there is tendency for model to overestimate water temp in the warming phase, which could lead to overestimation of GSDD
could be helped by adding separate logistic function for each warming/cooling phase to allow different slopes
can also estimate peak DOY where change in phase occurs
see https://github.com/poissonconsulting/ok-ko-runtiming-22/blob/main/stan/temperature-logistic.stan
there is tendency for model to overestimate water temp in the warming phase, which could lead to overestimation of GSDD could be helped by adding separate logistic function for each warming/cooling phase to allow different slopes can also estimate peak DOY where change in phase occurs see https://github.com/poissonconsulting/ok-ko-runtiming-22/blob/main/stan/temperature-logistic.stan