Ecological assessment of water quality in the Kabul River, Pakistan, using statistical methods

The Peshawar Valley is the most distinct region within the entire province of Khyber Pakhtunkhwa, Pakistan. This region was divided into five administrative units: Peshawar, Nowshera, Charsadda, Mardan, and Swabi Districts. The valley is located between 34°07’58’’N and 71°41’45’’E and has an altitude of about 335 m above sea level. The valley represents an irregular ellipse with its longer axis (116 km) extending from east to west and the shorter axis (84 km) extending from north to south (Salim and Khan 1973).

Except for the eastern region, the valley is surrounded by mountains, which are offshoots of the Hindu Kush mountain ranges. In the east, these mountains have a small opening near Attock through which the Kabul River drains into the Indus River. The area of the valley does not include the tribal territories adjoining the districts, which is approximately 9583 sq. km, but with the tribal territories the area increases to about 13 087 sq. km (Government of Pakistan 1998; Salim and Khan 1973).

The geology of the Peshawar Valley is diverse, and its geological history is closely related to the Great Himalayan Mountains. The mountains surrounding the valley date from the Tertiary period. No fossils are known from this area. The Kabul River has a complex basin system. Sedimentary limestones and shales mostly represent the lower basin, while the headwaters of the main tributaries rise between very complex groups of igneous and metamorphic rocks (Government of Pakistan 1998; Salim & Khan 1973; Yousafzai et al. 2010).

The Peshawar Valley is fed by four main rivers: the Kabul River, the Swat River, the Bara River, and the mighty Indus River (Figure 1). The Swat River originates in Kalam and flows southward up to the place where it empties into the Kabul River. The Bara River originates in the Tirah Hills in the south and flows toward the northeast up to the place where it joins the Kabul River that originates in the Wonay Pass, Afghanistan, and flows from west to east in the center of the valley. The Kabul River flows into the mighty Indus River (originating in Mansarwar, Tibet) in the east. The Indus River forms the eastern boundary of the valley and flows from north to south (Government of Pakistan 1998; Salim & Khan 1973).

Figure 1

The climate in the Peshawar Valley has been described as a modified Mediterranean type of climate. The summers are hot and the winters are cold. The range of temperature varies between 50°C (highest) and -3°C (lowest). Frosts may occur any time between December and February. Four seasons are recognized in the Peshawar Valley: autumn (September-November), winter (December-February), spring (March-May), and summer (June, July, and August) (Government of Pakistan 1998; Salim & Khan 1973).

The Kabul River is a transboundary river between Afghanistan and Pakistan. It is 700 km long, 563.27 km of which is in Afghanistan and 136.79 km in Pakistan (Gresswell & Huxley 1965). It emerges in northeastern Afghanistan at the base of the Wonay Pass in the Paghman Range (Selsela-e-Kuh-e-Paghman) and ends in the Indus River near Khairabad, Attock District, northwest of Islamabad, Pakistan (Government of Pakistan 1998). The depth-width ratio is highly varied in different seasons and at different sites. Although no accurate data are available for Pakistan, the depth can vary from approximately 1.5 to 3.0 m and the width from 30 to 90 m. The average discharge at the Warsak Dam is 20.5 × 10³ m³ s^-1. The differences in the river flow result from seasonal rains, as well as from glacial and snow melting (Yousafzai et al. 2010).

The river splits into three branches in the place where the Peshawar Valley drains into the Kabul River. Two of these branches flow into the Peshawar district, while the main tributary flows into Charsadda. These three tributaries, i.e. Shah Alam, Naguman, and the main Kabul River can be seen while traveling by road from Peshawar to Charsadda. The three tributaries meet at Jangi Tapoo near Prang. The Swat River originates in Kalam, flows through the hills of the Swat Valley and enters the plains of the Peshawar Valley near Abazai, where it divides into two branches: Abazai and Khiali. The two branches meet and flow into the Kabul River near Jalabella, District Charsadda, forming a single mighty Kabul River that receives water from the Bara River and the Jindai River before emptying into the Indus River near Khairabad, Attock District (Government of Pakistan 1998).

