Adding a couple of figures based on the scenario where there is a nitrate flux threshold and we are evaluating our model's ability to correctly forecast the exceedance of this threshold. I arbitrarily chose thresholds that were ~75-90th percentile of nitrate flux for each site, but these can be altered if we find a threshold that is more meaningful.
I'm not sure if I get a whole lot out of the fraction of forecasts exceeding threshold fig but it is helpful to get an idea of where the threshold is.
I find the forecasting skill figure useful and a good summary fig:
Adding a couple of figures based on the scenario where there is a nitrate flux threshold and we are evaluating our model's ability to correctly forecast the exceedance of this threshold. I arbitrarily chose thresholds that were ~75-90th percentile of nitrate flux for each site, but these can be altered if we find a threshold that is more meaningful.
I'm not sure if I get a whole lot out of the fraction of forecasts exceeding threshold fig but it is helpful to get an idea of where the threshold is.
I find the forecasting skill figure useful and a good summary fig:
see #59 and #58 for more details on decisions.