Table 5 Model skill scores descriptions

Accuracy (ACC)

\(\mathrm{AC}=\frac{\mathrm{YY}+\mathrm{NN}}{\mathrm{YY}+\mathrm{YN}+\mathrm{NY}+\mathrm{NN}}\)

What fraction of the forecasts were correct

0 to 1

Probability of Detection (POD)

POD = YY/YY + NY

What fraction of the observed “yes” events were correctly forecast

0 to 1

Equitable Threat Score (ETS)

\(\mathrm{ETS}=\frac{\mathrm{YY}-{\mathrm{YY}}_{\mathrm{random}}}{\mathrm{YY}+\mathrm{NY}+\mathrm{YN}-{\mathrm{YY}}_{\mathrm{random}}}\)

\({\mathrm{YY}}_{\mathrm{random}}=\frac{\left(\mathrm{YY}+\mathrm{YN}\right)*(\mathrm{YY}+\mathrm{YN})}{\mathrm{YY}+\mathrm{YN}+\mathrm{NY}+\mathrm{NN}}\)

How well did the forecast “Yes” events correspond to the observed “yes” events (accounting for hits that would be expected by chance

− 1 to 1

False Alarm Ration (FAR)

FAR = YN/YY + YN

What fraction of the predicted “yes” events actually did not occur

0 to 1