Influence of COVID-19 lockdown on river water quality and assessment of environmental health in an industrialized belt of southern Western Ghats, India

Abstract

The COVID-19 pandemic and sudden lockdown have severely hampered the country’s economic growth and socio-cultural activities while imparting a positive effect on the overall fitness of the environment especially air and water resources. Increased urbanization and rapid industrialization have led to rising pollution and deterioration of rivers and associated sectors such as agriculture, domestic and commercial needs. However, various available studies in different parts of the country indicate that the COVID-19 pandemic has changed the entire ecosystem. But it is noted that studies are lacking in the southern Western Ghats region of India. Therefore, the present study attempts to investigate how the continuous lockdowns affect the River Water Quality (RWQ) during lockdown (October 2020) and post-lockdown (January 2021) periods in the lower catchments (Eloor-Edayar industrialized belt) of Periyar river, Kerala state, South India. A total of thirty samples (15 samples each) were analyzed based on drinking water quality, irrigational suitability, and multivariate statistical methods to evaluate the physical and chemical status of RWQ. The results of the Water Quality Index (WQI) for assessing the drinking water suitability showed a total of 93% of samples in the excellent and good category during the lockdown, while only 47% of samples were found fit for drinking during the post-lockdown period. Irrigational suitability indices like Mg hazard, KR, PI, SAR, and Wilcox diagram revealed lockdown period samples as more suitable for irrigational activities compared to post-lockdown samples with site-specific changes. Spearman rank correlation analysis indicated EC and TDS with a strong positive correlation to Ca²⁺, Mg²⁺, Na⁺, K⁺, TH, SO₄²⁻, and Cl⁻ during both periods as well as strong positive correlations within the alkaline earth elements (Ca²⁺ and Mg ²⁺) and alkalis (Na⁺ and K⁺). Three significant components were extracted from principal component analysis (PCA), explaining 88.89% and 96.03% of the total variance for lockdown and post-lockdown periods, respectively. Variables like DO, BOD, Ca²⁺, NO₃⁻, and Cl⁻ remained in the same component loading during both periods elucidating their natural origin in the basin. The results of health risk assessment based on US EPA represented hazard quotient and hazard index values below the acceptable limit signifying no potential noncarcinogenic risk via oral exposure except As, suggesting children as more vulnerable to the negative effects than adults. Furthermore, this study also shows rejuvenation of river health during lockdown offers ample scope to policymakers, administrators and environmentalists for deriving appropriate plans for the restoration of river health from anthropogenic stress.

Introduction

Coronavirus (COVID-19) pandemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has brought a tremendous change to the entire global economic activities through complete and continuous lockdowns imposed by the governments around the globe. The USA, Italy, and Spain were the worst affected countries with 2 million cases of fatalities reported all over the world (Yunus et al. 2020). Since the new disease resembles the attributes of a global disaster, on 11 March 2020, the World Health Organization (WHO, 2020) declared COVID-19 disease a pandemic with its fast outbreak, geographical expansion, and complex consequences (Manoiu et al. 2022). Initially reported in Wuhan, China, in December 2019, the transmission of disease occurred from person to person through the respiratory droplets of an infected person. Since no vaccine or medicine was available at first, WHO and other governmental bodies recommended the people for practicing social distancing and self-isolation. Numerous countries and territories started to impose strict rules and actions to keep their citizens at home which include shutting down schools, industries and businesses, suspending travel, and closing the international and state boundaries. All types of human activities, industrial production, trade, and traffic suddenly came to a standstill, conceivably in modern history for the first time. Day-to-day life of citizens were interrupted from place to place round the globe since February 2020 (Yunus et al. 2020), and thus, the anthropogenic stress imparted onto the environment was tremendously reduced.

Kerala was one among the most affected states in India during the initial stages of COVID-19 outbreak in the country (Thomas et al. 2020). A state-wide lockdown was implemented by Kerala state on 23 March 2020 which was further followed by a nationwide lockdown from 24 March 2020 onwards enforced by the Government of India till the end of May amongst a population of approximately 1.38 billion (Worldometer 2020). Later, the nationwide lockdown was further extended with varying exemptions in different parts of the country based on the COVID-19 spread rates. The lockdown implemented in Kerala continued till July 2020 with the zonal classification of districts based on COVID-19 transmission rates with several unlock phases and relaxations. On 11 September 2020, the Government of Kerala reported more than 1 lakh confirmed cases (https:// dashboard.kerala.gov.in) in the state. Some of the lockdown-imposed restrictions include ban on roadways and water ways transport services except for the procurement of essential commodities or services. All the government, its autonomous, and subordinate offices including corporations remained closed throughout the lockdown. All commercial and private activities, manufacturing industries, hotels, homestays, lodges and motels, construction activities, etc. were completely shut down during the initial phase of lockdown. Only twenty persons were allowed to attend marriages and death-related ceremonies. Only the utmost essential service sectors were excepted from the lockdown restrictions. However, in the unlock phase/partial unlock phase, activities/sectors like banking, financial and insurance services, ration shops (under Govt PDS) dealing with food, groceries, fruits and vegetables, dairy products, meat and fish, animal fodder, poultry and cattle feed, bakeries, print and electronic media, broadcastings, petrol pumps, LPG, petroleum, power generation, transmission and distribution units and services, private establishments involved in the production and supply of equipment required to contain COVID-19 including masks, sanitizers, drugs, personal protective equipment (PPE), etc. were allowed to function with restrictions. Private vehicles of persons travelling for vaccination against COVID-19 (showing their vaccination registration), movement of persons from other states /countries to destination from railway station and airport (with proof of ticket), interstate road transport for goods, and emergency services were permitted.

Across the world, limited industrial and tourism activities have reduced the movement of people leading to the improvement of our environment and ecosystem. Emissions from the transportation and industrial sectors were remarkably decreased leading to a significant reduction in the levels of environmental pollution (Arora et al. 2020; Muhammad et al. 2020; Paital 2020). Numerous studies were conducted to analyze the ambient air quality in several major cities and industrial areas in different parts of the world and revealed improvement (AQI = 50) in the quality of air (Pal et al. 2022; Ghosh and Ghosh 2020; Sarkar et al., 2020; Thomas et al. 2020). Likewise, many studies have tried to quantify the impacts of lockdown on the water systems of the world as well. Dutta et al. (2020), Khan et al. (2021), Muduli et al. (2021), Chakraborty et al. (2021a), Chakraborty et al. (2021b), Pant et al. (2021), Sharma and Gupta (2022), and many more had investigated the positive effects of lockdown on different river systems in India. Studies have highlighted that notable improvements in river water quality for prominent rivers like Ganga, Yamuna, Damodar, and Gomti were observed during the lockdown period. Dutta et al. (2020), Dhar et al. (2020), Mani (2020), Duttagupta et al. (2021), Muduli et al. (2021), and Shukla et al. (2021) have stated that the water quality of the Ganga river regained drinking suitability designation after several years due to the complete nationwide lockdown, which cleaned the Ganges more than “Ganga Action Plan” and “Namami Gange Mission.” A temporary restriction on industrial and urban activities presented an opportunity and scope for restoration of river water quality and river health from the exploitation pressure of urbanization and industrialization (Chakraborty et al. 2021a).

While numerous studies have linked COVID-19 and water quality status in India, very few studies have reported the water quality allied to COVID-19 conditions in Kerala state except Yunus et al. (2020) and Aswathy et al. (2021) on the reduction of suspended particulate matter in lake Vembanad and Ashtamudi wetland system, respectively. No studies were available with respect to the freshwater quality aspects, which is very much linked with the environmental health of people. Periyar river is the lifeline of Kerala and as a major drinking water source for Kochi corporation and its suburban panchayaths, deserves much attention for COVID-19 linked studies. The river meets the domestic needs of lakhs of people in Ernakulam district. Since the lockdown has imposed several restrictions on the different activities, it is interested to study the activity-oriented aspects especially in the river catchments. Eloor industrial belt area, also known as the Industrial hub of Ernakulam District, is the most polluted zone in the Periyar river. Being an urbanized and industrialized area, numerous hospitals, industrial drainage systems, municipalities, etc. dispose their untreated effluents at this zone into the river (Khalid et al. 2018) and several cases of fish kills/fish decline and significant damage to the paddy fields and other crops in the region (Anjusha et al. 2020) were reported from this area. Therefore, in the present study, an attempt has been made to investigate the influence of continuous lockdowns on river water quality (RWQ) of Periyar river, a tropical Western Ghats (WG) river, hosting a prominent industrial cluster in Kerala state (Eloor-Edayar industrial belt). Since no previous literature is available regarding the effect of lockdown on river water quality in Kerala state, this study made the first attempt to quantify the ambient RWQ during the lockdown (LD) and post-lockdown (PLD) periods of COVID-19 spread. Moreover, this data generated during the lockdown period might serve as baseline data, because during this period the hydrochemical characteristics were least affected by human intervention. Hence, any data generated on river water quality aspects in future will be helpful for assessing the role of industries in determining the water quality and its deterioration.

