Variability in Oil Sardine and Indian Mackerel Fishery of Southwest Coast of India during 1991-2008 - Term Paper Example

Variability in Oil sardine and Indian mackerel fishery of southwest coast of India during 1991-2008: a statistical approach Marine fishery is distinguished by the wide variations in landings due to spatial-temporal factors as well as environmental effects. Fish populations respond to these variations in different ways: as short term (seasonal) and as long term (decadal). Special efforts have been made to modernize coastal fisheries and these efforts led to the major increase in fishery. The analysis of an 18 year time-series of oil sardine and mackerel landings from southwest coast of India is provided in the given work (Malabar Coast). The variability in landings during the period 1991-2008, which was a period with frequent and intense climatic events, has been investigated. Trend analysis has been conducted in order to reveal, whether the landings keep a short term as well as a long term trend. In the time-series analysis, a statistical prediction model has been validated using ARIMA with the observed landing data. Approximately 90% of the variation in predicted landings can be explained by the observed landings using this model. The outcomes help understand the variability in fish landings and are also important for assist management of fishery resources. Key words: Oil sardine, Indian mackerel, southwest coast of India, trend analysis, ARIMA model, prediction Introduction Fisheries play an important role in the world economy, thus, exploitation of fishery resources is widely discussed on global scale. About 200 million people worldwide depend on fishing and related industries for livelihood. Now the quantity of landings is declining in thirteen of the fifteen major marine areas (Garcia and Newton, 1994; Bathal and Pauly, 2008; Zeller et al., 2009). Marine fish populations have been known for extensive variation in landings (Ljungman, 1882). For various fisheries in many parts of the world’s oceans the variability in landings has been recognized (Kawasaki and Omori, 1995; Lluch-Cota et al., 1999). In the Indian Ocean, there is deterioration in quality caused by decline in the sizes of highly valued species and the move to bulk landings of lower-value species due to climate change (Regier and Baskerville, 1986). Fish production of India has increased more than fivefold since 1950. Special efforts have been made to promote extensive and intensive inland fish farming, modernize coastal fisheries, and encourage deep-sea fishing through joint ventures. These efforts led to the significant increase in fishery. At present, India stands seventh place among fishing countries. The South Eastern Arabian Sea (SEAS) undergoes strong upwelling during the southwest monsoon period. The Malabar Coast falling under the domain of strong upwelling is the landing area of oil sardine and mackerel. They form the major support of Indian pelagic fisheries along the west coast of India (Madhupratap et al., 1994). Along the peninsular India 80-95% of the catch of oil sardine and mackerel is from the west coast and the commercial fishery of these species is restricted to 8°N-18°N. These fisheries are shallow water based, the usual depth of fishing confined to the foreshore area within 20-25m depth. Like other parts of the world’s oceans, this fishery exhibits dramatic fluctuations from year to year. The investigations have been held to relate this fishery to the variability to seawater temperature, salinity, rainfall, upwelling etc., but the success of these studies was limited. (Madhupratap et al., 1994; Jayaprakash, 2002; Krishnakumar and Bhat, 2008). Here the variability in landings of oil sardine and mackerel are investigated along the southwest coast of India (Fig. 1). For this purpose landing data of these species for the period 1991-2008 has been used. To understand the variability in landings of these fishes “Trend Analysis” and “Time Series Analysis’ have been conducted. The statistical prediction has been validated using observed landing data. The dependence structure in the annual landings of oil sardine and mackerel are also examined. Data and Methodology Landing data of Oil sardine and Mackerel The quarterly landings of the two major pelagic species (oil sardine and mackerel) for Kerala, Karnataka and Goa representing the southwest coast as a whole are collected from the Central Marine Fisheries Research Institute (CMFRI) and used for the research. CMFRI has been following a suitable sampling design for estimating marine fish landings to arrive at species wise estimates since 1950. The landings are estimated by using a stratified multi-stage random sample design that takes into account landing centres, days and boat net combinations in fishing operations (Srinath et al., 2005). On the west coast, highest abundance of oil sardine and mackerel mainly occur along the coasts of Kerala, Karnataka and Goa. Hence the landings of oil sardine and mackerel of southwest coast of India, for the period 1991 – 2008 is utilized for the analysis. Time series models are applied in the attempt to explore the dependence structure in the landings data. The statistical packages Statistical Package for the Social Sciences (SPSS-Version 15.0) and R are used for the data analysis. Trend analysis Time series analysis is applied to reveal the main components of the observed chronological data on oil sardine and mackerel landings of southwest coast of India, comprising of the three coasts (Kerala, Karnataka and Goa). The former analysis is used to delineate the two main components namely long term variation or trend and short term variation or the seasonal index after suppressing the cyclical variation along with trend by suitable choice of the period of moving average equal to the period of the cycles, if cycles occur. The observed value (O) is assumed to be the multiplicative product of trend (T), seasonal variations (S), cyclical variations (C) and irregular fluctuations (I). O = T *S * C * I -------------------------------------------------(1) Trend is estimated by centered moving average method with period equal to 4 (Waymire and Gupta, 1981). S*C*I = (O*100)/T------------------------------------------------(2) Raw seasonal index, S= average of (S*C*I) average over the years Correct seasonal index = (Raw seasonal index*100) Average of the raw seasonal indices-------------(3) Cycles are removed by proper choice of the period of the moving average and irregular variations are removed by the process of averaging. Results and discussion Oil sardine and Mackerel landings during 1991-2008 The quarterly landings of oil sardine and mackerel for southwest coast of India during the period 1991-2008 have been presented in the figures 2a and 2b. Figures demonstrate that the contribution of oil sardine to the fishery is more as compared to mackerel. The oil sardine landings showed a decreasing trend from 1991 onwards and reached a lowest catch of that decade in 1994 which was a significant positive IOD year (extreme oceanic event). We know the relationships between ocean conditions and fish behaviour result in the variability of fish catch. Influence of the physical processes on temporal fluctuations in landings has been studied by Longhurst and Wooster (1990), Lluch-Cota et al. (1999) and Madhupratap et al. (2001a). After 1994 the catch picked up and showed an increasing trend from 1995 onwards. After three consecutive La Niña events 1998-99, 1999-2000 and 2000-01, the oil sardine landings showed a tremendous increase till 2008. The lowest landings of the oil sardine are recorded during 1991-2000, which was a decade with the occurrence of frequent and intense climatic events. The mackerel landings showed higher values during this period, because mackerel is more susceptible to severe environmental conditions. This abundance of mackerel during this decade caused to predate over the oil sardines in that area led to the collapse of the oil sardine fishery, in addition to the influence of extreme climatic events. This inverse behaviour of oil sardine and mackerel was postulated also by Hornell (1910) and Nair and Chidambaram (1951). They explained this inverse relationship in terms of competing species in a multiple fishery ecosystem. The quarterly landings data of oil sardine shows the maximum catch during Jan-Mar and Oct-Dec (the first and fourth quarters) in most of the years. Instead, mackerel shows maximum catch in the third & fourth quarter. From 1991 to 1994 there was a decreasing trend in oil sardine landings with minimum catch ( Read More