To assess the algal diversity and ecology of the Kabul River, four study sites were selected in the Peshawar Valley (Figure 1). The sites included Warsak (01), Sardaryab (02), Nowshera (03), and Khairabad (04) as shown in the map. The quality of water in the Kabul River is affected by miscellaneous industrial and domestic wastewaters, and therefore the water is more polluted down the river (Ullah et al. 2013). In the process of sampling and analysis of samples, we assume that the pollution is higher at the Sardaryab, Nowshera, and Khairabad sites because they are densely populated areas. Furthermore, small mills (Mabel factories) are located at these sites and they directly dump their wastes into the river. The Warsak site is located upstream and can be considered as a reference.

Sampling was carried out during the autumn and spring seasons of 2014-2015. A total of 30 samples were collected by scratching all substrates at the sampling sites. Samples were collected within a 100-meter radius at each individual location. At each site, samples were collected from different habitats, including phytoplankton and periphyton (epiphyton, epilithon, and epipsammon). Samples were brought in standard specimen bottles to the Phycology and Tissue Culture Laboratory, Department of Botany, University of Peshawar.

Microscopic morphology of the non-diatomaceous isolates was determined using the wet-mount staining method (Edler & Elbrächter 2010). This procedure was performed using a sterile micromanipulator to pick up algal filaments from temporarily preserved samples and then placed onto a clean glass slide on which a drop of distilled water was added beforehand. A drop of lactophenol cotton blue stain was added, and preparations were covered with clean cover slips. The slides were subsequently viewed under 10×, 20×, 40×, 60× and 100× Nikon Eclipse E200 microscope. Images of the taxa were taken with a BRESSER digital microscope.

The diatomaceous isolates were cleaned using a peroxide (H₂O₂) technique (Swift 1967). This process involves the oxidation of organic matter in a sample so that only silica walls of diatom cells remain. The empty frustules were then mounted and analyzed for their morphology.

Standard references were followed for taxonomic identification (Bellinger & Sigee 2015; Wehr 2002; Cox 1996; Prescott 1962; Desikachary 1959; Tiffany & Britton 1952; Transeau 1951; Collins 1909).

The Water Quality Index (WQI) was used to determine the class and status of the river (Mitchell and Stapp 1992).

WQI=ΣSIi $${\rm{WQI}} = \Sigma {\rm{SIi}}$$

The 100-point index is divided into several ranges corresponding to the general descriptive terms shown in Table 1.

Indices of saprobity S (Sládeček 1973; 1986) were calculated based on the identified species for each community and quantitative analysis of phytoplankton as:

(Eq. 1) S=∑i=1n(Si×ai)∑i=1nai $$S = {{\sum\limits_{i = 1}^n {({S_i} \times {a_i})} } \over {\sum\limits_{i = 1}^n {{a_i}} }}$$

where: S – Saprobity index of algal community; s_i – species-specific saprobity index; a_i – species abundance.

The integral index of river pollution (RPI) was calculated for critical chemical variables over all sampling sites for each river, based on the integral method of Sumita (1986) and Watanabe et al. (1986), and developed by S. Barinova (Barinova et al. 2006; Barinova 2011). The integral index of river pollution is based on the pollution estimates (such as saprobity indices) for each of the sampling sites. The integral indices are calculated according to equation 2 (below):

(Eq. 2) RPId=Σ(Di+Dj)×I2L $$RP{I_d} = {{\Sigma (Di + Dj) \times I} \over {2L}}$$

where: Di, Dj – saprobity indices for adjacent sites i, j; l – distance between two adjacent sites (km); L – total river length

RPIs are very conservative for stable ecosystems and can be used as reference variables for the river ecosystem stability assessment (Sumita 1986).

The integral index of aquatic ecosystem sustainability (WESI) was constructed based on the results of our studies (Barinova et al. 2006; Barinova 2011). Calculations of the index are based on the water quality ranks as determined by Sládeček’s saprobity indices and nitrate (or phosphates) concentrations.

