The present study was carried out at the Thycattussery area in Vembanad lake and assessed the ichthyo-diversity, variability in climatic variables, the exploited status of predominant small-scale sector fisheries (SSFs) and impact of climatic variables upon an existing SSF. Fish, as well as clam specimens collected from the study area, were identified. Diversity indices and dominance curves helped to identify monsoon (June–September) as the diverse season in the wetland. SIMPER (similarity percentage) analysis indicated that Villorita cyprinoides (clam) was the predominant species in the wetland with a mean relative abundance of 16.1%. Canonical correspondence analysis (CCA) of variables with clam production identified calcium hardness and rainfall with axis loadings of –0.56 and 0.50 respectively as the variables predominantly contributing to clam production. Stepwise regression indicated that temperature and rainfall were the determinants of clam production. A decrease in rainfall and an increase in temperature at an annual rate of 0.02 and 0.8%, respectively, decreased the clam production by 5.37%/year in the study area. Further, the generalized linear model (GLM) indicated stagnancy in clam production until 2035 below 400,000 kg per year. The study introduced diversification of livelihood systems using clam culture in climate resilient pen structures (CRPS) as an adaptation strategy.

• The study identified climate change as the major stressor affecting small scale fishers in Thycattussery region of Vembanad lake (wetland).

• The study exposed a huge decline in the production of a predominant single species (Villorita cyprinoides) fishery to the tune of 5.37% per year and attributed the decline to climatic variables such as temperature and rainfall.

• The study predicted that the clam production will have a stagnant phase till 2035 below 400 tonnes per year in the study area, which in turn will have an unanticipated loss in catch per unit effort of the fishers.

• The study proposes suitable fishing technology with lesser investments (Climate Resilient Pen Structures) as the appropriate adaptation strategy for restoring the declining fishery in the region.

• This strategy brought about by Central Inland Fisheries Research Institute helps the small-scale fishers through diversification of livelihood systems, wherein farmers switch between farming in CRPS (Climate Resilient pen Structures) and fishing in response to seasonal and inter-annual variation in black clam fisheries.

### Graphical Abstract

Graphical Abstract
Graphical Abstract

Inland fisheries are declining rapidly due to the increase in anthropogenic pressures which include climate change and overexploitation (Sarkar et al. 2019a, 2019b). Anthropological attributes such as growth and development have significantly impacted fisheries and the various assets the system brought in to the communities (FAO 2018). The FAO (2017) revealed that nearly 90% of inland fisheries of the growing economies was contributed by small-scale fishers. Small-scale sector fisheries (SSFs) help in the generation of derived foreign exchange and in greater contributions to the gross domestic product (via income multiplier effects and taxes). They therefore have enabled a stable and growing economy, food and nutrition security, and buffer against extreme poverty and vulnerability (FAO 2017) in emergent nations. The opportunistic nature of SSF has supported the livelihood of the associated local communities during various catastrophic events (Sarkar & Borah 2018). Despite being adept in the fisheries environment, stressors such as competition for resources, over-harvesting, pollution, environmental degradation and rapid development caused the decline in small-scale fisheries (FAO 2017). Apart from these stressors, climate change was identified by Salagrama (2012) as an additional layer in a complex web of factors affecting fisheries and fishing communities. Some of the principal climatic factors affecting fisheries are sea-level rise, sea level salinity, seasons and seasonality patterns, wind processes, rainfall, tidal amplitude, etc. Frequent incidences of juveniles in the catch, spatiotemporal alteration of inland fishery resources, negative impacts on postharvest and trade arrangements (Salagrama 2012), destruction of traditional breeding grounds, the proliferation of dead fishing zones and pollution sinks (FAO 2017) represent the effects of climate change on the livelihood of SSFs. These vulnerable small-scale fishers required an appropriate adaptation strategy in the form of shifting of fishing operations, using better fishing gear and technologies with lesser investments in capital and operating costs. Understanding the response of small-scale fishers to this unprecedented challenge is critical for developing more efficient management strategies and alternative options in regional socio-ecological systems.

With this background, the present study assessed the ichthyo-diversity, the variability in hydrological and climatic variables, the exploited status of black clam fisheries and the impact of climatic variables upon the fisheries of a tropical inland wetland in India. The study provided a better adaptation strategy to improve the resilience and productivity of small-scale fisheries of Villorita cyprinoides (black clam) associated with Vembanad Lake.

Vembanad, the largest tropical wetland system which is spread over 2,033 km2, is bordered by the Alappuzha, Kottayam and Ernakulam districts of Kerala (Kurup et al. 1990). The geographical location of the wetland is explained by its extent (latitude 9.51°N–10.19°N and longitude 76.16°E–76.43°E). Vembanad Lake was designated as a wetland of international importance under the Ramsar Convention in 2002 (GOI 2008) and a critically vulnerable coastal area (Singh 2016) after recognizing its ecological significance as an indispensable ecosystem service provider and an essential habitat. The Thanneermukkom saltwater barrier, a man-made barrier, divided the lake into two sectors based on the salinity of the water as a brackish water zone (towards the north of the lake) and a freshwater zone (towards the south of the lake). The area is noted for two fishery resources (Kurup et al. 1990), namely black clam (Villorita cyprinoides) and Pearl spot (Etroplus suratensis).

### Sampling area

The sampling area (Figure 1) is geographically located at latitude 9.77°N and longitude 76.34°E and is situated at Thycattussery Panchayat in the Alappuzha district of Kerala, India. The sampling period extended over two years (2016–2018) across six seasons, encompassing two cycles of pre-monsoon (February–May), monsoon (June–September) and post-monsoon (October–January) seasons to establish a relevant relationship between the ichthyo-diversity, black clam fisheries and associated variables at the sampling site.

Figure 1

Sampling station at Thycattussery in Vembanad lake in Kerala, India.

Figure 1

Sampling station at Thycattussery in Vembanad lake in Kerala, India.

Close modal

### Data acquisition

The physical and chemical variables of water were measured monthly during the period 2016–2018. The sampled monthly data were pooled into yearly data for better analysis and interpretation to obtain a more precise estimate of the mean value of each variable (annexure). The time-series data of climatic variables such as temperature and rainfall (in the yearly interval) procured from the Indian Meteorological Department (IMD) during the period 1987–2018 were used in the study taking into consideration its relevance in a tropical monsoon country. The month-wise data of the various fish and clam species collected from the sampling station were pooled together into yearly data to obtain a more precise estimate of the mean value of each variable (Table 1).

