, Bakan, G., & Büyükgüngör, H. (2011). Multivariate statistical
techniques for the assessment of surface water quality at the mid-Black Sea coast of Turkey. Water Air
Soil Pollution, 216, 21–37
APHA. (1998). Standard methods for the examination of water and waste water. (20th ed., p. 1998). Ameri-
can Public Health Association.
Aras, E., Togan, V., & Berkun, M. (2007). River water quality management model using genetic algorithm.
Environmental Fluid Mechanics, 7, 439–450
ASTM. (2000). American society for testing and materials. (p. 20402). Published by United States Environ-
mental Protection Agency.
Avdan, Z. Y., Kaplan, G., Goncu, S., & Avdan, U. (2019). Monitoring the water quality of small water bod-
ies using high-resolution remote sensing data. International Journal of Geo-Information (MDPI), 8,
553
Azamathulla, H. M., & Ghani, A. A. (2011). Genetic programming for predicting longitudinal dispersion
coefficients in streams. Water Resources Management, 25, 1537–1544
BIS. (1986). Indian standard specification for irrigation water. IS: 11624. Indian Standard Institute, India.
Bonansea, M., María, C. R., Lucio, P., & Susana, F. (2015). Using multi-temporal landsat imagery and lin-
ear mixed models for assessing water quality parameters in Río Tercero Reservoir (Argentina). Remote
Sensing of Environment, 158, 28–41
Bordalo, A. A., Teixeira, R., & Wiebe, W. J. (2006). A water quality index applied to an international shared
river basin: The case of the Douro River. Environmental Management, 38, 910–920
Boyacioglu, H. (2010). Utilization of the water quality index methods: A classification tool. Environmental
Monitoring and Assessment, 167, 115–124
Brown, R. M., McClelland, N. I., Deininder, R. A., & Tozer, R. G. (1970). A water quality index- do we
dare? Water Sewage Works, 117(10), 339–343
Chander, G., & Markham, B. (2003). Revised Landsat-5 TM radiometric calibration procedures and post
calibration dynamic ranges. IEEE Transactions on Geoscience and Remote Sensing, 41, 2674–2677
Chapagain, S. K., Pandey, V. P., Shrestha, S., Nakamura, T., & Kazama, F. (2010). Assessment of deep
groundwater quality in Kathmandu valley using multivariate statistical techniques. Water Air Soil Pol-
lution, 210, 277–288
Chen, L., Tan, C. H., Kao, S. J., & Wang, T. S. (2008). Improvement of remote monitoring on water quality
in a subtropical reservoir by incorporating grammatical evolution with parallel genetic algorithms into
satellite imagery. Water Resources, 42, 296–306
Chu, H. B., Lu, W. X., & Zhang, L. (2013). Application of artificial neural network in environmental water
quality assessment. Journal of Agriculture Science and Technology, 15(2), 343–356
Clifford, C. (2020). The Water in Venice, Italy’s Canals Is Running Clear amid the COVID-19Lockdown—
Take a Look. Retrieved 17 April 2020 from https://​www.​cnbc.​com/​2020/​03/​18/​photos-​water-​in-​ven-
ice-​italys-​canals-​clear-​amid-​covid-​19loc​kdown.​html.
CPCB. (2012). Reconnaissance survey of pollution load of River Kali. Central Pollution Control Board.
Dunca, A. M. (2018). Water pollution and water quality assessment of major transboundary rivers from
Banat (Romania). Journal of Chemistry (Article ID 9073763).
Ekercin, S. (2007). Water quality retrievals from high resolution IKONOS multispectral imagery: A case
study in Istanbul, Turkey. Water Air Soil Pollution, 183, 239–251

