Elsevier

Ecological Engineering

Volume 169, November 2021, 106314
Ecological Engineering

Fuzzy logic approach for the assessment of trophic state of water bodies

https://doi.org/10.1016/j.ecoleng.2021.106314Get rights and content

Abstract

Eutrophication has emerged as one of the main threats to surface water bodies. The parameters for assessing eutrophication are not generally measured regularly from the lakes of India as part of monitoring programmes and it is necessary to express the trophic status in terms of secondary indicator variables that are included in the routine analysis by the State/Central government. In this paper, the trophic status of Ashtamudi Lake has been studied, considering its international importance and socio-economic relevance. The cause and response variables such as total phosphorus, Secchi disc depth, chlorophyll-a and the secondary indicator variables such as pH, turbidity, DO, EC, Salinity, TDS and BOD were analysed during the pre-monsoon season. The lake was predominantly classified as eutrophic, with some areas coming under a hypereutrophic state. A comparison was made among the five methods of water quality index by the use of secondary indicator variables to identify the appropriate method for simulating trophic status and the WQI based on the logarithmic method considering pH, turbidity, DO and BOD was identified to predict the trophic state very well. Further, the application of numerical strategy such as fuzzy logic was used to determine water quality indexes and trophic status, which can define the quality of a water body as a consequence of the variation of environmental parameters. The developed approach was further validated by using historical data for the years 2013–2015, which were regularly monitored by the Kerala State Pollution Control Board. The trophic state predicted by the approach was found to be in agreement with that estimated using Carlson's method. Thus the developed model, based on secondary indicator parameters that are readily available from government agencies, can be used to assess the trophic state of lakes, thereby assisting policy makers to frame regulations to minimize eutrophication.

Introduction

Eutrophication is the enrichment of nutrients, mainly nitrogen and phosphorus, in water bodies. United Nations Environmental Protection (UNEP) reported that globally 30–40% of lakes and reservoirs show a tendency towards varying degrees of eutrophication (Farley, 2012). It is mainly caused due to human activities, increased land usage and the application of fertilizers, which is a major source of nutrients. Due to natural processes like flood, heavy runoff causes these nutrients to enter into the water bodies which may further result in eutrophication (Yang et al., 2008). It severely deteriorates water quality leading to increased turbidity, cyanobacterial blooms, loss of biodiversity, health hazards, diminishing aquatic growth caused by depletion of oxygen, and foul taste and odour (Havens, 2008). Increased phytoplankton biomass leading to algal blooms, venomous or poisonous plant species, raised macro algae biomass, reduced transparency of water, depletion of oxygen or hypoxia, slashed species diversity and also the changes of dominant biota are some of the major ecological impacts of eutrophication. This, in turn, creates socioeconomic challenges, such as increased water treatment costs, difficulties in fulfilling the criteria for disinfection by-products, and aesthetic damage (Chislock et al., 2013). Eutrophication management is, hence, the primary step towards the conservation of water bodies.

Eutrophication has been recognized as a challenge in freshwater systems for several years, but concern about the widespread occurrence of eutrophic conditions in estuarine systems has only been increasing in the last three decades. For a proper formative assessment of eutrophication, water resources must be categorized into various trophic states followed by quantitative analysis of all those states. Many researchers employed various criteria to measure the trophic status of lakes through the quantification of the trophic state index (El-Serehy et al., 2018).

Palmer (1969) developed two pollution indices on the basis of available knowledge and perspectives about organic pollution resistance, namely the Palmer algae genus organic pollution index and the Palmer algae species organic pollution index. Carlson's Trophic State Index (CTSI) is commonly accepted and applied to the lakes, where freshwater bodies have been categorized into four potential trophic states namely, oligotrophic, mesotrophic, eutrophic, and hyper-eutrophic based on chlorophyll-a (chl-a), total phosphorus (TP) and Secchi disc depth. Several criteria like nutrient concentration, species composition, transparency and various measures of biomass are analysed for determining trophic state of water bodies. Aizaki et al. (1981) evaluated the relationships between the index values and the parameters related to lake trophic status by examining the possibility of the application of Carlson's index on Japanese lakes. Even though Carlson's index is used widely, there arises a confusion among the assessment of three different values of trophic state index (TSI). Osgood (1982) strengthened this by using variations in the three indexes to help determine the lakes' water quality. Canfield and Hodgson (1983) used data from Florida lakes to develop models for prediction of chl-a concentrations and Secchi depth. The model provides unbiased estimates of chl-a and Secchi depth across a wide variety of lake types.