Table 1

Hydro-climatic variables at Thycattussery

Hydro-climatic variablesMaxMinMean ± SD
Specific conductivity (μS) 79,300.0 241.0 2,6975.5 ± 3022
pH 7.8 6.8 7.1 ± 0.3
Carbon dioxide (ppm) 4.3 0.0 1.8 ± 1.3
Alkalinity (ppm) 120.0 16.0 36.3 ± 12.0
Calcium hardness (ppm) 7.0 0.0 3.4 ± 2.8
DO (ppm) 10.4 2.1 4.8 ± 2.3
Temperature (°C) 32.8 27.0 28.8 ± 1.7
Rainfall (mm) 213.3 5.5 84.3 ± 71.5
Clam (numbers) 594 273 380.3 ± 111.7
Fish (numbers) 324 112 256 ± 53
Hydro-climatic variablesMaxMinMean ± SD
Specific conductivity (μS) 79,300.0 241.0 2,6975.5 ± 3022
pH 7.8 6.8 7.1 ± 0.3
Carbon dioxide (ppm) 4.3 0.0 1.8 ± 1.3
Alkalinity (ppm) 120.0 16.0 36.3 ± 12.0
Calcium hardness (ppm) 7.0 0.0 3.4 ± 2.8
DO (ppm) 10.4 2.1 4.8 ± 2.3
Temperature (°C) 32.8 27.0 28.8 ± 1.7
Rainfall (mm) 213.3 5.5 84.3 ± 71.5
Clam (numbers) 594 273 380.3 ± 111.7
Fish (numbers) 324 112 256 ± 53

The hydrological variables such as temperature, pH and conductivity were measured in-situ using a Eu-tech multi-parameter probe and transparency was measured using Secchi-Disc Depth. The chemical characteristics of water, such as alkalinity, dissolved oxygen (DO) and hardness, were measured using a titrimetric method (APHA 2012). Fish specimens were procured at Thycattussery by experimental fishing using gillnets (based on the number of hours of operation) and from market surveys monthly. The specimens were then brought to the laboratory for taxonomical identification (Talwar & Jhingran 1991). The black clam was procured by raking (based on the number of hours of operation).

### Software used

Plymouth Routines In Multivariate Ecological Research (PRIMER) is a statistical package used for analyzing species sampling data for studies pertaining to community ecology and the environment in the scientific community (Clarke & Gorley 2015). R is a programming language and free software environment widely used among statisticians and data miners for statistical computing and graphics (R Core Team 2013). The pooled year-wise data of all the collected variables in the study (hydrological and climatic variables and abundance data of fish and clam species) were analyzed using various packages in softwares such as Primer 7.0 (Clarke & Gorley 2015) and R (R Core Team 2013).

### Data analysis

#### Data transformation and biodiversity indices

The study describes fish and clam abundance data as a measure of the number of fish or clam in a given area (McLachlan & Brown 2006). The fish and clam abundance data were normalized using log (x + 1) transformation to moderate the influence of extreme values of fish assemblage (such as zero) on temporal scales and to estimate various indices which assess the status of fish assemblage in the area of study. Four major biodiversity indices, namely dominance index, Pielou's evenness index, Margalef's richness index and Shannon–Weiner's diversity index, were used to estimate the discrepancy of fish communities or populations in this system.

The dominance index is a uniformity index which indicates a high dominance of a species against other species (Harper 1999). The minimum index value is 1. The higher value indicates higher dominance of the species. This index helps to identify the dominant fish species in the selected area of study. It is calculated by the following equation (Harper 1999):
(1)
where ni is the number of individuals of taxon i and n is the total number of individuals in the sample.
Margalef's species richness is a measure of the number of species found in a sample or community and its value ranges between 0 and 1. This index provides details about the species richness of the fish species in the study area selected. It is calculated using the following equation (Margalef 1968):
(2)
where N is the total number of individuals; ln is the natural logarithm; S is the number of taxa; e is the natural logarithm equal to 2.718.
Pielou's evenness index is a measure of biodiversity which indicates homogeneity of species in a sample or community (Ulfah et al. 2019). The value of the index ranges between 0 and 1. An evenness value of one indicates all species are equally abundant. It is calculated as (Pielou 1966):
(3)
where H′ is Shannon diversity index and S is the number of taxa.
The Shannon diversity index (H′) is a measure of species abundance and richness to quantify the diversity of the species. The index ranges from 1.5 to 3.5 in most ecological studies (Ulfah et al. 2019). It is calculated as (Shannon & Wiener 1963):
(4)
where Pi is the proportion within the sample of the number of the individuals of ith species, ni is the number of individuals in ith species and N is the total number of individuals.

### Data processing

A dominance curve was plotted for the study area employing the transformed data. The dominance curve remains an effective tool for measuring abundance trends in communities over time. The curves are obtained by plotting the cumulative ranked abundance against a log species rank (Jennings et al. 2001). The dominance curves across three seasons (monsoon, pre-monsoon and post-monsoon) were plotted to determine the rate of instability of the system during various seasons. The steepest and most elevated curve shows the lowest diversity and the most perturbed system state.

Further, similarity percentage (SIMPER) analysis was carried out using Primer 7.0 to identify (Sandhya et al. 2019) the species that was contributing predominantly to the fishery at Thycattussery. SIMPER analysis is a descriptive method which helps in identifying the taxa that are primarily responsible for an observed difference between two or more sample groups (Clarke & Gorley 2015). The SIMPER analysis helped in identifying the predominant fish and clam species that was contributing to the ichthyo-diversity of the study region.

On identifying the predominant species in the study area, a vulnerability assessment was conducted using a structured schedule to ascertain the level of vulnerability faced by the species despite assuring food security of the associated local population. Trend analysis of the production of species from 1987 to 2017 was evaluated to substantiate that the species was vulnerable. Canonical correspondence analysis (CCA) was carried out to establish the relationship of the prominent fishery with hydrological and climatic variables (TerBraak & Smilauer 2002) and to identify the variables which impacted the species abundance in the study area (Wei et al. 2017).