13
Remote sensing‑based water quality index estimation using… 18275

Farhad Yousefabadi, L.O., Shariati, F., & Mardookhpour, A. (2013). A Comparison of water quality
indices for Haraz River. Department of Environmental Engineering Lahijan Branch, Islamic, 3(3),
30–36.
Ferreira, C. (2001). Gene expression programming: a new adaptive algorithm for solving problems.
Complex Systems, 13(2), 87–129
Fulazzaky, M. A., Seong, T. W., & Masirin, M. I. M. (2010). Assessment of water quality status for the
Selangor River in Malaysia. Water Air Soil Pollution, 205, 63–77
Ghavidel, Z. Z. S., & Montaseri, M. (2014). Application of different data-driven methods for the predic-
tion of total dissolved solids in the Zarinehroud basin. Stochastic Environmental Research and Risk
Assessment, 28, 2101–2118
Gürsoy, Ö., & Atun, R. (2019). Investigating surface water pollution by integrated remotely sensed and field
spectral measurement data: A case study. Polish Journal of Environmental Studies, 28, 2139–2144
Häder, D. P., Banaszak, A. T., Villafañe, V. E., Narvarte, M. A., González, R. A., & Helbling, E. W.
(2020). Anthropogenic pollution of aquatic ecosystems: Emerging problems with global implica-
tions. Science of Total Environment, 713, 136586
Hafeez, S., Wong, M. S., Ho, H. C., Nazeer, M., Nichol, J., et al. (2018). Comparison of machine learn-
ing algorithms for retrieval of water quality indicators in case-II waters: A case study of Hong
Kong. Remote Sensing, 11(684), 1–26
Hashmi, M. Z., Shamseldin, A. Y., & Melville, B. W. (2011). Statistical downscaling of watershed pre-
cipitation using gene expression programming (GEP). Environmental Modelling and Software, 26,
1639–1646
Hoseinzadeh, E., Khorsandi, H., Wei, C., & Alipour, M. (2014). Evaluation of aydughmush river water
quality using the national sanitation foundation water quality index (NSFWQI), river pollution
index (RPI), and forestry water quality index (FWQI). Desalination and Water Treatment, 54,
2994–3002
Hussain, M., Ahmed, S. M., & Abderrahman, W. (2008). Cluster analysis and quality assessment of logged
water at an irrigation project, eastern Saudi Arabia. Journal of Environment Mangement, 86(1),
297–307
Jindal, R., & Sharma, C. (2011). Studies on water quality of Sutlej River around Ludhiana with reference
to physicochemical parameters. Environmental Monitoring and Assessment, 174(1–4), 417–425
Koza, J. R. (1999). Genetic programming: On the programming of computers by means of natural selec-
tion. The MIT Press.
Kuo, J., Hsieh, M., Lung, W., & She, N. (2007). Using artificial neural network for reservoir eutrophica-
tion prediction. Ecological Modelling, 200, 171–177
Li, H., Liu, C. G., Fan, J., et al. (2009). Application of back-propagation neural network for predicting
chlorophyll-A concentration in rivers. China Water and Waste Water, 25(5), 75–79
Liu, L. W., & Wang, Y. M. (2019). Modelling reservoir turbidity using Landsat 8 satellite imagery by
gene expression programming. Water (MDPI), 11, 1479. https://​doi.​org/​10.​3390/​w1107​1479
Lotfinasabasl, S., Gunale, V. R., & Khosroshahı, M. (2018). Applying geographic information systems
and remote sensing for water quality assessment of mangrove forest. ActaEcologicaSinica, 38, 135
Markogianni, V., Dimitriou, E., & Karaouzas, I. (2014). Water quality monitoring and assessment of
an urban Mediterranean lake facilitated by remote sensing applications. Environmental Monitoring
and Assessment, 186(8), 5009–5026
Marquez, L. C. G., Bejarano, F. M. T., Espınoza, A. C. T., & Rodríguez, I. R. H. (2018). Use of LAND-
SAT 8 images for depth and water quality assessment of El Guájaro reservoir, Colombia. Journal of
South American Earth Sciences, 82, 231
Martí, P., Shiri, J., Duran-Ros, M., Arbat, G., De Cartagena, F. R., & Puig-Bargués, J. (2013). Artificial
neural networks vs. gene expression programming for estimating outlet dissolved oxygen in micro-
irrigation sand filters fed with effluents. Computers and Electronics in Agriculture, 99, 176–185
Maurya, P. K., & Malik, D. S. (2016). Distribution of heavy metals in water, sediments and fish tissue
(Heteropneustisfossilis) in Kali River of western UP India. International Journal of Fisheries and
Aquatic Studies, 4(2), 208–215
MilanovićPešić, A., Brankov, J., & MilijaševićJoksimović, D. (2020). Water quality assessment and
populations’ perceptions in the National park Djerdap (Serbia): Key factors affecting the envi-
ronment. Environment Development and Sustainability, 22, 2365–2383. https://​doi.​org/​10.​1007/​
s10668-​018-​0295-8
Mishra, S., Kumar, A., Yadav, S., & Singhal, M. K. (2015). Assessment of heavy metal contamination in
Kali river, Uttar Pradesh, India. Journal of Applied and Natural Science, 7(2), 1016–1020