Study area

The Kerala state is an elongated strip of land located in southwest India bordered by the Arabian Sea on the west and the Western Ghats mountain range on the east. Numerous rivers originate in this water tower system. Periyar is the longest river in Kerala state and serves as the “lifeline of Kerala” providing water for domestic and agricultural activities (Krishnakumar et al. 2022). The river originates from Sivagiri hills in WGs and debouches into the Arabian Sea traveling a distance of 244 km² passing through 3 districts (Ernakulam, Thrissur, and Idukki), 88 villages, 183 sub-watersheds, encompassing highland, midland and lowland terrains covering a watershed area of 5398 sq km. During 2018, the low-lying areas of this river were severely affected due to floods popularly known as the “2018 Kerala floods” due to extreme rainfall (164% > annual average rainfall) (Krishnakumar et al. 2022; Sudheer et al. 2019; Kondapalli et al. 2019; Hunt and Menon 2020) as a consequence of global climate change phenomenon. Sixty thousand hectares of agricultural land have been destroyed in the state, and more than 220,000 people were displaced. The flood has damaged 83,000 km of roadways, including 10,000 km of important highways (Sphere India, 2018; Krishnakumar et al. 2022).

Geologically, a variety of rock types, including sedimentaries from the Tertiary and Quaternary periods and crystalline rocks of Pre-Cambrian age, make up the basin’s geological composition. The crystallines are composed of quartz-feldspar-hypersthene granulites (charnockites), charnockite gneiss, hypersthene-diopside gneiss, hornblende gneiss, hornblende-biotite gneiss, quartz-mica gneiss, and pink granite. These crystalline rocks have undergone significant polymetamorphic and polydeformational processes. Acidic and basic rocks encroach on Pre-Cambrian crystalline surfaces. Laterites cover the Pre-Cambrian crystallines and Tertiary sedimentaries (Krishnakumar et al. 2023).

Forest loam and laterite soil types with site-specific textural and geochemical properties constitutes about 60% of the study area. The study area experiences humid tropical rainforest climate (Singh et al. 2022), and the rainfall is dual monsoon dependent, i.e., Southwest Monsoon (SWM), occurs from June to September and Northeast Monsoon (NEM) from October to December. The average annual rainfall is 3000 mm due to Southwest and Northeast monsoons and the mean temperature ranges from 23 to 27 °C (Saranya et al. 2020).

Eloor-Edayar industrial belt in Periyar basin

Among the different states of India, Kerala ranks 12th position in terms of the total number of industries (http://mospi.nic.in) and more than 60% of the industries are related to food, nonmetallic minerals, tobacco, wood, rubber, and plastic products (Government of Kerala 2020). The lowland catchment area of the Periyar river hosts many prominent industries. Eloor is considered as the largest industrial belt of Kerala with maximum varying clusters of chemical industries (Table 1) including pesticides, petrochemical products, rare earth goods, fertilizers, leather products and inks located on the banks of Periyar river. Toxic chemicals like Endosulphan and DDT were once manufactured at Eloor. More than 300 factories of various chemical industries, bone meal factories and tanneries are currently functional at Eloor on an 11.21 sq km area. Out of these factories, 79 factories are categorized under the red category list of industries producing excessive emissions, effluents, hazardous waste, and over-exploitation of natural resources. According to many reports, industrial sectors at Eloor has very poor- and low-quality mechanisms for effluent treatment. The river serves as a chief source of water for the entire industrial belt and townships and also acts as the principal source for the disposal of wastes and industrial effluents which harms the RWQ considerably. Many times, various print/visual media reported the poor quality of water and related allergic effects to the common people residing in the study area. The data on various communicable diseases reported from this area reiterates these views.

Table 1.

Major and minor industries located in the study area

Industry	Established year	Products
FACT	1944	Fertilizers, Caprolactan and associated by products
Travancore Cochin Chemicals	1950	Caustic soda
Indian Rare Earth Ltd	1951	-
Hindustan Insecticides Ltd	1954	Insecticides, fungicides, herbicides, fertilizers, sanitizers, bio-pesticides, micronutrients, etc
Cochin Minerals and Rutiles	1989	Aqua ferric chloride
SM Industries	2004	Manufacturer of water tank, RCC pipe, cement pipe
Amcos Paints Factory	2003	Paints, Epoxy
Organo Fertilizers India Pvt Ltd	2005	Poultry meal, organic fertilizers, and feed fats
Salfa Chemicals	1995	Sulfur products
Marksmen Marine products Ltd	2015	Fishmeal and fish oil
Parakkal Industries	2017	Aqua feed supplements
Periyar Chemicals Ltd	1968	Ammonium chloride, carbon black, castor oil, caustic soda flakes and soaps, detergent powder and cakes
Active Char Products Pvt Ltd	2005	Coconut shell charcoal based steam activated carbon
Sud-Chemie India Pvt Ltd	1969	Catalyst manufacturing
Cochin Petromins Factory	1993	Industrial chemicals, chemical acids, and general solvents
Indo-German Carbons Ltd	1998	Activated carbon, coconut shell based activated carbon products and solutions
Njavallil latex	1992	Surgical and examination gloves, contraceptive, balloons, elastic threads, erasers, foam, rubber bands
Valeth Hightech Composites Pvt Ltd	-	High silica fabric
Kochiplast solutions Pvt Ltd	2017	-
Rubber O’ Malabar Products Pvt Ltd	2004	Various types of textile reinforced rubber conveyor belts
Glasstech Industries	2004	Glass and aluminum processing machineries
Vtuff Glass Pvt Ltd	2004	Glass
Chilton Factory Kerala	-	Water chillers
Cochin Polymers Pvt Ltd	1995	Water tanks and tank
Cella Space Ltd (Sree Sakthi Paper Mills)	2007	Kraft paper and duplex board
Hindalco IndustriesLtd	1938	Aluminum
RDC Concrete Pvt Ltd (Neptune readymix)	1993	Ready mixed concrete
Southern Gas Ltd	1963	Gas mixtures, nitrogen argon mixture gas, and food grade
Geeyes Industries	-	Orid flour and cattle feed

Methodology

The present study has been confined to about 46 km stretch (total length 244 km) in the lower catchment of Periyar river (from Neeliswaram to Kochi Kayal encompassing Eloor-Edayar industrial belt) as depicted in Fig. 1. The elevation of the study region ranges from 13 to 1 m amsl. This region is highly suitable with all available infrastructural facilities for industrial activities due to landscape morphology, increased population, rapid urbanization, tourism, and travel accessibility (road, rail, waterways, and air connectivity). As a result, many large- and small-scale industries and recreational developments are established near the river banks, which directly impose severe pressure on the riverine system. The tributaries serve as discharging sites for industrial effluents.

Fig. 1.

Location and sampling sites in the study area

Sample collection and analysis

Riverine samples were collected from 15 locations in Periyar river near to the industrial discharge sites in Ernakulam district (Fig. 1) employing stratified random sampling technique. The sampling had been carried out during monsoon and post-monsoon seasons, i.e., during October (NEM) 2020 and January (post-monsoon) 2021, and the possible effect of dryness in the study area is thus eliminated. Initially, samples were taken during October 2020 (partial lockdown) when the lockdown restrictions were partially relaxed (though lockdown was imposed, urgent and important travelling were allowed during this period and hence we could collect samples). Again, the samples were taken during January 2021 (after unlock phase when all activities were beginning to be normal) from the same locations to get a clear idea about how the industrial activities influence the RWQ. Pre-cleaned polyethylene bottles were used for sample collection. All samples were carried properly for further analysis to the laboratory for various physicochemical attributes to assess the RWQ for drinking and irrigational purposes.