Stepwise regression was further used to ascertain the level of influence of each hydrological and climatic variable on the species abundance and to identify the predominant variable affecting the predominant fishery in the study area. The stepwise regression excludes one variable at a time, especially one with the lowest F-to-remove statistic during each iteration to find the most influential variable. The variable with the highest regression coefficient usually was considered the most predominant variable among all the variables represented in the model (Schneider et al. 2010). Based on CCA and stepwise regression, the predominant variables were segregated and used for further mathematical analyses.

Trend analysis of the predominant variables was carried out subsequently to predict the future variations in these variables based on recently observed trend data (Radhakrishnan et al. 2017). The moving average method was used for the trend analysis. The mathematical model or equations thus retrieved from trend analysis of predominant variables were used for predicting the status of the fishery of the study area. Further, generalized linear modeling (GLM) was carried out to predict the state of the fishery in the event of predicted variations in the chosen predominant variables by fitting the best model.

The ichthyo-diversity at Thycattussery was assessed and it was found that 39 species belonging to 35 genera, 24 families and ten orders were present in the study area. Various biodiversity indices estimates indicated that monsoon season (June–August) was the diverse season, though all the fish and clam species were evenly distributed across all seasons. However, diversity represented on a dominance curve indicated predominance of a single species in all seasons. The analysis using (Similarity Percentage) SIMPER indicated that Villorita cyprinoides (black clam) was the major contributor to the species abundance at Thycattussery (16.1%). Further, the vulnerability assessment specified that Villorita cyprinoides, despite being a good supplier of provisioning services, was declining in production at Thycattussery by 5.4% per year during the period 1987–2018. Canonical correspondence analysis (CCA) indicated that calcium hardness and rainfall had a predominant role in clam production. The stepwise regression analysis indicated that specific conductivity, pH, carbon dioxide and temperature were the variables influencing clam production. The best fit formula derived from stepwise regression showed the predominance of temperature on clam production (as indicated by the highest regression coefficient of the variable in the model). Thus, this study identified temperature and rainfall as the predictor variables based on the analysis (stepwise regression and CCA) carried out on hydrological and climatic variables. Taking the predominant inputs from CCA and stepwise regression, i.e. rainfall and temperature, the prediction model was fitted using the generalized linear model in order to identify the future status of clam production. The prediction model indicated that with the combined effect of temperature (increasing at an annual rate of 0.02%) and rainfall (decreasing at an annual rate of 0.8%), the clam production will have a stagnant phase in production till 2035 below 400,000 kg per year. The vulnerability of the species towards climatic variables resulted in stagnancy of the production, which would cause an unanticipated loss in catch per unit effort of the fishers. It was found that small scale sector fisheries (SSFs) based on black clam fisheries exhibited regular fluctuations. The Central Inland Fisheries Research Institute (CIFRI) has come up with an adaptation strategy to cope with climate change by diversification of livelihood systems, wherein farmers switch between farming in CRPS (climate resilient pen structures) and fishing in response to seasonal and inter-annual variation in black clam fisheries.

### Icthyo-diversity

The ichthyo-diversity at Thycattussery was represented by 39 species belonging to 35 genera, 24 families and ten orders. Similar studies were carried out by Narayanan et al. (2005), Mogalekar et al. (2015) and Ansar et al. (2017) at Aymanam, Kumarakom and Panangad-Kumbalam regions of Vembanad lake, respectively. Narayanan et al. (2005) reported 37 fish species belonging to 18 families and nine orders from Aymanam panchayath, in the Vembanad wetland. Mogalekar et al. (2015) recorded 39 species of fin fishes belonging to 27 families, 11 orders and 31 genera from the Panangad region of Vembanad lake. Ansar et al. (2017) recorded 52 fin fish species from the Kumarakom region. A comparative assessment of the studies on the ichthyo-diversity conducted at various areas of Vembanad lake pointed out that diversity was more towards the freshwater side of the wetland represented by the Kumarakom region compared to the brackish water side of the lake represented by Panangad and Kumbalam regions. Saswata et al. (2018) examined the diversity status during the periods 1987–88 and 2012–2013 and identified changes in species richness in Vembanad lake consequent to a top-down trophic cascade effect.

The seasonal variation of the species diversity across two years was represented by a presence or absence table (Table 2) which indicated that diversity in terms of species abundance was more during monsoon. The Shannon diversity index of 2.90 during monsoon followed by post-monsoon (2.30) and pre-monsoon (1.91) indicated that the system was highly diverse during the monsoon season followed by other seasons. This variation in the diversity index (Table 3) may be attributed to the availability of preferred food brought in by the feeding river systems during the monsoon season for the fish and shellfish. This is in accordance with the study by Miranda (2001), which stated that an increase in water levels during monsoon temporarily floods vegetation, increases food and shelter supply and thereby increases the number of species. Similar results were evidenced by Ansar et al. (2017) in Vembanad lake and Gbaguidi et al. (2016) in Southern Benin, West Africa. Diversity (2.90) and evenness (0.88) indices during monsoon season in the study area indicated a habitat which was characterized by the equitable distribution of individuals in the community among the wide variety of species (Magurran 1988). Gbaguidi et al. (2016) had similar findings in a study in the man-made Bewacodji lake in southern Benin in West Africa. Mogalekar et al. (2015) studied the diversity indices in Vembanad lake and concluded that Shannon Wiener diversity (H’) ranged from 2.9 to 3.4, similar to the findings of the present study. The study pointed out that the proximity of Pielou's evenness index to unity indicated a stable environment with species evenly distributed across the study area irrespective of the seasons. Gbaguidi et al. (2016) reported similar results in Pielou's evenness index in the studies conducted at Bewacodji lake in West Africa.

