International Journal of
Fisheries and Aquaculture

  • Abbreviation: Int. J. Fish. Aquac.
  • Language: English
  • ISSN: 2006-9839
  • DOI: 10.5897/IJFA
  • Start Year: 2010
  • Published Articles: 193

Full Length Research Paper

Stochastic modelling of Lake Malawi Engraulicypris sardella (Gunther, 1868) catch fluctuation

Rodgers Makwinja
  • Rodgers Makwinja
  • Department of Physics and Biochemical Sciences, University of Malawi, The Polytechnic, Private Bag 303, Chichiri, Blantyre 3, Malawi.
  • Google Scholar
Wales Singini
  • Wales Singini
  • Department of Fisheries Science, Mzuzu University, Private Bag 201, Mzuzu 2, Malawi.
  • Google Scholar
Emmanuel Kaunda
  • Emmanuel Kaunda
  • Aquaculture and Fisheries Science, Lilongwe University of Agriculture and Natural Resources., Bunda Campus, P. O. Box 219, Lilongwe, Malawi.
  • Google Scholar
Fanuel Kapute
  • Fanuel Kapute
  • Department of Fisheries Science, Mzuzu University, Private Bag 201, Mzuzu 2, Malawi.
  • Google Scholar
Mwamad M’balaka
  • Mwamad M’balaka
  • Monkey Bay Fisheries Research Station, P. O. Box 27, Monkey Bay, Malawi.
  • Google Scholar


  •  Received: 02 September 2017
  •  Accepted: 23 November 2017
  •  Published: 30 April 2018

Abstract

Lake Malawi continues experiencing serious depletion of most valuable fish species. Presently, commercial and artisanal fishery are forced to target less valuable fish species. Evidently, economic importance of Engraulicypris sardella in Malawi cannot be negated as it currently contributes over 70% of the total annual landings. However, such highest contribution could be a sign of harvesting pressure. Therefore, as the species continues being increasingly exploited, the development of scientific understanding through application of stochastic models is particularly relevant for present and future policy making and formulation of strategies to sustain the resource in the lake. Thus, the study was designed to forecast the annual catch trend of E. sardella from Lake Malawi. The study used time series data from 1976 to 2015 period obtained from Monkey Bay Fisheries Research Station of the Malawi Fisheries Department. The study adopted Box-Jenkins procedures to identify appropriate Autoregressive Integrated Moving Average (ARIMA) model, estimate parameters in ARIMA model and conducting diagnostic check. The study findings showed that ARIMA (2,1,1) model had least Normalized Bayesian Information Criterion (NBIC) value making it a appropriate model for the study. ARIMA (2,1,1) model  showed  that E. sardella  annual catches are positively fluctuating. Again, the model  predicted that E. sardella annual catches from Lake Malawi will increase from the annual total landings  of 71,778.47 metric tons to 104,261.20 metric tons in the next 10 years (ceteris paribus). 

Key words: Box-Jenkins, Engraulicypris sardella, Lake Malawi, autoregressive integrated moving average (ARIMA), Modelling, Usipa, Stochastic.