The samples were collected for cation and anion analysis in pre-cleaned HDPE bottles that have been rinsed with sample water three times prior to use following the methods of APHA (2012) and the sampling locations (co-ordinates) were determined using Global Positioning System (GPS). Parameters such as temperature, pH, Electrical Conductivity (EC), Dissolved Oxygen (DO), and Total Dissolved Solids (TDS) were determined in situ by Aquaread multiparameter water quality analyzer (AP-2000-D). Samples were filtered using 0.45 µm membrane filters to separate suspended particles prior to hydrochemical analysis. The concentration of major cations (Ca²⁺, Mg²⁺, Na⁺, and K⁺) were determined using Microwave Plasma Atomic Emission Spectroscopy (MP-AES, Model: Agilent 4210) while that of major anions like NO₃⁻, SO₄²⁻, and Cl⁻ were estimated by Continuous Flow Analyzer (CFA, Model: Skalar SAN + +). Bicarbonate ions (HCO₃⁻) were determined by acid titration. Trace metal ion determination was done with Microwave Plasma Atomic Emission Spectroscopy (MP-AES, Model: Agilent 4210). The charge balance between the total dissolved cations and total dissolved anions were estimated as Eq. 1 (Freeze and Cherry 1979) to evaluate the accuracy of the results. Normalized inorganic charge balance (NICB) of most of the samples was found to be ≤ 10%, however, a few samples have NICBs slightly ≥ 10%.

$$NICB\left(\%\right)=100\times \frac{(\sum {\mathit{TZ}}^{+}-\sum {\mathit{TZ}}^{-})}{(\sum {\mathit{TZ}}^{+}+\sum {\mathit{TZ}}^{-})}$$ 1

where TZ⁺ and TZ⁻ are the sum of total cations and anions, respectively.

Water quality index

The Bureau of Indian Standard Water Quality Index (BIS WQI) is the Indian national WQI standard parameter which is defined by the Government of India under IS: 10,500. However, to identify the overall status of all the parameters, the widely accepted water quality index (WQI) technique was employed. The 5 important steps applied to determine WQI are:

• Step 1: Collection of data regarding the relevant physicochemical water quality parameters.

• Step 2: Calculation of proportionality constant “K” value using the formula:

  $$K=\left[\frac{1}{1/\sum nisi}\right]$$ 2

where “si” is standard permissible for the nth parameter.

• Step 3: Calculation of quality rating for the nth parameter (Qn) where there are n parameters.where Vn = estimated value of the nth parameter of the given sampling station.

  $$Qn=100\times \left\{\frac{Vn-Vi}{Sn-Vi}\right\}$$ 3

Vi = ideal value of the nth parameter in pure water. And Sn = standard permissible value of the nth parameter.

• Step 4: Calculation of unit weight for the nth parameter.

  $$Wn=\left(\frac{K}{\mathit{Sn}}\right)$$ 4
• Step 5: Calculate the water quality index (WQI) using the formula,

  $$WQI=\frac{\sum Wn\times Qn}{\sum Wn}$$ 5

The characterization of WQI status was assessed based on the WQI values followed by Brown et al. (1970).

Irrigational suitability

Indicators of irrigation quality were determined using sodium absorption ratio (SAR), sodium percentage (% Na), magnesium hazard ratio (MH), and Kelly’s ratio (KR) using the mathematical relationships as follows:

$$SAR=\frac{{\mathit{Na}}^{+}}{\sqrt{\frac{({\mathit{Ca}}^{2+}+{\mathit{Mg}}^{2+})}{2}}}$$ 6

$$\%Na=\frac{({\mathit{Na}}^{+}+K^{+})}{({\mathit{Ca}}^{2+}+{\mathit{Mg}}^{2+}+{\mathit{Na}}^{+}+K^{+})}\times 100$$ 7

$$MH=\frac{{\mathit{Mg}}^{2+}}{{\mathit{Ca}}^{2+}+{\mathit{Mg}}^{2+}}\times 100$$ 8

$$KR=\frac{{\mathit{Na}}^{+}}{{\mathit{Ca}}^{2+}+{\mathit{Mg}}^{2+}}$$ 9

Multivariate statistical analysis

Multivariate statistical methods were employed to study relationships existing among samples and various RWQ variables. Spearman rank correlation analysis and principal component analysis (PCA) were performed to classify the samples and to identify the relationship existing between the chemical parameters and the sample. Prior to multivariate analysis, the raw hydrochemical data were transformed using centered log-ratio (clr) transformation, which destroys the closure effect revealing the inherent patterns in the data structure (Reimann et al. 2012; Buccianti and Grunsky 2014; Islam et al. 2020). The clr transformation was carried out using CoDaPack, a freely available Excel-based software. In PCA, principal components were extracted in a way that the greatest variance of the observed data is accounted by the first component, the second greatest variance by the second component, and so on (Selle et al. 2013). Principal components with an eigen value > 1 are considered to account for the maximum variance in the observed parameter (Kolsi et al. 2013). Varimax rotations were applied to all the extracted principal components to reduce the contribution of the variables that are not significant (Closs and Nichol 1975).

Since hydrochemical data are nonparametric, ordinal and variables are less likely to be normally distributed, a Spearman rank correlation analysis was performed. Moreover, it is less sensitive to outliers and will yield an exact line of best fit and correlation coefficient among variables. It is a type of bivariate correlation which investigates the relationship between two sets of data based on their data ranks.

Health risk assessment

Ingestion is usually considered as the common exposure pathway of pollutant entry into human beings followed by dermal exposure. Therefore, in this study, the exposure pathway of drinking water intake was considered to assess and evaluate the degree of risks caused by different metal ion contaminants on human health. The exposure dose and risk assessments were evaluated based on the recommended model by the USEPA (Mohammadi et al. 2019). The lifetime average daily intake dose (mg/kg/day) via ingestion can be computed from Eq. 10 (Bhattacharya et al. 2020; Joardar et al. 2021; Naik et al. 2021a; Zhang et al. 2018).

$$Intake\left(O,r,a,l\right)=\frac{C\times IR\times EF\times ED}{BW\times AT}$$ 10

where C is the concentration of contaminant (mg/L), IR is the rate of intake (1 L/day for children and 2.5 L/day for adults) (Das et al. 2020). EF denotes the exposure frequency (365 days) (Naik et al. 2021a; Su et al. 2018). ED represents exposure duration (65 years for adults and 6 years for children) (Das et al. 2020; Naik et al. 2021b). BW is the average body weight and is taken as 70 kg for adults and 15 kg for children (Bhattacharya et al. 2020; Njuguna et al. 2020). AT denotes average lifetime in days and is calculated as the duration of total exposure (EF × ED). Since all the analyzed elements in the study are observed in slight amounts, the noncarcinogenic hazard quotient through oral exposure was computed by Eq. 11 for As, Zn, Pb and Cd.

$$HQj=\frac{Intake(Oral)}{RfD(Oral)}$$ 11

where j represents the element and RfD is the noncarcinogenic reference dose limit through oral exposure (US EPA, 2019). The site specific cumulative noncarcinogenic risk posed by all the heavy metals is the Hazard Index (HI)(unitless) which can be determined using Eq. 12 (Bhattacharya et al. 2020).

$$HI={\sum }_{j=1}^{n}HQj$$ 12

When HQ or HI > 1, adverse effect on human health may occur or the health risk is unacceptable while when HQ or HI < 1 indicates no adverse effects on human health or is acceptable (US EPA 2004; Naik et al. 2022). Alver (2019) and Rahim et al. (2019) classified HI as HI < 0.1, negligible noncarcinogenic risk; 0.1 ≤ HI < 1, low noncarcinogenic risk; 1 ≤ HI < 4, moderate noncarcinogenic risk; 4 ≤ HI, high noncarcinogenic risk (Alver 2019; Barzegar et al. 2019).

Carcinogenic risk caused by carcinogenic chemical pollutants were calculated by

$$\mathrm{CR}=\mathrm{Intake}\left(\mathrm{oral},\right)\times \mathrm{SF}$$ 13

where CR is the carcinogenic risk, SF is the oral cancer slope factor (mg/kg/day). When CR < 10⁻⁶ indicating carcinogenic risk could be negligible, CR > 10⁻⁴ high risk of causing cancer on humans, 10⁻⁶ < CR < 10⁻⁴ acceptable risk to human beings (US EPA 2004).

Results and discussion

Table 2 shows the descriptive statistics of the widely measured physicochemical parameters of RWQ during the sampling periods. From the analytical results, sampling stations showed a notable temporal variability in October 2020 compared to January 2021. All the chemical parameters were found to be reduced during the lockdown (LD) period compared to the post-lockdown (PLD) period. During LD, RWQ exhibited a chemical concentration trend of HCO₃²⁻ > TH > Na⁺ > Cl⁻ > SO₄²⁻ > Ca²⁺ > Mg²⁺ > K⁺ > NO₃⁻, while in PLD, an order of Cl⁻ > TH > Na⁺ > HCO₃²⁻ > SO₄²⁻ > Mg²⁺ > Ca²⁺ > K⁺ > NO₃⁻ was observed. The change in the order of chemical composition of RWQ can be directly associated with the effect of anthropogenic activities practiced in the study area. Since during the post-lockdown period, the partial reopening of industries started in the Eloor-Edayar region which transformed the LD scenario of reduced pollutants from the transportation and industrial sectors.