Table 2

Presence or absence table of species at Thycattusery

OrderFamilySpeciesPost-monsoonPre-monsoonMonsoon
Perciformes Pristolepidae Pristolepis rubripinnis √
Leiognathidae Leiognathus dussumeiri √ √
Ambassidae Ambassis ambassis √ √
Parambassis sp. √
Gerridae Gerres limbatus √ √ √
Percidae Etroplus maculatus √ √ √
E. suratensis √ √ √
Carangidae Caranx ignobilis √
Glossogobidae Glossogobius giuris √ √
Lethrinidae Lethrinus sp. √
Lutjanidae Lutjanus argentimaculatus √ √
Scatophagidae Scatophagus argus √ √
Sillaginidae Sillago sihama √
Sciaenidae Johnius dussumieri √
Clupeiformes Clupeidae Thryssa malabarica √
Stolephorus indicus √
Nematalosa nasus √
Anabantiformes Anabantidae Anabas testudineus √
Channidae Channa striatus √ √
Channa marulius √ √
Heteropneustidae Heteropneustes fossilis √
Siluriformes Mystidae Mystus malabaricus √
Ariidae Arius maculates √
Siluridae Ompok malabaricus √
Bagridae Horabagrus brachysoma √ √
Beloniformes Hyporhamphidae Hyporhamphus limbatus √ √
Belonidae Xenentodon cancila √ √
Pleuronectiformes Cynoglossidae Cynoglossus macrostomus √ √
Euroglossa orientalis √ √
Mugiliformes Mugilidae Mugil cephalus √
Cypriniformes Cyprinidae Dawkensia filamentosa √ √
Gibelion catla √
Puntius sarana √ √ √
Puntius mahecola √ √
Elopiformes Megalopidae Megalopa cyprinoides √ √ √
Venerida Cyrenidae Villorita cyprinoides √ √
OrderFamilySpeciesPost-monsoonPre-monsoonMonsoon
Perciformes Pristolepidae Pristolepis rubripinnis √
Leiognathidae Leiognathus dussumeiri √ √
Ambassidae Ambassis ambassis √ √
Parambassis sp. √
Gerridae Gerres limbatus √ √ √
Percidae Etroplus maculatus √ √ √
E. suratensis √ √ √
Carangidae Caranx ignobilis √
Glossogobidae Glossogobius giuris √ √
Lethrinidae Lethrinus sp. √
Lutjanidae Lutjanus argentimaculatus √ √
Scatophagidae Scatophagus argus √ √
Sillaginidae Sillago sihama √
Sciaenidae Johnius dussumieri √
Clupeiformes Clupeidae Thryssa malabarica √
Stolephorus indicus √
Nematalosa nasus √
Anabantiformes Anabantidae Anabas testudineus √
Channidae Channa striatus √ √
Channa marulius √ √
Heteropneustidae Heteropneustes fossilis √
Siluriformes Mystidae Mystus malabaricus √
Ariidae Arius maculates √
Siluridae Ompok malabaricus √
Bagridae Horabagrus brachysoma √ √
Beloniformes Hyporhamphidae Hyporhamphus limbatus √ √
Belonidae Xenentodon cancila √ √
Pleuronectiformes Cynoglossidae Cynoglossus macrostomus √ √
Euroglossa orientalis √ √
Mugiliformes Mugilidae Mugil cephalus √
Cypriniformes Cyprinidae Dawkensia filamentosa √ √
Gibelion catla √
Puntius sarana √ √ √
Puntius mahecola √ √
Elopiformes Megalopidae Megalopa cyprinoides √ √ √
Venerida Cyrenidae Villorita cyprinoides √ √

√ Monsoon: June–September; Post-monsoon: October–January; Pre-monsoon: February–May. Absence represented by x and presence represented by √.

Table 3

Season-wise diversity indices at Thycattussery

SeasonsN (number of species)S (number of specimens)D (dominance index)J′ (Peilou's evenness index)Shannon diversity
Pre-monsoon 15 551 2.22 0.71 1.91
Monsoon 27 610 4.05 0.88 2.90
Post monsoon 25 837 3.57 0.72 2.30
SeasonsN (number of species)S (number of specimens)D (dominance index)J′ (Peilou's evenness index)Shannon diversity
Pre-monsoon 15 551 2.22 0.71 1.91
Monsoon 27 610 4.05 0.88 2.90
Post monsoon 25 837 3.57 0.72 2.30

Monsoon: June–September; Post–monsoon: October–January; Pre-monsoon: February–May.

Further, the dominance curve was plotted (Figure 2) to identify the most stable environment based on the season dependent diversity of fish and shellfish in the study area. The dominance curve indicated an unstable ecosystem represented by steep slopes across the seasons. Figure 2 indirectly showed the predominance of a single species to the tune of 94% and above in all the seasons.

Figure 2

Dominance curve at Thycattussery.

Figure 2

Dominance curve at Thycattussery.

Close modal

SIMPER analysis in Primer 7.0 (Clarke & Gorley 2015) was used to identify the single dominant species contributing to the local fishery in the study area. The mean relative abundance of distinguishing species in the study area was then compared and species with percentage contributions >10% were identified as major contributors to local fishers (Wakefield et al. 2013). Thus, the SIMPER analysis (Table 4) indicated that Villorita cyprinoides was the major contributor to the species abundance at Thycattussery (16.1%) followed by Etroplus suratensis (13.8%) and Gerres limbatus (12.3%). The study thus proved the existence of a local fishery of Villorita cyprinoides and Etroplus suratensis in the study area.

Table 4

SIMPER analysis at Thycattussery

SpeciesMean abundance% Contribution to fish abundance
Villorita cyprinoides 4.9 16.1
Etroplus suratensis 4.4 13.8
Gerres limbatus 3.7 12.3
Megalopa cyprinoides 2.6 7.9
Ambassis ambassis 2.8 5.0
SpeciesMean abundance% Contribution to fish abundance
Villorita cyprinoides 4.9 16.1
Etroplus suratensis 4.4 13.8
Gerres limbatus 3.7 12.3
Megalopa cyprinoides 2.6 7.9
Ambassis ambassis 2.8 5.0

### Vulnerability assessment

In order to assess the importance of black clam as a critical food source and a key element in determining the food security and nutrition strategies of the local population, a vulnerability assessment was conducted using a structured schedule. The assessment indicated that Villorita cyprinoides (black clam) was a good supplier of provisioning services which was one the ecosystem services exhibited by the species. All the fishers claimed a considerable decline in clam production at Thycattussery. The trend analysis of black clam production data (Figure 3) obtained from the Thycattussery Black Clam Co-operative societies substantiated the assessment and indicated a declining trend of 5.4% per year during the period 1987–2018. The first decadal analysis (1987–2008) of clam production indicated a huge decline to the tune of 8.8% per year compared to the last decade (2009–2018), which exhibited a gradual decrease to the tune of 0.9% per year.