13
18276 S. Said, S. A. Khan

Mohamed, E., Ioannis, G., Anas, O., Jarbou, B., & Petros, G. (2019). Assessment of water quality param-
eters using temporal remote sensing spectral reflectance in arid environments Saudi Arabia. Water,
11(3), 556
Mohammadpour, R., Shaharuddin, S., Chang, C., Zakaria, N., Ghani, A. A., & Chan, N. (2015). Predic-
tion of water quality index in constructed wetlands using support vector machine. Environmental
Science and Pollution Research, 22, 6208–6219
Mohammadpour, R., Shaharuddin, S., Zakaria, N.A., Ghani, A. A., Vakili, M., & Chan, N. W. (2016).
Prediction of water quality index in free surface constructed wetlands. Environmental Earth Sci-
ences, 75, 139. https://​doi.​org/​10.​1007/​s12665-​015-​4905-6.
Mondal, M., & Satpati, L. (2020). Human intervention on river system: a control system—A case study
in Ichamati River, India. Environment Development and Sustainability, 22, 5245–5271. https://​doi.​
org/​10.​1007/​s10668-​019-​00423-3
Nabizadeh, R., Amin, M. V., Alimohammadi, M., Naddafi, K., Mahvi, A. H., & Yousefzadeh, S. (2013).
Development of innovative computer software to facilitate the setup and computation of water qual-
ity index. Journal of Environmental Health Science and Engineering, 11, 1
Nas, B., Ekercin, S., Karabork, H., Berktay, A., & Mulla, D. J. (2010). An application of Landsat-5TM
image data for water quality mapping in lake Beysehir, Turkey. Water Air Soil Pollution, 212,
183–197
Nasri, M. (2010). Application of Artificial Neural Networks (ANNs) in prediction models in risk man-
agement. World Applied Science Journal, 10(12), 1493–1500
Nathan, N. S., Saravanane, R., & Sundararajan, T. (2017). Application of ANN and MLR models on
groundwater quality using CWQI at Lawspet, Puducherry in India. Journal of Geoscience and
Environment Protection, 5, 99–124. https://​doi.​org/​10.​4236/​gep.​2017.​53008
Ni, Q., Wang, L., Zheng, B., & Sivakumar, M. (2012). Evolutionary algorithm for water storage fore-
casting response to climate change with small data sets: The Wolonghu Wetland, China. Environ-
mental Engineering and Science, 29, 814–820
Prabu, P. C., Wondimu, L., & Tesso, M. (2011). Assessment of water quality of Huluka and Alaltu Riv-
ers of Ambo, Ethiopia. Journal of Agricultural Science and Technology, 13(1), 131–138
Rajurkar, M. P., Kothyari, U. C., & Chaube, U. C. (2004). Modelling of Daily Rainfall runoff relation-
ship with artificial neural network. Journal of Hydrology, 285, 96–113.
Ramakrıshnaıah, C. R., Sadashıvaıah, C., & Ranganna, G. (2009). Assessment of water quality index for
the groundwater in Tumkur Taluk, Karnataka State, India. E-Journal of Chemistry, 6(2), 523–530
Reynolds, C. S., & Maberly, S. C. (2002). A simple method for approximating the supportive capacities
and metabolic constraints in lakes and reservoirs. Freshwater Biology, 47(6), 1183–1188
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning representations by back propagating
errors. Nature, 323, 533–536
Said, S., & Hussain, A. (2019). Pollution mapping of Yamuna river segment passing through Delhi
using high resolution GeoEye-2 imagery. Applied Water Science, 9, 46. https://​doi.​org/​10.​1007/​
s13201-​019-​0923-y
Said, S., Kothyari, U. C., & Arora, M. K. (2008). ANN-based soil moisture retrieval over bare and veg-
etated areas using ERS-2 SAR data. Journal of Hydrologic Engineering, 13(6), 461–475
Sargaonkar, A., & Deshpande, V. (2003). Development of an overall index of pollution for surface water
based on a general classification scheme in Indian context. Environmental Monitoring and Assess-
ment, 89, 43–67
Sharaf Essam, E. D., Zhang, Y., & Suliman, A. (2017). Mapping concentrations of surface water quality
parameters using a novel remote sensing and artificial intelligence framework. International Jour-
nal of Remote Sensing, 38(4), 1023–1042
Sharma, G., Said, S., & Hussain, A. (2018). Water quality mapping of Yamuna River stretch passing
through Delhi sate using high resolution GeoEye-2 imagery. International Journal of Applied Geo-
spatial Research, 9(4), 23–35
Singh, G., Patel, N., Jindal, T., Srivastava, P., & Bhowmik, A. (2020). Assessment of spatial and tem-
poral variations in water quality by the application of multivariate statistical methods in the Kali
River, Uttar Pradesh, India. Environmental Monitoring and Assessment, 192, 394
Sirohi, S., Sirohi, S. P. S., & Tyagi, P. K. (2014). Impact of industrial effluents on water quality of Kali
River in different locations of Meerut, India. Journal of Engineering Technology and Research, 6,
4347
Song, K. S., Li, L., Li, S., Tedesco, L., Hall, B., & Li, L. H. (2012). Hyperspectral remote sensing of
total phosphorus (TP) in three central Indiana water supply reservoirs. Water Air Soil and Pollu-
tion, 223, 1481–1502