Table 2.

Descriptive statistics of different water quality parameters during LD and PLD periods

Period	Parameters	pH	EC (µS/cm)	TDS (mg/L)	DO (mg/L)	BOD (mg/L)	Ca2+ (mg/L)	Mg2+ (mg/L)	Na+ (mg/L)	K+ (mg/L)	TH (mg/L)	SO42− (mg/L)	NO3− (mg/L)	Cl− (mg/L)	HCO32− (mg/l)
LD	Mean	6.48	324.87	211.16	6.72	0.61	6.02	4.01	20.76	1.93	31.52	7.19	0.46	19.72	37.33
	Standard deviation	0.46	787.19	511.68	1.00	0.22	7.17	11.12	66.74	4.53	62.74	21.39	0.11	11.65	27.65
	Kurtosis	1.72	14.78	14.78	− 0.64	2.95	9.69	14.96	14.94	14.90	14.39	14.81	− 0.07	6.07	11.45
	Skewness	0.02	3.84	3.84	− 0.52	1.55	2.97	3.87	3.86	3.86	3.77	3.84	0.35	2.44	3.20
	Minimum	5.46	80.00	52.00	4.80	0.40	1.66	0.88	1.40	0.34	7.97	0.10	0.26	12.94	10.00
	Maximum	7.50	3162.00	2055.30	7.90	1.20	29.85	44.20	261.80	18.30	256.42	84.30	0.66	55.00	132.00
PLD	Mean	7.69	2040.43	1326.28	8.72	3.82	34.43	37.28	230.67	13.65	239.39	74.27	0.66	409.62	85.37
	Standard deviation	0.72	3695.13	2401.83	1.53	2.33	64.61	74.35	461.32	24.69	465.06	151.63	1.59	807.31	13.09
	Kurtosis	0.38	8.15	8.15	5.27	2.65	8.01	9.16	8.93	8.26	8.97	9.33	14.91	8.91	0.89
	Skewness	− 0.01	2.72	2.72	2.35	1.68	2.71	2.88	2.83	2.73	2.84	2.90	3.86	2.83	− 0.87
	Minimum	6.23	178.46	116.00	7.61	1.79	2.45	0.78	2.16	0.96	10.15	0.12	0.11	7.75	54.16
	Maximum	9.10	13,929.23	9054.00	13.20	10.03	241.56	281.68	1739.99	93.28	1762.29	574.48	6.40	3050.00	103.68
Relative change in water quality w.r.t LD during PLD		1.2 times	6 times	6 times	1.3 times	6 times	5.6 times	9 times	11.5 times	7 times	7.7 times	10 times	1.4 times	21 times	2 times
WHO (2017)		6.5–8.5	500	500	5	-	100	50	200	20	200	250	-	250	200
BIS (2012)		6.5–8.5	500	500	5	-	75	30–100	-	-	200	200–400	-	250–1000	600

Hydrochemical properties of river water

In the LD period, the pH values ranged from 5.46 to 7.50 with an average of 6.48, and during PLD, the values varied from 6.23 to 9.10 with an average of 7.69. The slightly lower value of pH observed in the LD is due to the less or no mixing of industrial effluents from the nearby regions. The actual acidic pH of precipitation was reflected during LD while the slightly alkaline pH of PLD is directly linked to the mixing of industrial effluents containing high alkaline solutions. The EC values of the study varied between 80 and 3162 µS/cm and 178–13,929 µS/cm during LD and PLD periods, respectively, clearly indicating the mixing of various soluble industrial effluents. Similar trends were reported earlier by Chakraborty et al. (2021a) and Chakraborty et al. (2021b) in river Damodar, and Kutralam-Muniasamy et al. (2022) in Santiago river during the unlock phase of COVID-19 lockdown. TDS values also showed an increased trend correlating with EC values during the PLD phase. The concentration of TDS was found to be significantly lower during LD as compared to PLD which might be due to less discharge of pollutants in LD. According to BIS, the desirable limit of TDS in drinking water is 500 mg/L. In the study, the mean TDS value during PLD has exceeded this limit reflecting the lower quality for drinking.

DO and BOD are the two essential biological RWQ parameters. The DO values represent the availability of oxygen, and BOD indicates the utilization of oxygen in water. Numerous chemical reactions, decomposition of organic matter, and microbial respiration can have an effect on these variables. DO in the LD phase ranged from 4.80 to 7.90 (mg/L), whereas in the PLD phase, DO values were observed between 7.61 and 13.20 mg/L. BOD values varied between 0.40–1.20 in LD and 1.79–10.03 in PLD.

The observed values of BOD in LD period with that of PLD period shows that there is a slight decreasing trend due to the comparative cleaner status of the locations by the lesser growth and survival of microbes. On the other hand, during PLD, the industrial discharges cause for the proliferation of microbes and a favorable environment for algal growth. The comparative higher values of DO is observed during the PLD period with that of LD may be due to the active phase of the riverine locations due to turbulence and mixing activities from the discharges as well.

The mean concentrations of Ca²⁺, Mg²⁺, Na⁺, and K⁺ in the study area were 6.02 (mg/L), 4.01 (mg/L), 20.76 (mg/L), and 1.93 (mg/L) during LD and 34.43 ppm, 37.28 ppm, 230.67 ppm, and 13.65 ppm during PLD periods. It is clear from the results that river water exhibited reduced concentration of metal ions due to the low input of chemical elements through industries. The mean concentrations of SO₄²⁻, NO₃⁻, Cl⁻, and HCO₃²⁻ observed were 7.19 (mg/L), 0.46 mg/L, 19.72 (mg/L), and 37.33 (mg/L) in LD and 74.27 (mg/L), 0.66 (mg/L), 409.62 (mg/L), and 85.37 (mg/L) in PLD period. It is evidently noticed that the analyzed cations as well as anions exhibited a reduced concentration during LD period implying that the closing of industries in the present study stretch of Periyar river has highly reduced the concentration of RWQ parameters. An increase in Na⁺ and Cl⁻ concentration was observed in PLD which might be due to the direct discharge of effluents as well as the wastewater released from sodium hypochlorite (NaOCl) production industries. During lockdown period, detergents and disinfectants were produced by the industries and used in large scale as a weapon against coronavirus (Haque et al. 2023) in hospitals, houses, office establishments, etc. Moreover, Cl⁻ can also be attributed to increased sewage discharges during lockdown period.

In the study area, the concentration range of TH fluctuated between 7.97–256.42 mg/L during LD and 10.15–1762.29 (mg/L) during PLD period. Role of anthropogenic signatures including sewage, chemical fertilizers, and domestic wastes in imparting the higher values during PLD period is very evident from the observed higher values.

Metal concentration

The results of heavy/trace metal ion concentrations in study area shown in Table 3 revealed a significant variation among the sites. Metal ions like Cu, Ni, Co, Mn, and Cr were found below the detection limits in almost all the sampling sites whereas metals like Zn, Cd, As, Th, and Pb exhibited varying concentrations in each site. Cd concentration varied from BDL-0.55 mg/L which is higher than the BIS standard for drinking water (0.003 mg/L). As and Pb values observed during PLD were between BDL-0.46 mg/L and BDL-2.85 mg/L respectively against the standard values of 0.01 mg/L. The reason may be the low levels of suspended particles during the lockdown period that would aid in the retention of As in sediments from the above water column through binding or co-precipitation (Smedley and Kinniburgh 2002; Rubinos et al. 2003; Barral-Fraga et al. 2020). Moreover, the increase in As can be attributed to the rise of usage of disinfectants during LD period which was reflected in PLD season. It was noticed from the results that trace metals were observed during the PLD phase whereas during LD phase, the metal ions exhibited BDL status which can be attributed by reduced industrial operations. When the industrial operations become full-fledged, the quantity of discharge also increases and cause for the enhanced degradation. The long-term addition of these discharges into the system gradually causes health issues, since the raw water is not purified in its desired manner. Severe threats to the physiological functions of the body processes like cardiovascular disease, neurological effects, diabetics, respiratory disorders, etc. may happen due to the usage of such type of water as observed elsewhere.

Table 3.

