Two double-limit tobit models are used to identify significant risk factors that most affect farm-raised catfish losses from weather-related events and from disease outbreaks. Results of the weather loss model indicate that the variables for operator education level, number of ponds, pond water depth, production management strategy, past experience with severe losses from low oxygen levels from off-farm power outages, past experience with severe losses from diseases, and being in the South are statistically significant. Results of the disease loss model indicate that the variables for operator experience and pond water depth are significant. Development of models explaining weather and disease losses through observable variables provides a better understanding of the interrelation between the loss perils and explanatory variables so management strategies can be developed to mitigate losses from identified risk factors.
