State-space models (SSMs) are a popular tool for modeling animal abundances. Inference difficulties for simple linear SSMs are well known, particularly in relation to simultaneous estimation of process and observation variances. Here we identify improvements for inference about nonlinear stage-structured SSMs fit with biased sequential life stage data using both theoretical and simulation-based assessments. The model is applied to modelling the life cycle of an endangered fish, Delta Smelt, which has been the focus of much political and legal controversy. This talk includes the historical and political background that motivated the statistical work and discusses the centrality of statistical methodology to guiding and justifying management actions aimed at protecting the fish.