PLoS ONE (Jan 2018)
Threshold temperatures and thermal requirements of black soldier fly Hermetia illucens: Implications for mass production.
Abstract
Efforts to recycle organic wastes using black soldier fly (BSF) Hermetia illucens into high-nutrient biomass that constitutes a sustainable fat (biodiesel) and high-quality protein ingredient in animal feeds have recently gained momentum worldwide. However, there is little information on the most suitable rearing conditions for growth, development and survivorship of these flies, which is a prerequisite for mass production technologies. We evaluated the physiological requirements for growth and reproduction of H. illucens on two diets [spent grains supplemented with brewers' yeast (D1) and un-supplemented (D2)]. Development rates at nine constant temperatures (10-42°C) were fitted to temperature-dependent linear and non-linear day-degree models. Thereafter, life history table parameters were determined within a range of favourable temperatures. The thermal maximum (TM) estimates for larval, pre-pupal and pupal development using non-linear model ranged between 37.2 ± 0.3 and 44.0 ± 2.3°C. The non-linear and linear day-degree model estimations of lower developmental temperature threshold for larvae were 11.7 ± 0.9 and 12.3 ± 1.4°C for D1, and 10.4 ± 1.7 and 11.7 ± 3.0°C for D2, respectively. The estimated thermal constant of immature life stages development of BSF was higher for the larval stage (250±25 DD for D1 and 333±51 for D2) than the other stages evaluated. Final larval wet weight was higher on D1 compared to D2. The population growth rate was most favourable at 30-degree celsius (°C) with higher intrinsic rate of natural increase (rm = 0.127 for D1 and 0.122 for D2) and shorter doubling time (5.5 days for D1 and 5.7 days for D2) compared to the other temperatures. These results are valuable for the optimization of commercial mass rearing procedures of BSF under various environmental conditions and prediction of population dynamics patterns using computer simulation models.