Vollenweider et al. (1998) proposed a new trophic index (TRIX) in order to categorise the trophic state of inland waters focusing chl-a, total nitrogen, pHosphorous, oxygen saturation and mineral, where the index is scaled from 0 to 10. Bricker et al. (2003) addressed management options in estuaries by the development of an index for the assessment trophic status of estuaries namely, ASSETS by ranking its eutrophication status. Further studies in Pernambuco, Brazil (Alves et al., 2013) and Guanabara Bay, Rio de Janeiro, Brazil (Santos, 2015) using TRIX indicated that the performance of TRIX is better than the TSI using pH and DO developed by O'Boyle et al. (2013) as these estuaries have good sea water exchange, but the application of TRIX is localized. Gupta (2014) further modified the index by adopting nitrite‑nitrogen, chl-a and Secchi disc depth for Chilka Lagoon. In the Northern Beibu Gulf of China, Lai et al. (2014) conducted a comparative study between TRIX and ASSETS and concluded that TRIX was an indicator of organic process status for the assessment of eutrophication, while ASSETS demonstrated sensitive capacity in coastal areas.

An updated lake trophic classification model by Farnaz Nojavan et al. (2019) was presented recently that used a Bayesian method which constitutes a proportional odds logistic regression (POLR). This model operates on the current categorization of trophic status and reassesses the creation and classification methods for the TSI, thereby rethinking the classification and index for the lake trophic state.

Only a limited number of studies has been conducted on Kerala Lakes for assessing the trophic status and most of them followed the classical Carlson TSI (Carlson, 1977) approach. Sheela et al. (2011a) performed the estimation of the trophic state of Aakkulam-Veli Lake system using Carlson TSI. Sheela et al. (2011b) used Indian Remote Sensing imagery to analyse the trophic status profile based on Secchi depth and chlorophyll-a of the Akkulam-Veli Lake. Neena et al. (2019) determined the trophic status of Vellayani freshwater lake in Thiruvananthapuram district using Carlson TSI considering chlorophyll-a, total phosphorus and Secchi disk depth as parameters. Generally, the data collection of response variables such as chl-a and Secchi depth is challenging and not monitored regularly in the lake systems of Kerala, while water quality parameters are monitored regularly and readily available with Kerala State Pollution Control Board (KSPCB). Hence developing TSI based on the secondary indicator variables (water quality parameters) is highly essential for monitoring the trophic status of water bodies.

Water Quality Index (WQI) may be a valuable and distinctive rating employing a specific phrase that is valuable in choosing an appropriate recovery strategy to meet the problems involved. Horton (1965) was the first to propose the concept of indices to represent gradations in water quality in which he used the method of arithmetical aggregation to evaluate WQI. Unlike Horton (1965), Brown et al. (1970) also used simple arithmetic weight, but without the variables that multiply. The National Sanitation Foundation (NSF) sponsored this initiative by choosing the water quality parameters using the Delphi methodology (Dalkey, 1969). Dinius (1987) established an index with a diminishing range backed by multiplicative aggregation, including measures expressed in percentage of excellent quality of water that adore 100%. Extensive testing was also conducted by Helmer and Rescher (1959), Dalkey and Helmer (1963), making improvements to the Delphi system (Dalkey, 1969). McClelland (1974) introduced the geometric mean method of rating to WQI because he had been anxious about the process of eclipsing, a characteristic process in which the numerical mean lost tolerance to variables of low value. Later, Dojlido et al. (1994) used the harmonic mean to evaluate the WQI, in which there is no usage of weights for the individual indicators.

WQI applications for surface or groundwater are far more than that of coastal waters. In 2012, the USEPA evaluated the quality of coastal water using an index backed by the percentage of quality parameters ranging from excellent to poor (USEPA, 2012). Nguyen et al. (2013) suggested an updated mathematical WQI for Ha Long Bay, Vietnam, while Darko et al. (2013) used the Solway model for coastal waters in Ghana to measure WQI. Throughout modelling advanced environmental issues, there exists a tangle in making precise statements of inputs and outputs; at this time of view, fuzzy logic plays a major role in changing advanced input variables into easy output (McKone and Deshpande, 2005). Pereira et al. (2012) compared the WQI acquired through the application of the fuzzy sets with the strategy planned by the National Sanitation Foundation. By using fuzzy logic, de Oliveira et al. (2014) developed a study to assess the index of crude water quality. Gorai et al. (2016) used eleven physicochemical variables and developed a groundwater analysis study through the application of fuzzy logic. Li et al. (2016) explained the water quality of the Qu River in China using fuzzy logic which relied on Principal Component Analysis methodology to choose the most important parameters and stress how fuzzy set theory generates an additional correct result compared to various indices in the study. In general, the fuzzy logic provides several blessings in comparison to various strategies for evaluating water quality ratings, stressing the need for environmental variables not to validate weights. For this reason, formal reasoning was implemented for the production of this work, for each one of the WQI and as such the TSI, and a much stronger analysis of the large body of water was permitted in an extremely quick yet additional detailed approach.

