Feedback on Compare_dataset module

[Rwema25]

1. Inaccurate precipitation

Here are some issues with the output:

• The precipitation value in Overall Period Statistics seems to be unrealistic, especially maximum. We can remove this block of Overall Period Statistics.

Command line used:

python -m climate_tookit.compare_datasets.compare_datasets --sources era_5 chirps nasa_power imerg nex_gddp agera_5 chirts soil_grid terraclimate tamsat --lat -1.286 --lon 36.817 --start 1990-01-01 --end 2020-12-31 --model MRI-ESM2-0 --scenario ssp245 --format report

Output:

[Sammyjoseph999]

Done.

[Rwema25]

Check if you pushed the updated code to address this issue. i am still observing error in precipitation (maximum). The output does not change:

[Sammyjoseph999]

I have committed it already, I will be pushing about 6 changes together in a moment

[Rwema25]

• Here we are getting exactly the same values at different coordinates. I checked this for Nairobi, Kigali, and Kisangani.
• Annual total values are somewhat small; we need to verify why we are getting this low values.

Command line:

python -m climate_tookit.compare_datasets.compare_datasets --sources era_5 chirps nasa_power imerg nex_gddp agera_5 chirts soil_grid terraclimate tamsat --lat -1.286 --lon 36.817 --start 1990-01-01 --end 2020-12-31 --model MRI-ESM2-0 --scenario ssp245 --format report

Output:

Nairobi: --lat -1.286 --lon 36.817

Kigali: --lat -1.969 --lon 30.139

[Sammyjoseph999]

Thanks for flagging this. I’ve just completed a verification run for the 1990–2020 period, and the results are now showing clear spatial differentiation. For instance, Nairobi (1055 mm), Kigali (1264 mm), and Kisangani (802 mm) are no longer returning identical values. Additionally, the annual totals now sit consistently within the 800–1500 mm band, which aligns much better with historical benchmarks for these regions compared to the previous outputs.

[Rwema25]

Fixed.