Metal concentrations during LD and PLD periods (mg/L)

Sample stations	Zn		Cd		Cu		Ni		As		Co		Th		Pb		Mn		Cr
	LD	PLD	LD	PLD	LD	PLD	LD	PLD	LD	PLD	LD	PLD	LD	PLD	LD	PLD	LD	PLD	LD	PLD
S1	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	0.07	BDL	BDL	BDL	BDL	BDL	BDL	BDL
S2	BDL	BDL	BDL	0.01	BDL	BDL	BDL	BDL	BDL	0.01	BDL	BDL	0.06	0.29	BDL	0.07	BDL	0.01	BDL	BDL
S3	BDL	BDL	BDL	0.01	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	0.06	BDL	BDL	BDL	BDL	BDL	BDL	BDL
S4	BDL	BDL	BDL	0.03	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	0.02	BDL	BDL	BDL	BDL
S5	0.01	BDL	BDL	0.02	BDL	BDL	BDL	BDL	BDL	0.02	BDL	BDL	BDL	0.45	BDL	0.1	BDL	BDL	BDL	BDL
S6	BDL	0.01	BDL	0.02	BDL	BDL	BDL	BDL	BDL	0.01	BDL	BDL	BDL	0.34	BDL	0.09	BDL	0.01	BDL	BDL
S7	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	0.04	BDL	BDL	BDL	BDL	BDL	BDL	BDL
S8	BDL	0.01	BDL	0.02	BDL	BDL	BDL	BDL	BDL	0.01	BDL	BDL	0.03	0.29	BDL	0.07	BDL	0.02	BDL	BDL
S9	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	0.04	BDL	BDL	BDL	BDL	BDL	BDL	BDL
S10	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL	0.05	BDL	BDL	BDL	BDL	BDL	BDL	BDL
S11	BDL	0.22	BDL	0.32	BDL	BDL	BDL	BDL	BDL	0.31	BDL	BDL	0.01	6.56	BDL	1.48	BDL	0.09	BDL	BDL
S12	BDL	0.32	BDL	0.11	BDL	BDL	BDL	BDL	BDL	0.08	BDL	BDL	0.06	2.62	BDL	0.64	BDL	BDL	BDL	BDL
S13	BDL	0.08	BDL	0.1	BDL	BDL	BDL	BDL	BDL	0.08	0.01	BDL	0.08	2.38	BDL	0.69	BDL	BDL	BDL	BDL
S14	BDL	BDL	BDL	0.43	BDL	BDL	0.01	BDL	BDL	0.32	BDL	BDL	0.06	7.27	BDL	2.14	BDL	BDL	BDL	BDL
S15	BDL	0.02	BDL	0.55	BDL	BDL	BDL	BDL	BDL	0.46	BDL	BDL	BDL	11.2	BDL	2.85	BDL	BDL	0.01	BDL

Impact on overall water quality

WQI evaluates a collective result of analyzed parameters on the overall water quality for human consumption, domestic, agriculture usages, and aquatic life. Initially, Horton (1965) developed WQI, was modified by Brown et al. (1970) (Table 4), and the subsequent developments of WQIs is described by Tyagi et al. (2013). Later, several water quality indices have been formulated by national and international organizations all over the world. The assessment of RWQ based on drinking purposes was evaluated using weighted arithmetic water quality index method. The results of WQI ranged from 5.71 in excellent category to 55 in poor category during LD period with very significant positive changes in RWQ, while in PLD period, the WQI varied from 19.56 in excellent category to 175 indicating unsuitable for drinking category (Table 5).

Table 4.

Status of water quality based various index levels (after Brown et al. 1970)

Water quality index level	Water quality status	Possible usage
0–25	Excellent	Drinking, irrigation, industrial
26–50	Good	Drinking, irrigation, industrial
51–75	Poor	Irrigation and industrial
76–100	Very poor	Irrigation
> 100	Unsuitable for drinking	Treatment required before use

Table 5.

Water quality index (WQI) during LD and PLD periods of the sampling locations

Sampling sites	LD		PLD
S1	35.83	Good	56.94	Poor
S2	30.75	Good	49.96	Good
S3	25.98	Excellent	19.57	Excellent
S4	16.79	Excellent	68.21	Poor
S5	25.36	Excellent	52.51	Poor
S6	30.60	Good	28.73	Good
S7	42.31	Good	60.28	Poor
S8	5.71	Excellent	44.56	Good
S9	36.92	Good	34.61	Good
S10	30.69	Good	45.54	Good
S11	21.28	Excellent	61.83	Poor
S12	21.23	Excellent	53.48	Poor
S13	26.26	Good	50.74	Good
S14	37.08	Good	83.91	Very poor
S15	74.00	Poor	175.67	Unsuitable

Lockdown has increased the domestic water demand as frequent cleaning of hands has become a habit and has decreased the nondomestic demand (Balamurugan et al. 2021). The comparison of WQI values between the two sampling periods shows that during LD, 40% samples were seen in the excellent category, 53% samples in good, and 7% samples in poor category whereas during PLD period, only 7% samples were observed in the excellent category, 40% samples in good, 40% samples in poor, 6% in very poor, and 7% in unsuitable for drinking category. Inverse distance weighted algorithm (IDW) using Arc.GIS 10.3 software was applied to classify the spatial distribution of water quality in the river stretch. The results of spatio-temporal variations of WQI in Fig. 2 clearly indicates a shift to poor category in PLD compared to LD period for certain locations (sample sites 1, 4, 5, 7, 11, 12, and 14) indicating the deterioration of RWQ in unlock phase due to mixing of industrial effluents. The overall RWQ of the present study showed a declining pattern from LD to PLD with differing magnitudes.

Fig. 2.

Spatial variation of river water quality index during the two sampling periods (the lower catchment area of 46 km.² of Periyar river was digitized in ArcGIS 10.3 software)

Impact on irrigation

Percent Na (%Na)

EC is an important parameter for determining the suitability of water for irrigation. It was observed that 88% of samples were good for irrigation in LD period whereas the suitability reduced to 66% during PLD phase indicating the deterioration of study area due to industrial and sewage effluent discharge. Wilcox diagram was used to assess the irrigational suitability plotting sodium percentage (%Na) versus EC (Wilcox,1955). The diagram (Fig. 3) shows that 93% of the samples were excellent to good for irrigation during LD phase with 7% of samples indicating unsuitable for irrigation. Meanwhile, PLD period samples exhibited 66% of samples as excellent to good for irrigation, 20% of samples were unsuitable, and 14% of samples were doubtful to unsuitable for irrigation.

Fig. 3.

Wilcox classification of samples for LD and PLD periods

SAR

Sodium adsorption ratio (SAR) is a commonly used criterion for estimating the sodium hazard associated with an irrigation water supply. For US salinity diagram, proposed by USSL (1954), SAR was plotted against EC in Aquachem software 2014.2 (Fig. 4). It was observed from the results that during LD, 7% of samples were in the C2S1 zone while in PLD, 14% samples occurred in C2S1 indicating good water for irrigation. Eighty-seven percent of the samples occurred in the C1S1 zone (good water for irrigation) in LD whereas the number reduced to 66% during PLD suggesting the mixing of industrial effluents with river water. The situation may become problematic after the regular functioning of industrial establishments in a full manner. At present, the establishments are slowly picking up its pace. The C1S1 and C2S1 samples can be used for irrigation in almost all types of soils with little danger of exchangeable sodium (Marghade et al. 2012). Only 7% of samples were seen in the C4S2 zone during LD; however, the sample percentage has increased to 20% (C4S3 and C4S2 zones) in PLD clearly indicating the effect of unlock phase influence on RWQ.

Fig. 4.

USSL classification of LD and PLD samples

Magnesium hazard (Mg Haz)

It is a well-established fact that the higher levels of Mg²⁺ in water support an advanced production of transferrable Na⁺ in irrigated soils which not only deteriorates soil structure but also has an adverse effect on the crop yield-affecting overall production. Water with Mg Haz less than 50% is appraised for irrigation. The Mg Haz values in LD period exhibited 93% of samples in the suitable category and 7% of sample in the unsuitable category, while during PLD, 66% of samples were seen in the suitable category and remaining 44% of samples in the unsuitable category (Fig. 5).

Fig. 5.

Spatiotemporal variation of irrigation suitability indices values at each sampling station

KR

Kelly’s ratio (KR)/ Kelly's Index (KI) is an important parameter used in the assessment of water quality for irrigation based on the Na⁺, Ca²⁺, and Mg²⁺ ion concentrations in the water. As per standard classification, water with a KR value higher than unity is considered nonsuitable for irrigation. In the present study, 93% of the samples were observed in the safe category while 7% of the samples were in the unsafe category during LD period. While during PLD, 60% of samples were in the safe category and 40% of samples in the unsafe category for irrigation. Higher values of KR can be ascribed to the presence of excessive Na⁺ ions in the river water due to the addition of wastewaters from several industries. This may be credited to the direct disposal of domestic and industrial wastewaters into the river leading to increased concentrations of the different ions in the water. Higher KR values for the surface water have also been reported in the literature suggesting the unsafe nature of water for irrigation.