Figure 3

Black clam production data at Thycattussery (trend analysis).

Figure 3

Black clam production data at Thycattussery (trend analysis).

Close modal

### Influence of hydro-climatic variables on clam production

Canonical correspondence analysis (CCA) was used to identify the predominant hydro-climatic variables influencing the clam production at Thycattussery. Axis 1 of CCA (Table 5) with an eigenvalue of 0.035 explained 66.91% of the variations among the various components compared to axis 2 (23.48%) with an eigenvalue of 0.012. Eigenvalue determines variance in the data in a particular direction. The component or axis with the highest eigenvalue is therefore considered the major component or axis which should be considered for interpretation.

Table 5

Eigenvalues of CCA of Thycattussery

AxisEigenvaluePercentage of variation explained
0.035 66.91
0.012 23.48
0.004 6.851
0.001 2.383
AxisEigenvaluePercentage of variation explained
0.035 66.91
0.012 23.48
0.004 6.851
0.001 2.383

Since the eigenvalues of axes 1 and 2 were higher compared to other axes, they were chosen for explaining the variations in the study. Table 5 further indicates that axis 1 alone could explain the variations among various components in the study area (Thycattussery) with clam production. However, a bi-plot diagram (Figure 4) utilizes two axes, mainly those with higher eigenvalues (axis 1 and axis 2) for representation of the findings. The variables of axis 1 and axis 2 in CCA with their respective scores are represented in Table 6.

Table 6

Components of CCA and their scores

Hydro-climatic variables at ThycattusseryComponent scores
Axis 1Axis 2
Specific conductivity –0.02 –0.11
pH 0.01 0.06
Co2 –0.10 –0.08
Alkalinity 0.04 –0.09
Calcium hardness –0.56 0.29
D0 –0.13 –0.14
Temperature 0.02 0.06
Rainfall 0.50 0.15
Clam –0.01 0.06
Hydro-climatic variables at ThycattusseryComponent scores
Axis 1Axis 2
Specific conductivity –0.02 –0.11
pH 0.01 0.06
Co2 –0.10 –0.08
Alkalinity 0.04 –0.09
Calcium hardness –0.56 0.29
D0 –0.13 –0.14
Temperature 0.02 0.06
Rainfall 0.50 0.15
Clam –0.01 0.06
Figure 4

Canonical correspondence analysis of hydro-climatic variables on clam production.

Figure 4

Canonical correspondence analysis of hydro-climatic variables on clam production.

Close modal

The CCA showed that clam production was positively influenced by calcium hardness with axis loadings of –0.56 and negatively influenced by rainfall which has an axis loading of 0.50 (Table 6). The study thus indicated that calcium hardness and rainfall had a predominant role in clam production. Robert (2000) found similar results wherein mean abundance (number of clams/sq.m) of Pisidium casertanum was found to be positively related to calcium hardness of the water. Similar findings were also reported in land molluscs by Wäreborn (1969). The present study was insufficient to estimate the level of influence of hydrological and climatic variables on clam production. So, a stepwise regression was used to identify the level of influence of hydrological and climatic variables on clam production and to identify the predominant variable affecting the black clam fishery in the study area.

The stepwise regression between the dependent variable, clam abundance and hydro-climatic variables indicated that a combination of physicochemical variables, such as specific conductivity, pH, carbon dioxide and temperature, impacted clam abundance and the relation was represented with this model:
(5)
where C is clam abundance in thousand kgs, S is specific conductivity in μS, P is pH of the waterbody, CO is carbon dioxide in ppm and T is the temperature (°C).

The stepwise regression carried out in the study (Table 7) indicated that variables with the highest F statistic, such as specific conductivity, pH, carbon dioxide and temperature, could influence the clam abundance as implied by the AIC (Akaike Information Criterion) value of –61.33. This model, which is the best fit formula, indicated the predominance of temperature on clam production (as indicated by the higher regression coefficient of the variable in the model) as evidenced by Kamal et al. (2012). Similar findings depicting the predominant role of temperature in seasonal abundance and other attributes of bivalvia was reported by Mark & Brown (2015). Research findings by Garg et al. (2009) and Singh (2017) substantiated the role of specific conductivity in the abundance of black clam. Soucek et al. (2001) and Garg et al. (2009) found clam abundance was positively influenced by water column pH, similar to the findings in the present study. Further, studies carried out by Garg et al. (2009) supported the findings that the clam abundance was positively influenced by carbon dioxide concentration which may be attributed to an increase in the temperature.

Table 7

Stepwise regression in R between hydro-climatic variables and clam production

Step: AIC = −61.33.
C ∼ S + P + CO + T
Degree of freedom (Df)Sum of squares (SS)Residual sum of squares (RSS)Akaike information criterion (AIC)
none  0.129 –61.33
+ CH 0.010 0.117 –60.71
+ DO 0.002 0.126 –59.66
–S 0.039 0.168 –59.39
+ AL 9.8 × 10–5 0.129 –59.34
+ R 9 × 10–6 0.129 –59.33
–T 0.040 0.169 –59.30
–P 0.064 0.193 –57.3
–CO 0.086 0.215 –55.66

Call: lm(formula = C ∼ S + PH + CO + T, data = A)
Coefficients:
(Intercept) pH CO
–0.881 –0.122 0.409 1.313 2.216
Step: AIC = −61.33.
C ∼ S + P + CO + T
Degree of freedom (Df)Sum of squares (SS)Residual sum of squares (RSS)Akaike information criterion (AIC)
none  0.129 –61.33
+ CH 0.010 0.117 –60.71
+ DO 0.002 0.126 –59.66
–S 0.039 0.168 –59.39
+ AL 9.8 × 10–5 0.129 –59.34
+ R 9 × 10–6 0.129 –59.33
–T 0.040 0.169 –59.30
–P 0.064 0.193 –57.3
–CO 0.086 0.215 –55.66

Call: lm(formula = C ∼ S + PH + CO + T, data = A)
Coefficients:
(Intercept) pH CO
–0.881 –0.122 0.409 1.313 2.216

Normally, the variable with the highest regression co-efficient is the most determining independent variable in a study (Schneider et al. 2010) and temperature was that independent variable in this study. In India, being a monsoon based tropical country, there always existed an inverse relation between rainfall and temperature (Kamal et al. 2012). Since the results from CCA indicated that rainfall was also a determining factor for clam abundance, the study has taken rainfall as a predictor to estimate the level of influence of variation in rainfall on clam production in a climate-changing context along with temperature. Thus the study identified temperature and rainfall as the predictor variables based on the analysis (stepwise regression and CCA) carried out on hydrological and climatic variables.

