Happy new year everyone. Some of you are probably working with some kind of forecasting models. In Remote Sensing we are using metrics such as producer accuracy (PA) and user accuracy (UA, and others) to report a classification accuracy assessment. Now, I stumbled upon some other metrices (e.g. Heidke skill score (HSS), the Peirce skill score (PSS), the Clayton skill score (CSS), the Gilbert skill score (GSS)) and I am curious what the can do possibly better than the ones I am familiar with. These metrics seem to be used in weather forecasting and I was hoping that someone here has some knowledge about them and is able to compare them to remote sensing scores.

Cheers,

René