Abstract (from http://onlinelibrary.wiley.com/doi/10.1111/jawr.12178/abstract): We developed a stochastic hourly stream temperature model (SHSTM) to estimate probability of exceeding given threshold temperature ( T ) for specified durations (24 and 96 h) to assess potential impacts on freshwater mussels in the upper Tar River, North Carolina. Simulated daily mean stream T from climate change (CC) and land-use (LU) change simulations for 2021-2030 and 2051-2060 were used as input to the SHSTM. Stream T observations in 2010 revealed only two sites with T above 30°C for >24 h and T s were never >31°C for more than 24 h at any site. The SHSTM suggests that the probability, P , that T will exceed 32°C for at least 96 h in a given [...]