Changes in seasonal patterns of climatic variables will alter the hydrologic characteristics of aquatic systems, affecting species composition and ecosystem productivity (Poff et al. 2002). In India, being a tropical monsoon fed country, the climatic variables have predominant roles in determining the production from aquatic systems (Saud et al. 2012). Trend analysis of climatic variables such as temperature and rainfall was also carried out to predict the future variations in these variables based on recently observed trend data (Radhakrishnan et al. 2017). The moving average method was used for the trend analysis. The mathematical model or equations thus retrieved from trend analysis were used for the predictions. The study thus investigated the trend in temperature and found an increase in the temperature at an annual rate of 0.02% during 1987–2018, as represented in Figure 5. Further, the trend of clam production (Clam Prod) was related to the temperature and it was found that for every 0.02% increase in temperature, the clam production decreased by 5.37% (Figure 6).

Figure 5

Trend in temperature during 1987–2017.

Figure 5

Trend in temperature during 1987–2017.

Close modal
Figure 6

Time series analysis of clam production and temperature.

Figure 6

Time series analysis of clam production and temperature.

Close modal

The trend in rainfall was also investigated and is represented in Figure 7 which clearly shows that rainfall (RF) decreased at the rate of 0.8%/year during 1987–2018. Further, the clam production in relation to rainfall is represented in Figure 8 which indicates that for every 0.8% decrease in rainfall, the clam production decreased by 5.37%.

Figure 7

Trend in rainfall during 1987–2017.

Figure 7

Trend in rainfall during 1987–2017.

Close modal
Figure 8

Time series analysis between clam production and rainfall.

Figure 8

Time series analysis between clam production and rainfall.

Close modal

### Relevance of clam culture at Thycattussery

The Thycattussery area had a clam bed of 28 ha area with an estimated production of 296.9 kg/ha in 1988–1989 and this further reduced to 17 kg/ha (Kurup et al. 1990). So the clam culture was carried out at the Thycattussery area as a part of the stock replenishment programme. The trophic status of Thycattussery, which was estimated at 48.5, indicated an oligo-mesotrophic condition (CIFRI 2018) further suggesting that the production from this ecosystem may be enhanced using pens or cages by utilizing the multi-trophic levels in the aquatic ecosystem.

### Prediction models for clam production

Taking the predominant inputs from CCA and stepwise regression (i.e. rainfall and temperature), the prediction model was fitted using the generalized linear model to identify the future status of clam production.

The prediction model (Equation (6)) for clam production at Thycattussery is as follows:
(6)
where R is rainfall measured in mm (annual) and T is temperature (°C).

The prediction model (Figure 9) indicated that with the combined effect of temperature (increasing at an annual rate of 0.02%) and rainfall (decreasing at an annual rate of 0.8%), the clam production would have a stagnant phase in production till 2035, below 400,000 kg per year. The vulnerability of the species towards climatic variables resulted in stagnancy of the production, which would cause an unanticipated loss in catch per unit effort of the fishers. The production is anticipated to decrease, further hampering the livelihood of the fishers.

Figure 9

Prediction model for clam production up to 2035.

Figure 9

Prediction model for clam production up to 2035.

Close modal

The climate change-induced vulnerabilities should be combated by developing climate resilient strategies that would accelerate the development of the associated and affected community and by implementing various adaptation strategies in the vulnerable ecosystem (FAO 2017; Sarkar et al. 2018, 2019a, 2019b). A crucial part of increasing the resilience of these individuals and communities must be to eradicate poverty and provide food security for them by adopting adaptation strategies based on smart culture-based fisheries to climate change which may be multidimensional and multi-sectoral (FAO 2018). Similar to many fishing communities in tropical countries, it was found that small-scale sector fisheries (SSFs) based on black clam fishery exhibited regular fluctuations (Easterling et al. 2007). The Central Inland Fisheries Research Institute (CIFRI) has devised an adaptation strategy to cope with climate change by diversification of livelihood systems, wherein farmers switch between farming in CRPS and fishing in response to seasonal and inter-annual variation in black clam fisheries (Allison & Ellis 2001). The technical design of CRPS is represented in Figure 10.

Figure 10

Technical design of pen at Thycattussery.

Figure 10

Technical design of pen at Thycattussery.

Close modal

The CRPS at Thycattussery was made of bamboo which is a locally available low-cost material. The bamboo was erected at 1 m gaps to withstand the current and wind action in the wetland. This also helped to disperse the wind borne floating weeds such as Eichornia sp. out of the culture system. Net walls made of HDPE were used to encircle the bamboo structure to delineate the culture area from the surroundings. The pen height was kept to 1.8–2.5 m to counter the flooded condition of the wetland during peak monsoon. The height of the pen wall is decided based on the highest depth encountered during peak monsoon (July–August). CRPS reduced the fisher's effort in harvesting of clams from the wild and further helped in targeted harvesting.

In this study, 650 kg of baby clams of mean size of 3 g in total weight and 3 mm in total length were stocked at the rate of 5,000 nos/m2 in the Central Inland Fisheries Research Institute (CIFRI) fabricated CRPS for one year of culture. The annual growth rate of clams was assessed in terms of length and weight as 14.2 mm/year and 48 g/year. The average production obtained from the pen was 2,000 kg. The culture earned a revenue of $383.43 from sales of clam meat and clam shell. The meat and the shell was separated, washed and sold at$1.46/kg, whereas the shell was sold at $51.03/1,000 kg. Thus, in a very short payback period of one year, a benefit cost ratio (BCR) of 1.26 was obtained, thereby substantiating the feasibility of the project. The BCR will act as a stimuli for other small scale sector fishers to use the technology and for researchers to improvise the system by utilizing the column niche. This culture activity assimilated 915 kg of carbon wherein the carbon cost was estimated at$ 0.58/1,000 kg. The advantages of the system included protection from adverse climatic events, resilience to fluctuating water levels in wetlands on account of tides and rainfall, better biosecurity compared to normal pen enclosures, tapping of natural productivity (food), environmentally viable, low maintenance cost and high shelf life, intensive utilization of space, ease of harvest and greater production in a limited space.

