Evaluation of a trawl to estimate the inventory of catfish in commercial ponds

Abstract

Most commercial catfish farms use multiple-batch production systems that contain fish sizes ranging from recently stocked fingerlings to fish weighing more than 1.5kg. Currently, there are no accurate and cost-effective techniques to estimate pond inventories of catfish. Three trials were conducted in earthen catfish ponds with a 12m×8m trawl (bar mesh at mouth of 3.8-cm; 2.54-cm at cod end). The first two trials included trawling with and without feed as bait in 4-ha commercial catfish ponds (Trial 1) and in 1.6-ha experimental ponds (Trial 2). In Trial 3, the trawl was pulled three times without feeding in 1.6-ha experimental ponds. Mean individual weights and size distributions of catfish caught by the trawl were statistically similar to those of catfish caught by the seine (Trial 1) and to stocking data (Trial 2). Single pulls with the trawl resulted in 73% error in estimated fish inventories in Trial 3. However, if pulled two times, the estimated catfish inventory error dropped to 7%, more accurate than other inventory methods, but was variable (coefficient of variation=186%). The trawl appears to have potential as a means to sample commercial catfish ponds to estimate means and frequency distributions of fish size, but additional refinement and correction would be needed for it to be used reliably to estimate total pond inventory.