The main objective of the study is to analyse the trophic state of Ashtamudi Lake using the readily available secondary indicator parameters. The trophic state of Ashtamudi Lake is estimated using Carlson TSI based on cause and response variables such as TP, Secchi disc depth and chlorophyll-a. An approach based on water quality index based on secondary indicator parameters such as pH, DO, turbidity and BOD is proposed for representing the trophic state of water bodies. An attempt has been made to develop a fuzzy inference system to determine WQI and TSI, which can define the trophic state of a water body as a consequence of the variation of environmental parameters. The proposed approach can be used to predict the TSI of a water body using secondary indicator variables that are regularly monitored by Government agencies. This will help in identifying the trophic state of the water body, thereby assisting policy makers to frame regulations to minimize eutrophication.

Ashtamudi Lake, Kerala's second largest and deepest estuarine ecosystem, is situated in the district of Kollam, Kerala (India), with an area of around 32 km2. The lake lies between 8° 53′- 9° 2′N latitude and 76° 31′– 76° 41′E longitude. The location map of Ashtamudi Lake is shown in Fig. 1. Ashtamudi wetland is included in the list of wetlands of worldwide importance, which is defined by the Ramsar Convention on wetland protection which property use. Ashtamudi is recognized as a coastal estuarine lake of brackish water (MOEF Classification).

Data available on the ecological and hydrological parameters of Ashtamudi Lake showed the abundance of phytoplankton (Threasimma and Nair, 1980). Divakaran et al. (1982) reported the seasonal variations in the lake ecology. In addition to this study, the mode of distribution of major organic and inorganic nutrients in Ashtamudi backwater, its general ecology, ecology of grass beds, fishery resources, benthic macrofauna sediments, mineral metals, heavy metals and their seasonal variations were also reported (Nair et al., 1983a, Nair et al., 1983b, Nair et al., 1983c; Nair et al., 1984a, Nair et al., 1984b). Numerous studies are available regarding the features of Ashtamudi Lake. The effects of man-made changes in the mangrove ecosystem of Ashtamudi were reported by Mohandas et al. (1994). Biodiversity status and restoration measures of the Ashtamudi backwater systems (Bijoy and Abdul, 1996), nutrient dynamics of Ashtamudi Lake (Sujatha et al., 2009) and sedimentary characteristics along the Ashtamudi estuarine system (Soumya et al., 2011) and analysis of the effluents discharged to Ashtamudi Lake from China clay industry (Suma et al., 2012), were also well documented. From the above observations, it is presumed that the lake is now suffering from the deterioration of ecosystem and water quality, accumulation of sediments, waste disposal, eutrophication and above all a loss of the aesthetic value. Thus monitoring programs are necessitated so as to assess the trophic status and water quality of this lake system thereby imposing appropriate management measures for the conservation of the lake.

Section snippets

Water sampling and analysis

Water samples were collected at regular intervals from fourteen locations in the Ashtamudi Lake, stretching from Neendakara harbour area to the Kallada river mouth (i.e. from saline zone to fresh water zone) during the pre-monsoon season (March 2020). Fig. 2 shows the location of sampling stations. The water samples were analysed for pH, turbidity, electrical conductivity (EC), salinity, total dissolved solids (TDS), total phosphorus (TP), dissolved oxygen (DO), biochemical oxygen demand (BOD),

General characteristics

The water samples were analysed for pH, turbidity, conductivity, salinity, total dissolved solids, total phosphorus, dissolved oxygen, BOD, Secchi depth and chlorophyll-a. The analyses of water samples for different physico-chemical parameters were followed from APHA (1999). A statistical summary of the parameters measured is given in Table 3.