(Eq. 3) WESI=RankSRankN−NO3 $$WESI = {{Rank{\rm{ }}S} \over {Rank{\rm{ }}N - N{O_3}}}$$

where: Rank S – the rank of water quality according to the range of Sládeček’s saprobity indices calculated for the sampling sites; Rank N-NO₃ – the rank of water quality according to the range of nitric-nitrogen concentrations.

At WESI ≥ 1, the photosynthetic level is positively correlated with the level of nitrate concentration. At WESI < 1, photosynthesis is suppressed (presumably due to toxic disturbances).

Physicochemical properties of the water from the sampling sites were measured in parallel with algal sampling. Temperature and water pH were measured at each site by using a HANNA HI98190 portable meter. Electrical conductivity and Total Dissolved Solids (TDS) were measured at each site by using a HANNA HI98192 meter. Turbidity was measured by using a HANNA HI98703 meter in the laboratory. Total Suspended Solids (TSS) values in the water samples were measured with a HACH TSS meter in the laboratory. Dissolved oxygen (DO) and Biochemical Oxygen Demand (BOD) were measured using a HANNA HI98193 meter in the laboratory. Alkalinity as CaCO₃ was analyzed by titration against standard sulfuric acid. Hardness as CaCO₃ was analyzed by complexometric titrations using EDTA (0.01M). Nitrites were analyzed by the spectrophotometric (colorimetric) method. Nitrates were analyzed by the spectrophotometric (colorimetric) method. Ammonia was analyzed by the spectrophotometric method using Nessler’s reagent. Sulfates were analyzed by complexometric titrations using EDTA (0.01M). Phosphates (PO₄^3-) were analyzed by the spectrophotometric method using Nessler’s reagent. Chlorides (Cl^-) were determined by titrating against silver nitrate (0.1N) using potassium chromate as an indicator.

Statistical analysis of the correlation between species diversity and the main water condition variables was calculated by distance-weighted least squares and Stepwise regression analysis using the Statistica 12.0 Program.

Statistical significance of variables was assessed using the Pearson correlation method from Wessa (2016).

Canonical Correspondence Analysis (CCA) from CANOCO for Windows 4.5 package (Ter Braak & Šmilauer 2002) was used to determine the relationships between species diversity in algal communities and their environmental variables.

Similarity between algal diversity and chemical variables was assessed with Sørensen indices using the GRAPHS program (Novakovsky 2004).

The dynamic water variables over the sampling sites in the Kabul River are presented in Table 2, which shows that water temperature, pH, conductivity, TDS, TSS, salinity, chlorides, and alkalinity increased down the river. Concentrations of nitrates, nitrites, and ammonia dramatically increased at site 2 immediately after the Swat River input. At the same time, dissolved oxygen, total hardness, and sulfates decreased at site 2, whereas phosphates increased, which led us to conclude that the influence of the Swat tributary on the water quality of the Kabul River is high because its water was slightly enriched with major ions but brought more pollutants to the waters of the Kabul River. A similar conclusion was reached by Yousafzai et al. (2010) on the basis of a detailed chemical investigation conducted in the same part of the river.

Chemical variables in the Songhua River, from climatically similar regions of China (Barinova et al. 2016a), also fluctuated in small ranges and reflected fresh, low-alkaline, low-to-temperate temperature, and low-to-moderately polluted waters saturated with oxygen, while the index of saprobity S ranging from 1.57 to 1.79 defined the water quality as Class III.