13
Remote sensing‑based water quality index estimation using… 18277

Srinivasamoorthy, K., Chidambaram, M., Prasanna, M. V., Vasanthavigar, M., John Peter, A., & Anand-
han, P. (2008). Identification of major sources controlling Groundwater Chemistry from a hard rock
terrain—A case study from Mettur taluk, Salem district, Tamilnadu. India. Journal of Earth System
Sciences, 117(1), 49–58
Stone, M. (2020). Carbon emissions are falling sharply due to coronavirus. But not for long. Retrieved
17 April 2020 from https://​www.​natio​nalge​ograp​hic.​com/​scien​ce/​2020/​04/​co-​ronav​irus-​causi​ng-​
carbon-​emiss​ions-​to-​fall-​but-​not-​for-​long/.
Suneetha, M., SyamaSundar, B., & Ravindhranath, K. (2015). Calculation of water quality index (WQI) to
assess the suitability of groundwater quality for drinking purposes in Vinukonda Mandal, Guntur Dis-
trict, Andhra Pradesh, India. Journal of Chemical and Pharmaceutical Research, 7(9), 538–545
Syahreza, S., MatJafri, M. Z., & Lim, H. S. (2012). Water quality assessment in Kelantan delta using remote
sensing technique. Proceedings of SPIE 8542, Electro-Optical Remote Sensing, Photonic Technologies
and Applications VI, 85420X.https://doi.org/https://​doi.​org/​10.​1117/​12.​978931
Wang, L., Li, X., & Cui, W. (2012). Fuzzy neural networks enhanced evaluation of wetland surface water
quality. International Journal of Computation and Applied Technology, 44, 235–240
Wang, Z., Su, Y., Jin, S., Shen, W., Ren, J., Zhang, X., & Clark, J. H. (2020). A novel unambiguous strategy
of molecular feature extraction in machine learning assisted predictive models for environmental prop-
erties. Green Chemistry, 22, 3867–3876
Wang, Z., Su, Y., Shen, W., Jin, S., Clark, J. H., Ren, J., & Zhang, X. (2019). Predictive deep learning
models for environmental properties: the direct calculation of octanol–water partition coefficients from
molecular graphs. Green Chemistry, 21, 4555–4565
Water Programme. (2007). Global drinking water quality index development and sensitivity analysis. In
Report of United Nations Environment Programme and Global Environment Monitoring System.
(GEMS)/Water Programme.
Waxter, M. T. (2014). Analysis of Landsat satellite data to monitor water quality parameters in Tenmile
Lake, Oregon. MSc. Thesis, Portland State University.
Xu, T. Y., & Qin, X. S. (2013). Solving water quality management problem through combined genetic algo-
rithm and fuzzy simulation. Journal of Environmental Information, 22, 39–48
Yao, H., Ni, T., & Zhang, T. (2020). Estimation of phosphorus flux into the sea through one reversing river
using continuous turbidities and water quality modeling. Environment Development and Sustainability,
22, 4251–4265. https://​doi.​org/​10.​1007/​s10668-​019-​00382-9
Zakaria, N. A., Azamathulla, H. M., Chang, C. K., & Ghani, A. A. (2010). Gene expression program-
ming for total bed material load estimation—A case study. Science of the Total Environment, 408,
5078–5085
Zhang, Y., Pulliainen, J. T., Koponen, S. S., & Hallikainen, M. T. (2003). Water quality retrievals from com-
bined Landsat TM Data and ERS-2 SAR data in the Gulf of Finland. IEEE Transactions on Geosci-
ence and Remote Sensing, 41, 622–629
Zhao, F., Zhu, F. Q., & Feng, Z. K. (2014). Study on water body information extraction method based on
ZY-3 Imagery. Bulletin of Surveying and Mapping, 3, 007
Zhu, L. (2013). Water quality analysis and evaluation of current situation in the riparian of west lake Taihu
in Yixing. Nanjing Forestry University.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

13