Skip to Main Content
The contingency table (Table 3) is a key tool to evaluate the occurrence of precipitation events in a binary manner (Yes/No): ‘Yes’ represents that the precipitation event occurs and ‘No’ depicts that the precipitation event does not happen. The threshold to separate the ‘Yes’ and ‘No’ precipitation event is 1 mm/day (Conti et al. 2014; Mantas et al. 2015). Therefore, there are four categories defined in the contingency table: ‘Hits’ presents a precipitation event that is observed by gauge precipitation and estimated by TRMM precipitation; ‘Misses’ presents a precipitation event that is observed by gauge precipitation, but not estimated by TRMM precipitation; ‘False alarms’ presents a precipitation event that is not observed by gauge precipitation, but estimated by TRMM precipitation; ‘Correct negatives’ presents a precipitation event that is not observed by gauge precipitation and not estimated by TRMM precipitation. As a result, the contingency table indices including probability of detection (POD), false alarm ratio (FAR), ACC, and critical success index (CSI) were used to verify the capability of TRMM precipitation to correctly estimate precipitation events at daily scale in this study. The indices are briefly described as follows:
formula
1
formula
2
formula
3
formula
4
Table 3

Contingency table

 Gauge precipitation
Contingency TableYesNo
TRMM precipitation 
 Yes Hits False alarms 
 No Misses Correct negatives 
 Gauge precipitation
Contingency TableYesNo
TRMM precipitation 
 Yes Hits False alarms 
 No Misses Correct negatives 
Close Modal

or Create an Account

Close Modal
Close Modal