Improvisation of existing CRPS into integrated multi-trophic pen culture structures by ensuring an ecosystem approach to fisheries and aqua-culture would increase the resilience of the small scale sector fisheries (SSF) and the associated communities in the coming years. Diversification of species in these CRPS would ensure conservation as well as enhancement of the vulnerable category of fishery resources by introducing various fish species in the column niche. Such culture-based fisheries fitted within an ecosystem management perspective can be a sustainable and successful food production strategy with stake-holders' participation.

This study identified this as a location specific suitable technology to support better livelihoods for the associated fisher communities and female family members in the inland aquatic system wherein double income from fish as well as black clams would be ensured. This green technology would not only provide a better livelihood but also act as a carbon sink which would help in reducing the carbon emissions from wetlands so as to curb global climate change. We recommended the application of a participatory approach to relegate the adverse impacts of climate change and the introduction of this new adaptive strategy can withstand the current climatic situation. This clam culture in CRPS in inland waters may be recommended as a context specific and community-based adaptation strategy for sustainable tropical fisheries in India. The study also emphasizes a bottom-up approach wherein the importance of proactive planning at primary stakeholders level is ensured.

The author would like to thank the Director, ICAR-CIFRI, for his kind support and guidance and financial assistance, and also CRIDA and NICRA for financial support.

All relevant data are included in the paper or its Supplementary Information.