PI

Permeability index has been developed as a standard for examining the aptness of water for irrigation. According to the standard classification, water belonging to class I and II with 75% or more of permeability is rendered good for irrigation whereas class III water with 25% of maximum permeability is contemplated as unfit. The PI classification revealed that 87% of samples in class I and 7% of samples each in class II and III during LD while for PLD period 72% of samples occupied class I and 28% of samples occupied class III categories.

Correlation analysis

Correlation between various RWQ variables were depicted through Spearman rank correlation coefficient analysis. It is a statistical tool that measures the relationship between various variables and provides information on magnitude of correlation and the direction of its relationship. EC and TDS exhibited a strong positive correlation with Ca²⁺, Mg²⁺, Na⁺, K⁺, TH, SO₄²⁻, and Cl⁻ during both periods (Tables 6 and 7). TDS was strongly positively correlated with EC at 0.001 level of significance because electric conductivity is directly dependent on the presence of TDS in water. DO exhibited no correlation with any variable signifying its independent characteristic. NO₃⁻ depicted a negative correlation between all the variables. Strong positive correlations were exhibited by Ca²⁺ with Mg²⁺, Na⁺, K⁺ TH, SO₄²⁻, and Cl⁻. pH showed correlations with all analyzed parameters except DO, NO₃⁻, and HCO₃⁻ during LD period whereas no correlation was observed during PLD period. A strong relation (p < 0.01) is observed within the alkaline earth elements (Ca²⁺ and Mg²⁺) and alkalis (Na⁺ and K⁺) in all the seasons.

Table 6.

Spearman’s correlation matrix of water quality parameters during LD period

	pH	EC	TDS	DO	Ca2+	Mg2+	Na+	K+	TH	SO42−	NO3−	Cl-	HCO3−
pH	1.00
EC	0.63	1.00
TDS	0.63	1.00	1.00
DO	− 0.24	0.01	0.01	1.00
Ca	0.62	0.94	0.94	0.13	1.00
Mg	0.62	1.00	1.00	0.00	0.93	1.00
Na	0.62	1.00	1.00	0.01	0.93	1.00	1.00
K	0.63	1.00	1.00	− 0.01	0.93	1.00	1.00	1.00
TH	0.63	1.00	1.00	0.03	0.96	0.99	1.00	0.99	1.00
SO4	0.62	1.00	1.00	0.01	0.94	1.00	1.00	1.00	1.00	1.00
NO3	− 0.39	− 0.53	− 0.53	− 0.45	− 0.49	− 0.52	− 0.52	− 0.51	− 0.52	− 0.52	1.00
Cl	0.62	0.87	0.87	0.12	0.94	0.85	0.86	0.85	0.89	0.87	− 0.48	1.00
HCO3	0.38	0.14	0.14	0.17	0.24	0.09	0.10	0.10	0.14	0.11	− 0.23	0.48	1.00

Table 7.

Spearman’s correlation matrix of water quality parameters during PLD period

	pH	EC	TDS	DO	Ca Ca2+	Mg2+	Na+	K+	TH	SO42−	NO3−	Cl-	HCO3−
pH	1.00
EC	− 0.09	1.00
TDS	− 0.09	1.00	1.00
DO	− 0.03	0.31	0.31	1.00
Ca	− 0.08	0.99	0.99	0.39	1.00
Mg	− 0.08	0.99	0.99	0.25	0.98	1.00
Na	− 0.09	0.99	0.99	0.27	0.98	1.00	1.00
K	− 0.10	0.99	0.99	0.23	0.97	1.00	1.00	1.00
TH	− 0.08	1.00	1.00	0.30	0.99	1.00	1.00	0.99	1.00
SO4	− 0.08	0.99	0.99	0.29	0.99	1.00	1.00	0.99	1.00	1.00
NO3	− 0.02	− 0.17	− 0.17	− 0.19	− 0.17	− 0.17	− 0.17	− 0.18	− 0.17	− 0.17	1.00
Cl	− 0.09	0.99	0.99	0.27	0.98	1.00	1.00	1.00	1.00	1.00	− 0.17	1.00
HCO3	0.17	0.22	0.22	− 0.33	0.20	0.20	0.18	0.18	0.20	0.18	0.19	0.18	1.00

Multivariate statistical analysis

The results of PCA for total element concentrations during LD and PLD are shown in Table 8. Three principal components (PCs) with eigenvalues higher than 1 (before and after rotation) were extracted. The results indicate that PCA leads to reduction of the initial dimension of the dataset to three components which explains 88.59% and 96.03% of the total variance of the data variation for LD and PLD periods, respectively. For LD, PC1 explains 46.74% of the total variance and loads heavily on pH (0.82), EC (0.93), TDS (0.93), DO (0.76), BOD (0.87), NO₃⁻ (0.83), and Cl⁻ (0.75). The second component was dominated by Mg²⁺ (0.75), K⁺ (0.83), and HCO₃⁻ (0.80) accounting for 24.16% of the total variance. The third component was very strongly correlated with Ca²⁺(0.92) and TH (0.81) and accounts for 18.67% of the total variance. The absence of sulphate loading in the principal component indicates the presence of nonpoint source pollution, such as decomposed organic waste from domestic activities. The lack of proper waste disposal facilities and the continuous lockdowns have forced the people to dump their household wastes in the nearby rivers and lakes. Whereas during PLD period, PC1 accounts for 55.63% of total variance and exhibited a strong positive correlation with pH (0.80), DO (0.82), BOD (0.83), Mg ²⁺ (0.84), Na⁺ (0.86), K⁺(0.79), TH (0.76), SO₄²⁻ (0.87), NO₃⁻ (0.72), Cl⁻ (0.85), and HCO₃⁻ (0.83). The second component explains 25.8% of total variance dominated by EC (0.86) and TDS (0.86). PC3 explains 14.59% of total variance with a strong affinity towards Ca²⁺ (0.76). In both periods, variables like pH, DO, BOD, NO₃⁻, and Cl⁻ are associated with PC1 explaining their common source of origin. EC and TDS had shifted from PC1 in LD to PC2 during PLD while Mg²⁺, Na⁺, and HCO₃⁻ linked with PC2 in LD changed to PC1 in PLD. Ca²⁺ was seen loaded with PC3 in both the seasons. It is also revealed from the results that the associations observed in correlation matrix is strongly exhibited through the component loadings (Fig. 6). Normal behavior of a freshwater system is reflected from the PC1 loadings in LD period, whereas during PLD, this behavior is influenced by the industrial effluents. PC1 loadings in PLD reflects the role of some external factors in bringing the changes in RWQ. Variables such as DO, BOD, Ca²⁺, NO₃⁻ and Cl⁻ remained same without any component change indicating their natural occurrence in the basin and unaltered by external forces.

Table 8.

Rotated component matrix of water quality variables

Parameters	LD			PLD
	PC1	PC2	PC3	PC1	PC2	PC3
pH	− 0.82	− 0.38	− 0.40	− 0.80	− 0.43	− 0.41
EC	0.93	0.27	− 0.14	0.40	0.86	0.31
TDS	0.93	0.27	− 0.14	0.40	0.86	0.31
DO	− 0.76	− 0.48	− 0.31	− 0.82	− 0.44	− 0.34
BOD	− 0.87	− 0.32	− 0.30	− 0.83	− 0.46	− 0.12
Ca2+	− 0.12	− 0.20	0.92	0.40	0.49	0.76
Mg2+	0.59	0.75	− 0.05	0.84	0.42	0.31
Na+	0.68	0.39	0.52	0.86	0.38	0.32
K+	0.31	0.83	− 0.11	0.79	0.51	0.09
TH	0.38	0.36	0.81	0.76	0.48	0.44
SO42-	0.53	0.16	0.40	0.87	0.20	0.34
NO3-	− 0.83	− 0.40	− 0.32	− 0.72	− 0.45	− 0.42
Cl-	0.75	0.60	0.12	0.85	0.39	0.33
HCO3-	− 0.25	− 0.80	− 0.41	− 0.83	− 0.34	x− 0.42
Percentage of variance	45.74	24.16	18.67	55.63	25.8	14.59
Cumulative percentage	45.74	69.90	88.58	55.63	81.44	96.03

-ve indicates inverse relation between factor and variable. PCA > 0.5 are shown in bold

Extraction method: principal component analysis; rotation method: varimax with Kaiser normalization

Fig. 6.

