April 9th forecast for some forecasts shows incorrect range

[ethanwhite]

Looks like there's something funny going on with at least the Sugar Maple forecasts which went from this on April 5th:

To this on April 9th:

I've checked about half a doze other species and don't see anything comparable, which is weird.

[ethanwhite]

Common winterberry is another example

[sdtaylor]

thanks ethan. this is a funny problem where the model says there are not enough warm days and that flowering essentially won't happen in the missing areas. It's only with 1 out of 5 of this days climate ensemble members though, but it gets propagated because of taking the average where that 1 value is nan. I can deal with ignoring nan values, but if the other issue becomes prominent than it will have to be dealt with in model selection/ building.

I'm rerunning it to put up some april 10 forecasts and they'll hopefully be back to normal.

[sdtaylor]

Sugar maple is one of the most well represented plants too. it's a shame the model failed so badly

[ethanwhite]

Yeah, I think dropping the nan's before averaging is a good idea. Happy to chat more about model selection and other changes if needed.