Salinity varies from 28.7 ppt to 6.35 ppt, from the coastal region to the riverside. The variation of pH was between 8.3 and 7 and a lower pH was found

Conclusions

The study analysed the trophic status of Ashtamudi Lake using Carlson TSI and proposed a methodology to categorise the trophic state of a water body using readily available secondary indicator parameters. The lake was classified as predominantly eutrophic with a few locations coming under hypereutrophic category according to Carlson TSI. The water quality parameters such as pH, DO, BOD and turbidity can be used as indicator variables of eutrophication parameters namely, chlorophyll-a, total

Declaration of Competing Interest

There is no conflict of interest.

Acknowledgements

This work was supported by the Centre for Engineering Research And Development (CERD) [No.KTU/RESEARCH 2/4068/2019], APJ Abdul Kalam Technological University, Trivandrum and a part of TEQIP Research Seed money project.

References (56)

  • D.E. Canfield

    Prediction of chlorophyll a concentrations in Florida Lakes: the importance of phosphorus and nitrogen

    J. Am. Water Resour. Assoc.

    (1983)
  • D.E. Canfield et al.

    Prediction of Secchi disc depths in Florida lakes: Impact of algal biomass and organic color

    Hydrobiologia

    (1983)
  • R.E. Carlson

    A trophic state index for lakes

    Limnol. Oceanogr.

    (1977)
  • M.F. Chislock et al.

    Eutrophication : causes, consequences, and controls in aquatic ecosystems

    Nat. Educ. Knowl.

    (2013)
  • N. Dalkey

    An experimental study of group opinion: the Delphi method

    Futures.

    (1969)
  • N. Dalkey et al.

    An experimental application of Delphi method to use of experts

    Manag. Sci.

    (1963)
  • H.F. Darko et al.

    A number description of ghanaian water quality—a case study of the Southwestern and Coastal Rivers systems of Ghana

    J. Environ. Prot. (Irvine Calif).

    (2013)
  • M.D. De Oliveira et al.

    Nova abordagem do Índice de Qualidade de Água Bruta utilizando a Lógica Fuzzy

    Eng. Sanit e Ambient

    (2014)
  • S.H. Dinius

    Design of an index of water quality

    J. Am. Water Resour. Assoc.

    (1987)
  • O. Divakaran et al.

    Seasonal variation of zooplankton of the Ashtamudi lake, south-west coast of India

    Mahasagar Bull. Natl. Inst. Oceanogr.

    (1982)
  • J.A.N. Dojlido et al.

    Water quality index applied to rivers in the vistula mg / l

    Environ. Monit. Assess.

    (1994)
  • M. Farley

    Eutrophication in fresh waters: an international review

  • A. Farnaz Nojavan et al.

    Rethinking the lake trophic state index

    PeerJ

    (2019)
  • A.K. Gorai et al.

    Prediction of ground water quality index to assess suitability for drinking purposes using fuzzy rule-based approach

    Appl Water Sci

    (2016)
  • M. Gupta

    A new trophic state index for lagoons

    J. Ecosyst.

    (2014)
  • K.E. Havens

    Cyanobacteria blooms: effects on aquatic ecosystems

    Adv. Exp. Med. Biol.

    (2008)
  • O. Helmer et al.

    On the epistemology of the inexact sciences

    Manag. Sci.

    (1959)
  • R.K. Horton

    An index number system for rating water quality

    J. Water Pollut. Control Fed.

    (1965)
  • Cited by (9)

    • Anthropogenic activities accelerated the evolution of river trophic status

      2022, Ecological Indicators
      Citation Excerpt :

      Trophic status refers to the water quality status of water bodies corresponding to the content of nutrients such as nitrogen and phosphorus, mainly including oligotrophic, mesotrophic, and eutrophic (Sruthy et al., 2021).

    • Groundwater hydrogeochemical characterization and quality assessment based on integrated weight matter-element extension analysis in Ningxia, upper Yellow River, northwest China

      2022, Ecological Indicators
      Citation Excerpt :

      Despite the widespread application of various methods, their limitations include weight determination, an inflexible structure, and insufficient input parameters, which are apparent (Abtahi et al., 2015; Gao et al., 2020c; Oladipo et al., 2021; Sruthy et al., 2021). Although the WQI based on classical mathematics had evaluated water quality, further studies gradually found that the WQI was inaccurate in practical applications (Oladipo et al., 2021; Sruthy et al., 2021). Because evaluating water quality is a complex problem, classical mathematics cannot fully consider the fuzziness and variability of parameters in evaluating water quality, and the incompatibility of evaluation indexes affects the accuracy of water quality evaluation results (Oladipo et al., 2021; Uddin et al., 2021).

    View all citing articles on Scopus
    View full text