There is an explanation here: https://chc.ucsb.edu/data/chirps

" However, estimates derived from satellite data provide areal averages that suffer from biases due to complex terrain, which often underestimate the intensity of extreme precipitation events. "

Dear Batuhan Kavlak,

The link you have provided compared an observational data with a reanalysis. The reanalysis products includes the model forecast that may have some estimation, which prevented the reanalysis product to give an actual result, especially in case of an extreme event (e.g., heavy rainfall). What you need to see is the variability (peaks and dips) in both the datasets are matching or not.

Secondly, its a accumulated rainfall within a day, so it is not underestimating I guess. It would be, if it considered the diurnal average.

Best

Both are remotely sensed with NASA having one degree sensitivity, CHIRPS has 0.05 degree sensitivity. CHIRPS is more accurate to local conditions, but neither is probably sufficient for most field based research level hydrology. The level of sensitivity suggested in the plot by the Lat-Long values is likely just the center of the location analyzed to 1 or 0.05 degrees. If these are used regardless of this, the NASA might be better for large basins or regional study, and the CHIRPS for watersheds. When comparing, it would take 400 CHIRPS size units to make one NASA unit, so when comparing these two data, the average of the 400 CHIRPS would be compared with the one georeferenced NASA location.