Allison
E. H.
Ellis
F.
2001
The livelihoods approach and management of small-scale fisheries
.
Mar. Policy
25
(
5
),
377
388
.
doi:10.1016/S0308-597X(01)00023-9
.
Ansar
H. S.
Mogalekar
C.
Sudhan
D. L.
Chauhan
A. G.
Canciyal
J.
2017
Finfish and shellfish diversity of Vembanad Lake in the Kumarakom region of Kottayam, Kerala, India
.
J. Entomol. Zool. Stud.
5
(
2
),
351
357
.
APHA
2012
Standard Methods for the Examination of Water and Waste Water
, 22nd edn.
American Public Health Association, American Water Works Association, Water Environment Federation
,
Washington, DC
,
USA
.
CIFRI
2018
Annual Report
. p.
148
. .
Clarke
K. R.
Gorley
R. N.
2015
PRIMER v7: User Manual/Tutorial
.
PRIMER-E Ltd
,
Devon
,
UK
.
Easterling
W. E.
Aggarwal
P. K.
Batima
P.
2007
Food, fibre and forest products
. In:
Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change
(
Parry
M. L.
Canziani
O. F.
Palutikof
J. P.
van der Linden
P. J.
Hanson
C. E.
, eds).
Cambridge University Press
,
Cambridge
,
UK
, pp.
273
313
.
FAO
2017
The State of Food and Agriculture. Leveraging Food Systems for Inclusive Rural Transformation
.
FAO
,
Rome
, pp.
122
123
.
FAO
2018
Small-scale Fisheries. Small-Scale Fisheries Around the World
.
FAO Fisheries and Aquaculture Department
,
Rome
.
Garg
R. K.
Rao
R. J.
Saksena
D. N.
2009
Correlation of molluscan diversity with physicochemical characteristics of water of Ramsagar reservoir, India
.
Int. J. Biodivers. Conserv.
1
(
6
),
202
207
.
GOI
2008
Report on Visit to VembanadKol, Kerala, A Wetland Included Under the National Wetland Conservation and Management Programme of the Ministry of Environment and Forests
.
Planning Commission, Government of India
, p.
11
.
Harper
D. A. T.
1999
Numerical Palaeobiology
.
John Wiley & Sons
,
Chichester
,
UK
, p.
216
.
Jennings
S.
Dinmore
T. A.
Duplisea
D. E.
Warr
K. J.
Lancaster
J. E.
2001
Trawling disturbance can modify benthic production processes
.
J. Anim. Ecol.
70
,
459
475
.
doi:10.1046/j.1365-2656.2001.00504.x
.
Kamal
A. F. M. C.
Ahmmed
S.
Mohammod
A. I. C.
2012
Trends of climatic variables (rainfall and temperature) at Sylhet, Bangladesh
.
SUST J. Sci. Technol.
19
(
5
),
87
93
.
Kurup
M. P.
Sebastian
M. J.
Sankaran
T. M.
Rabindranath
P.
1990
Exploited fishery resources of the Vembanad lake
.
Mahasagar
23
(
2
),
127
137
.
Magurran
A. E.
1988
Ecological Diversity and its Measurement
.
Princeton University Press
,
Princeton, NJ
.
Margalef
R.
1968
Perspectives in Ecological Theory
.
University of Chicago Press
,
Chicago, IL
, p.
111
.
Mark
P.
Brown
K. M.
2015
Introduction to Mollusca and the class Gastropoda
. In:
Thorp and Covich's Freshwater Invertebrates
, 4th edn.
Ecology and General Biology
. (
Mark
P.
Brown
K. M.
, eds).
,
USA
, pp.
383
421
.
McLachlan
A.
Brown
A. C.
2006
Beach and surf-zone
. In:
The Ecology of Sandy Shores
, 2nd edn (
McLachlan
A.
Brown
A. C.
, eds).
,
USA
, pp.
55
64
.
Miranda
L. E.
2001
A review of guidance and criteria for managing reservoirs and associated riverine environments to benefit fish and fisheries
. In:
Dams, Fish and Fisheries: Opportunities, Challenges and Conflict Resolutions
(
Marmulla
G.
, ed.).
FAO
,
Rome
, pp.
91
138
.
Mogalekar
H. S.
Ansar
C. P.
Raman
N. N.
Jayachandran
K. V.
Kolhe
S.
2015
Fish diversity of Vembanad lake in the Panangad- Kumbalam region of Kochi, Kerala, India
.
Pollut. Res.
34
(
2
),
345
349
.
Narayanan
S. P.
Thapanjith
T.
Thomas
A. P.
2005
A study on the ichthyofaunal of AymanamPanchayath in Vembanad lake, Kerala
.
Zoo's Print.
20
(
9
),
1980
1982
.
Pielou
E. C.
1966
The measurement of diversity in different types of biological collections
.
J. Theor. Biol.
13
,
131
144
.
Poff
N. L.
Brinson
M. E.
John
W. D.
2002
Potential Impacts on Inland Freshwater and Coastal Wetland Ecosystems in the United States Aquatic Ecosystems & Global Climate Change
.
Pew Center on Global Climate Change
,
Copenhagen
,
Denmark
, p.
34
.
K.
Sivaraman
I.
Jena
S. K.
Sarkar
S.
S.
2017
A climate trend analysis of temperature and rainfall in India
.
Clim. Change Environ. Sustainable
5
(
2
),
146
153
.
R Core Team
2013
R: A Language and Environment for Statistical Computing
.
R Foundation for Statistical computing
,
Vienna
,
Austria
.
Available from: www.R-project.org/
.
Robert
T. D.
Jr.
2000
The Ecology of Fresh Water Molluscs
.
Press Syndicate of university of Cambridge
,
Cambridge
,
UK
, pp.
334
335
.
Salagrama
V.
2012
Climate Change and Fisheries: Perspectives From Small-Scale Fishing Communities in India on Measures to Protect Life and Livelihood. SAMUDRA Monograph
.
International Collective in Support of Fishworkers
,
Chennai
,
India
, p.
53
.
Sandhya
K. M.
Lianthuamluaia
L.
Karnatak
G.
Sarkar
U. K.
Suman
K.
Mishal
P.
Vikash
K.
Debabrata
P.
Yousuf
A.
B. K.
2019
Fish assemblage structure and spatial gradients of diversity in a large tropical reservoir, Panchet in the Ganges basin, India
.
Environ. Sci. Pollut. Res.
26
,
18804
18813
.
Sarkar
U. K.
Borah
B.
2018
Floodplain wetland fisheries of India: with special reference to impact of climate change
.
Wetlands Ecol. Manage.
26
,
1
15
.
Sarkar
U. K.
Koushik
R.
Gunjan
K.
Saurav
K. N.
2018
Adaptive climate change resilient indigenous fisheries strategies in the floodplain wetlands of West Bengal, India
.
J. Water Clim. Change.
9
(
3
),
449
462
.
Sarkar
U. K.
Roy
K.
M.
Srivastava
P. K.
Bose
A. K.
Verma
V. K.
Gupta
S.
Nandy
S. K.
Sarkar
S. D.
Karnatak
G.
Sudheesan
D.
Das
B. K.
2019a
Minnows may be more reproductively resilient to climatic variability than anticipated: synthesis from a reproductive vulnerability assessment of Gangetic pool barbs (Puntius sophore)
.
Ecol. Indic.
105
,
727
736
.
Sarkar
U. K.
Mishal
P.
Karnatak
G.
Sarkar
S. D.
Das
A. K.
Das
B. K.
2019b
Climate resilient pen system in floodplain wetlands: an adaptive strategy for fisheries enhancement and biodiversity conservation
.
Indian Farm.
69
(
4
),
41
44
.
Saswata
M.
Harikrishnan
M.
Shibu
A. V.
Sureshkumar
S.
Ranjeet
K.
Nandan
B.
2018
Studies on temporal variations of exploited fishery resources and their trophic levels in a tropical estuary
.
Reg. Stud. Mar. Sci.
22
,
61
69
.
Saud
B. J.
Bhattacharjya
B. K.
Verma
V. K.
Kumar
D.
Debnath
D.
Das
M.
Sharma
A. P.
2012
Effect of climate change on aquatic life with special reference to north-east region of India
.
Environ. Ecol.
30
(
4A
),
1534
1537
.
Schneider
A.
Hommel
G.
Blettner
M.
2010
Linear regression analysis: part 14 of a series on evaluation of scientific publications
.
Deutsches Arzteblatt Int.
107
(
44
),
776
782
.
Shannon
E. R.
Wiener
W.
1963
The Mathematical Theory of Communication
.
University of Illinois Press
,
Urbana
,
Illinois
, p.
117
.
Singh
A.
2016
.
Econ. Polit. Wkly.
51
(
7
),
70
75
.
Soucek
D. J.
Schmidt
T. S.
Cherry
D. S.
2001
In situ studies with Asian clams (Corbaculafluminea) detect acid mine drainage and nutrient inputs in low-order streams
.
Can. J. Fish. Aquat. Sci.
58
(
3
),
602
608
.
Talwar
P. K.
Jhingran
A. G.
1991
Inland Fishes of India and Adjacent Countries
, Vol.
2
.
A.A. Balkema
,
Rotterdam
.
TerBraak
C. J. F.
Smilauer
P.
2002
CANOCO Reference Manual and CanoDraw for Windows User's Guide: Software for Canonical Community Ordination (Version 4.5)
.
Microcomputer Power
,
Ithaca, New York
,
USA
, p.
500
.
Ulfah
M.
Fajri
S. N.
Nasir
M.
Hamsah
K.
Purnawan
S.
2019
Diversity, evenness and dominance index reef fish in Krueng Raya Water, Aceh Besar
.
IOP Conf. Ser. Earth Environ. Sci.
348
,
012074
.
Wakefield
C. B.
Lewis
P. D.
Coutts
T. B.
Fairclough
D. V.
Langlois
T. J.
2013
Fish assemblages associated with natural and anthropogenically-modified habitats in a marine embayment: comparison of baited videos and opera-house traps
.
PloS One
8
(
3
),
e59959
.
doi:10.1371/journal.pone.0059959
.
Wei
L.
Lijuan
C.
Baodi
S.
Xinsheng
Z.
Changjun
G.
Yan
Z.
Manyin
Z.
Xu
P.
Yinru
L.
Wu
M.
2017
Distribution patterns of plant communities and their associations with environmental soil factors on the eastern shore of Lake Taihu, China
.
Ecosyst. Health Sustainability
3
(
9
),
1385004
.
doi:10.1080/20964129.2017.1385004
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).