Increasing pressure on freshwater ecosystems requires river managers and policy makers to take actions to protect ecosystem health. Species distribution models (SDMs) are identified as appropriate tools to assess the effect of pressures on ecosystems. A number of methods are available to model species distributions, however, it remains a challenge to identify well-performing models from a large set of candidate models. Metaheuristic search algorithms can aid to identify appropriate models by scanning possible combinations of explanatory model variables, model parameters and interaction functions. This large search space can be efficiently scanned with simple genetic algorithms (SGAs). In this paper, we test the potential of a variable length chromosome SGA to perform parameter estimation (PE) and input variable selection (IVS) for a macroinvertebrate SDM. We show that the SGA is an appropriate tool to identify fair to satisfying performing SDMs. In addition, we show that SGA performance and the uncertainty varies as a function of the chosen hyper parameters. The results can aid to further optimise the algorithm so models explaining species distributions can be identified and used for analysis in river management.