Altogether 209 species of algae and cyanobacteria were found at 4 sites of the Kabul River (Figure 2, Tables 3, 4). Table 3 reflects ecological preferences of some algal species from Figure 2. Table 4 presents qualitative (the number of species) and quantitative (abundance scores) data for the four sites. Species richness and algal abundance similarly increased down the river. The greatest abundance in communities was mostly represented by filamentous charophytes (Table 4). The index of saprobity S, calculated based on the species abundance, increased from 1.55 to 1.59 after site 2 at Nowshera and Khairabad sites. The index reflects moderately-polluted water with developed communities. The analysis of the algal community revealed that the Swat River (the main left-hand high-mountain tributary of the Kabul River) was not particularly rich in species, and species richness was strongly correlated with water temperature (Barinova et al. 2013). Nevertheless, the community structure in the lower part of the Swat River (below 1400 m) was similar to the studied part of the Kabul River with the dominance of filamentous algae and diatoms. It is remarkable that Charophyta algae represent about a quarter of the algal flora in both rivers, and Spirogyra as well as other filamentous algae were the most diverse groups. These characteristics can represent some regional features in the river basin with high agricultural activity. The common occurrence of green algae in the riverine community is correlated with the latitude of the river basin and therefore with sunlight intensity, as in the rivers in Ukraine (Bilous et al. 2012), China (Barinova et al. 2016a), and Israel (Barinova 2011; Barinova & Krassilov 2012).

Figure 2

Our bioindication approach to the Kabul River water quality assessment (Barinova et al. 2016b) shows differences between site 1 above the Swat River input and lower sites 2-4. Species indicators of salinity and organic pollution increased significantly after the Swat tributary.

Water quality of the Kabul River in the Peshawar Valley was measured using the Water Quality Index (WQI). Water quality was medium at Warsak (54.67 points) and Sardaryab (50.16 points) sites (Table 2). The water quality decreased at Nowshera (46.36 points) and Khairabad (39.39 points) sites. WQI changes show also that pollutants carried by the Swat tributary, flowing into the Kabul River (before the Sardaryab site), contaminate the Kabul River, and thus the water quality continues to decrease down the river.

We have no other examples of the WQI calculations for regional rivers in Pakistan, but similar results were published for the Yarqon River in Israel (Barinova et al. 2010b).

The River Pollution Indices (RPI) (Barinova 2011) were calculated for chemical and biological variables in the studied parts of the Kabul River. Table 2 shows that water in the river was low-alkaline, low-saline, fresh, but moderately organically polluted.

Our RPI calculations for a similar set of variables in other rivers such as the Songhua, the Lower Jordan, or the Southern Bug (Barinova 2011; Barinova & Krassilov 2012; Bilous et al. 2012; Barinova et al. 2010a; 2016a) show that the integral method for water quality assessment can yield relevant and stable results, which can be employed in the monitoring of regional water quality.

The Water Ecosystem Sustainability Index (WESI) was calculated on the basis of the classification rank of nitrates (Table 2) and the index of saprobity S (Table 2). As a result, WESI for each site as well as RPI-WESI were below 1.0. This calculation reflects the algal photosynthesis suppression and shows the Kabul River contamination with toxicants. Usually the concentration of phosphates in the rivers is unlimited, so was the case with the Kabul River. Therefore, we decide to calculate the WESI for phosphates in the Kabul River. The results (Table 2) confirm our conclusion on nitrate concentration. The WESI (phosphates) ranged from 0.44 to 0.67 with RPI-WESI of 0.597, which is also below 1.0.

The impact of pollution on the small tributary in the Peshawar Valley was reported in 2009 (Ullah et al. 2013) near the Peshawar city and shows that the pollution level in the river is rising from upstream (at the city’s entrance) to downstream (at city’s exit) due to the discharge of domestic wastewater effluents, agricultural activities, and solid-waste dumping directly into the river. This site is outside our research range but reflects the situation in the region.

Based on chemical analysis, Yousafzai et al. (2010) concluded that salinity as well as nutrients reach the Kabul River from industrial and domestic sources. Many authors considered the Kabul River as a large subtropical river. According to this view, we compared the studied river with rivers from similar climatic regions. Therefore, the same domestic and agricultural impact on algal communities was found in the lower Jordan River in Israel as well as in the Songhua River in China (Barinova et al. 2010a; 2016a). The WESI values were below 1.0 in both rivers at the sites affected by salinity and agriculture non-point runoff, which reflects a low-toxicity impact of algal photosynthetic activity. On the other hand, indices of saprobity S at the sites of the studied part of the Kabul River show high self-purification capacity of algal communities.