Plot of loading of three principal components in PCA results

Changes in factors influencing river water quality

River water quality is controlled by several natural and anthropogenic factors, including the climatic regime (temperature and rainfall), river regime, changes in land use and land cover (LULC), and land management practices in a river basin (Sarkar et al. 2021a, b). The lowland regions of PRB (including the present study area) experienced severe flood during 2018 and 2019 resulting in the historical Kerala Flood of 2018 after consecutive drought years of 2016 and 2017. In the present study, we observed the rainfall pattern, evapotranspiration, and basin LULC during the lockdown phases (March to October) and post-lockdown phases (January to May) to depict the influence of the various natural and anthropogenic forces on the river water quality. The study found that there were no significant differences between the LD and PLD phases. According to Mathew et al. (2021) the monsoon rainfall over the Kerala region has weakened during the past two decades. The recorded actual annual mean NEM rainfall in 2020 was 433 mm compared to the normal 519 mm rainfall resulting in 17% deficiency of rainfall. A reduction in the number of wet spell days in Kerala region is associated with an increase in number of prolonged dry days having a significant impact on the hydrological systems, in terms of increasing frequency of floods. As per the existing literature, the month of October showed a declining trend of rainfall pattern in lowlands, midlands, and highlands. Marked increase in evapotranspiration was observed only in the highlands and midlands of the Periyar basin (Mathew et al. 2022). The results of monthly actual evapotranspiration (AET) data (https://bhuvan.nrsc.gov.in/) revealed no significant role of evapotranspiration in the study area during the sampling period and hence not influential in water quality. The LULC during this short period did not change much, mainly due to the COVID pandemic. Thus, the marked fall in the concentration of the pollutants, especially the metal ions in the lock-down phase, is undoubtedly due to the inactive phase during COVID-19 pandemic. Since no anomalous precipitation pattern was evidenced in the study region during the study period, the results obtained in the study can be directly linked to the role of industries in bringing the changes in water quality during PLD compared to LD phase which was reflected during LD and after lockdown.

Human health risk assessment

Long-term exposure to trace elements may induce toxic deleterious effects to human body even in small amounts. In the study, trace elements like Zn, Cd, As, and Pb are found in quantities slightly above the BIS permissible limits during PLD period. Neslund-Dudas et al. (2018), Qing et al. (2020), and Rahimzadeh et al. (2017) stated that low-dose and long-term Cd exposure is associated with nephrotoxicity, osteoporosis, and neurotoxicity and plays a major role in prostate cancer. Zn exposure leads to intoxication, neuronal deficits, respiratory disorder, metal fume fever, and altered lymphocyte function (Plum et al. 2010). As absorption in the body causes cancer and skin lesions, cardiovascular disease, neurological effects, and diabetics (Hughes et al. 2011) while long-term Pb exposure results in cardiovascular diseases, kidney damage, high blood pressure, joint and muscle pain, difficulty in memory, reproductive health issues, etc. (Kim et al. 2014). Table 9 shows the daily intake concentration of toxic metal ions in the residents of the study area.

Table 9.

Daily intake of different metal ions in adults and children through oral exposure pathway

Sampling stations	Intake—adults				Intake—children
	Zn	Cd	As	Pb	Zn	Cd	As	Pb
S1	BDL	BDL	BDL	BDL
S2	BDL	0.0004	0.0004	0.003	BDL	BDL	BDL	BDL
S3	BDL	0.0004	BDL	BDL	BDL	0.001	0.0007	0.005
S4	BDL	0.0011	BDL	0.001	BDL	0.001	BDL	BDL
S5	BDL	0.0007	0.0007	0.004	BDL	0.002	BDL	0.001
S6	0.0004	0.0007	0.0004	0.003	BDL	0.001	0.0013	0.007
S7	BDL	BDL	BDL	BDL	0.0007	0.001	0.0007	0.006
S8	0.0004	0.0007	0.0004	0.003	BDL	BDL	BDL	BDL
S9	BDL	BDL	BDL	BDL	0.0007	0.001	0.0007	0.005
S10	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL
S11	0.008	0.011	0.0111	0.053	BDL	BDL	BDL	BDL
S12	0.011	0.004	0.0029	0.023	0.0147	0.021	0.0207	0.099
S13	0.003	0.004	0.0029	0.025	0.0213	0.007	0.0053	0.043
S14	BDL	0.015	0.0114	0.076	0.0053	0.007	0.0053	0.046
S15	0.0007	0.020	0.0164	0.102	BDL	0.029	0.0213	0.143

Heavy metal contamination in drinking water can seriously damage human health. The calculated noncarcinogenic HQ values for toxic metal ions varied between 0.001–0.038 for Zn, 7e − 4–0.04 for Cd, 0.03–1.17 for As, and 5e − 4–0.07 for Pb (Table 10) and were all below 1, indicating no adverse effects and potential noncarcinogenic risk on human health. While in children, the noncarcinogenic HQ values due to Zn, Cd, As and Pb varied in the range of 0.002–0.07, 0.002–0.98, 0.05–2.19, and 0.001–0.13, respectively, indicating that As might cause adverse effects and noncarcinogenic risk to human health. In most of the samples, the calculated HQ values were found to be less than 1 for adults; however, for children, As exhibited HQ values greater than 1 at selected locations. This demonstrates that adults are at a lower health risk than children. It has been noted from the total HQs that the order of the toxicity of the studied toxic metal ions to noncarcinogenic health risk as As > Pb > Zn > Cd for adults and As > Cd > Pb > Zn for children (Fig. 7). The results of the hazard index depicted that HI values of majority of the samples were found to be below 1, indicating noncarcinogenic risk on human health except certain locations where HI ranged between 0.03 and 1.29 for adults (Fig. 8) and 0.06 and 2.71 for children (Fig. 8). Although these locations have low noncarcinogenic risk, they may pose adverse health risk on human in the long term, and are to be attached significant attention. The pollution sources of high-risk sites might be attributed to point source pollution from intensive human, industrial, urban life, and agricultural activities especially during PLD period as it is gaining large amounts of industrial production wastewater and sewage.

Table 10.

Hazard quotient (HQ) values of noncarcinogenic human health risks posed in adults and children

Sampling stations	HQ—adults				HQ—children
	Zn	Cd	As	Pb	Zn	Cd	As	Pb
S1	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL
S2	BDL	0.0007	0.03	0.0018	BDL	0.0093	0.05	0.003
S3	BDL	0.0007	BDL	BDL	BDL	BDL	BDL	BDL
S4	BDL	0.0021	BDL	0.0005	BDL	0.003	BDL	0.0010
S5	BDL	0.0014	0.05	0.0026	BDL	0.013	0.10	0.0048
S6	0.0012	0.0014	0.03	0.0023	0.002	0.012	0.05	0.0043
S7	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL
S8	0.0012	0.0014	0.03	0.0018	0.002	0.0093	0.05	0.003
S9	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL
S10	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL
S11	0.0262	0.0229	0.79	0.038	0.049	0.197	1.48	0.070
S12	0.0381	0.0079	0.20	0.016	0.071	0.085	0.38	0.030
S13	0.0095	0.0071	0.20	0.018	0.018	0.092	0.38	0.033
S14	BDL	0.0307	0.82	0.055	BDL	0.285	1.52	0.102
S15	0.0024	0.0393	1.17	0.073	0.004	0.380	2.19	0.136
Min	0.0012	0.0007	0.03	0.0005	0.002	0.0027	0.05	0.0010
Max	0.0381	0.0393	1.17	0.0727	0.071	0.3800	2.19	0.1357
Avg	0.0147	0.0120	0.41	0.0234	0.028	0.1224	0.77	0.0437

Fig. 7.

HQ values of different metal ions in adults and children

Fig. 8.

Spatial distribution of hazard index (HI) values for noncarcinogenic risk for a adults and b children

Among the selected toxic metal ions, As served as the significant contributor to pose adverse effects on human body; therefore, the carcinogenic risk (CR) of As in adults and children were calculated in the study. The carcinogenic risk of Pb was not assessed because it was classified as probable human carcinogen by US EPA. According to CR classification recommended by US EPA, CR values of As varied from 0.001 to 0.025 in adults and 0.001 to 0.046 in children demonstrating the selected study area under the classification CR < 10⁻⁶, i.e., carcinogenic risk as negligible. Even though in the present study CR values of As was categorized as negligible, regular monitoring of As concentrations for drinking purposes should be strengthened as long tern exposure can induce irreversible potentially carcinogenic health risk. Furthermore, other trace metal ions such as Cd and Pb should also be attached to special attention, as they are nonbiodegradable in aquatic systems and could pose damaging health risk to human beings at lower concentrations for a long time (Tong et al. 2021).

