THE COMPARISON OF GROWTH WITH LENGTH-WEIGHT RELATION AND ARTIFICIAL NEURAL NETWORKS OF CRAYFISH, ASTACUS LEPTODACTYLUS, IN MOGAN LAKE

The comparison of growth with length-weight relation and artificial neural networks of crayfish, Astacus leptodactylus, in Mogan Lake

Semra Benzer, Çağlan Karasu Benli, Recep Benzer

Gazi Faculty of Education, Gazi University, Ankara, TURKEY
Institute of Natural and Applied Sciences, Gazi University, Ankara, TURKEY
Institute of Information, Gazi University, Ankara, TURKEY

Abstract

This study aimed to determine some morphological characteristics of freshwater crayfish, Astacus leptodactylus Eschscholtz 1823, populations in Mogan Lake, Turkey. Samplings were done between 2 July and 30 October in 2006 and 2007 with a random method. We present the relationships between total length (TL), carapace length (CL), chelae length (ChL), abdomen length (AL) and total weight (TW) for A. leptodactylus from Mogan Lake. Study was conducted in 112 individuals (14 female, 98 male). The research was found as 87.5 % male, 12.5 % female of crayfish thought investigation female and male ratios was of determined as to 0.14 /1.00. Avarage total length was 108.71 mm for female, 102.93 mm for male, average total weight was 28.64 g for female, 32.49 g for male. Length-weight relation equation was found for females W=0.0022 L2.01 for males W=0.00095 L2.23. The results obtained by artificial neural networks and length-weight relation equation are compared to those obtained by the growth rate of the crayfish caught from Mogan Lake. Length-weight relation and artificial neural network MAPE (mean absolute percentage error) results were examined. Artificial neural networks gives better results than length-weight relation. Artificial neural networks can be alternative as a evaluated for growth estimation.

Keywords: Artificial neural networks, crayfish, estimated, length-weight relation, Mogan Lake