Comparison of river water quality in LD and PLD with other studies

Physiochemical parameters in the water samples of the present study were compared with other similar studies conducted globally (Table 11). pH, EC, and TDS values were observed relatively similar during LD and PLD periods when compared with other studies. Water quality variables like DO, Ca²⁺, Mg²⁺, and Na⁺ showed decreased concentration during lockdown phase followed by an upsurge in unlock phase similar to the results of Haghnazar et al. (2022), Sharma and Gupta (2022), and Chakraborty et al. (2021a). The lesser concentration of toxic metal ions in the study area when compared with other studies internationally, might be due to the variation in the sampling period which was done before the complete operation of the industrial establishments. The present samples were collected immediately after when restrictions on industries and travel bans were relaxed.

Table 11.

Comparison of LD and PLD river water quality variables with other selected studies globally

Parameters		pH	EC (µS/cm)	TDS (mg/L)	DO (mg/L)	Na+ (mg/L)	Mg+ (mg/L)	Ca2+ (mg/L)	Fe (mg/L)	Mn (mg/L)	Ni (mg/L)	Cu (mg/L)	Zn (mg/L)	Pb (mg/L)	Cd (mg/L)	Cr (mg/L)	As (mg/L)
Present study	LD	6.48	324	211	6.72	20.76	4	6.02	-	BDL	BDL	BDL	BDL	BDL	BDL	BDL	BDL
	PLD	7.69	2040	1326	8.72	230	37	34.43	-	0.03	BDL	BDL	100	0.8	0.15	BDL	0.14
Haque et al. 2023	Pre-LD	6.2	-	80.35	8.12	54.5	7.2	45	0.88	118	146	154	181	67	33	161	55
	Post-LD	6.96	-	69.59	7.97	14	6.9	26	1.9	366	11.3	3.87	19.4	3.53	0.1	6.91	58
Haghnazar et al. 2022	Pre-LD	-	952.4	-	-	0.89	0.69	1.48	-	-	21	5.2	29	15	-	-	3.2
	Post-LD	-	1113.8	-	-	5.08	1.13	3.17	-	-	31	1.1	39	14	-	-	3.6
Tokatlı and Varol 2021	Pre-LD	8.89	1242	-	8.9	-	-	-	-	26.3	9.06	4.3	8.18	0.51	0.046	13.76	3.51
	Lockdown	6.65	1270	-	6.65	-	-	-	-	28.01	0.97	0.71	1.29	0.15	0.021	0.83	1.4
Sharma and Gupta 2022	Lockdown	7.44	296	276	8.68	-	43	108	-	-	-	-	-	-	-	-	-
	Post-LD	7.7	288	260	5.66	-	60	102	-	-	-	-	-	-	-	-	-
Chakraborty et al. 2021a	Pre-LD	7.46	1157	740	4.87	-	70	131	674	-	84	-	39	-	10	87	-
	Lockdown	6.92	820	524	7.29	-	27	64	132	-	6.5	-	53	-	3	33	-
	Post-LD	7.42	921	589	6.3	-	46	100	284	-	22.36	-	75	-	5	41	-
Kutralam-Muniasamy et al. 2022	Pre-LD	7.73	1187	-	3.7	-	-	-	-	-	-	-	-	0.012	-	-	0.008
	Lockdown	7.76	1220	-	3.36	-	-	-	-	-	-	-	-	0.010	-	-	0.010
	Post-LD	7.74	1416	-	3.66	-	-	-	-	-	-	-	-	0.013	-	-	0.0096
Pant et al. 2021	Pre-LD	8.09	956	563	1.59	44.43	15.54	50.55	-	-	-	-	-	-	-	-	-
	Post-LD	7.37	431	227	3.18	37.39	13.25	36	-	-	-	-	-	-	-	-	-

Conclusion

This work is first of its kind to present the impacts of COVID-19 lockdown in a prominent river system of Western Ghats. In the present study, the river water quality of Periyar river in an urban industrial belt has been assessed to comprehend the industrial impact during COVID-19 lockdown (October 2020) compared with the unlock phase or post-lockdown phase (January 2021). A metadata analysis of river water quality variables and various indexing methods indicates that lockdown procedures have resulted in significant reduction in industrial discharge of untreated wastes and thereby decreased concentrations of different parameters. Since the study was conducted immediately after COVID relaxations, results show the impact of reduced industrial and urban activities in improving the overall water quality. Complete closure of industries and commercial sectors helped to improve water quality by reduced mixing of effluents directly discharged to the river water.

From the analysis, it is evidenced that the hydrochemical parameters displayed bipolarity with a substantial difference between the LD and PLD samples indicating the influence of effluents of river side industries in bringing harmful effects in Periyar river lower catchments. The results of drinking water suitability suggested a total of 93% samples in excellent and good category during lockdown; on the other hand, only 47% of samples were found suitable for drinking during post-lockdown period depicting very significant positive changes in RWQ during lockdown. Indicators of irrigational quality determined using SAR, KI, PI, and Mg Haz showed a gradual unsuitable nature of RWQ towards post-lockdown coinciding with the results of WQI. The results of statistical analysis to measure the interrelationship and origin of various parameters indicated EC and TDS exhibited a strong positive correlation with Ca²⁺, Mg²⁺, Na⁺, K⁺, TH, SO₄²⁻, and Cl⁻ during both the periods as well as strong positive correlations within the alkaline earth elements (Ca²⁺ and Mg ²⁺) and alkalis (Na⁺ and K⁺). Health risk assessment based on dissolved trace elements like As, Pb, Cd, and Zn showed an order of As > Pb > Zn > Cd for adults and As > Cd > Pb > Zn for children. The HQ and HI values obtained were below 1 indicating acceptable limit signifying no potential noncarcinogenic risk via oral exposure for Zn, Pb, and Cd whereas As values were above 1 at some sites indicating adverse effects on human health may occur.

In general, it is evident from this assessment that COVID-19 lockdown served as a ventilator allowing the river water quality to improve. During post-lockdown phase, every public and private sector especially industrial and commercial activity resumed waste discharges to river water directly and as a consequence, RWQ is undergoing progressive deterioration. Findings from this study provide an example of how a highly industrialized river belt has responded positively (self-purification) to COVID-19 lockdown and that the rapid recovery from a heavily contaminated river basin is only feasible with limited industrial activities. The results further imply that there is a need for regular monitoring of river to reduce the industrial stress and recover the river resilience, water quality, structure, and ecological functions. Regular monitoring of industrial effluents may be checked through the concerned approved agency and ensure that it is meeting the desired standards for the discharge and the data should be validated with the data provided by the industries. Stringent measures may be taken to reduce the chemical contaminants of industrial discharge by adopting proper technologies. The study also suggested that it is worth controlling the rampant industrial/anthropogenic activities in the vicinity of the river to improve the water quality rather than investing a huge amount of economic resources for the restoration. Since the industrial establishments are responsible for major share of air pollutants in the study area, in future, it is highly appreciable that certain measures have to be done for accounting the concentration of various gases especially in climate change scenario. Also, the land-air interactions can be studied by analyzing the concentration of various parameters in air/soil and water is a topic of relevance in industrialized locations. Often the wasteyards near urban settlements catches fire releasing toxic air pollutants into the atmosphere. When it rains, these pollutants get mixed with rainfall and will eventually reach soil and water mediums on land, thus causing pollution and toxicity in these systems. Brahmapuram, a place near the study area, is one such example which gained attention nationally very recently. Hence, the proper assessment of the industrial vicinities in river basin deserves much attention in environmental health perspectives. Accordingly, the findings of the study could be useful to the researchers, policy-makers, and other concerned stakeholders for the sustainable management and restoration of polluted river systems across the globe.

Author contribution

SKA—conceptualization, sample collection, formal analysis, data curation, investigation, concept, writing original draft, validation, methodology, visualization. AK—conceptualization, formal analysis, data curation, supervision, concept, writing original draft, validation, project administration, resources. AKK—conceptualization, data curation, formal analysis, visualization.

Funding

Ministry of Earth Science, Govt of India.

Data availability

All data generated or analyzed during this study are included in this manuscript and further data related to this study if needed can be available from the corresponding author on reasonable request.

Declarations

Ethics approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

All authors agree to publish this manuscript if accepted.

Competing interests

The authors declare no competing interests.