I am writing to request clarification regarding the “Global bioclimatic indicators from 1979 to 2018 derived from reanalysis” dataset, specifically the monthly water vapour pressure indicator.
I have been trying to reproduce the monthly BIOCLIMATE water vapour pressure indicator on the 0.5° × 0.5° grid using ERA5 2 m dewpoint temperature (d2m). However, when I compare my calculated water vapour pressure with the published monthly BIOCLIMATE water-vapour-pressure product, I obtain large discrepancies that I have not been able to resolve.
I tested two approaches:
Calculating water vapour pressure directly from the ERA5 single-level monthly averaged 2 m dewpoint temperature (d2m).
Calculating water vapour pressure from hourly ERA5 2 m dewpoint temperature (d2m), then aggregating the result to daily means and subsequently to monthly means.
In both cases, the differences from the published BIOCLIMATE monthly water vapour pressure product remain large. For example, using the d2m-based calculation, I obtained the following comparison statistics against the CDS BIOCLIMATE monthly water vapour pressure indicator:
Bias: 478.77 Pa
RMSE: 818.64 Pa
MAE: 564.53 Pa
Correlation: 0.958
The spatial-temporal correlation is relatively high, but the magnitude and range are very different. My calculated values range globally from approximately 0.1 Pa to 4000 Pa, whereas the CDS BIOCLIMATE monthly water vapour pressure indicator ranges from approximately 18 Pa to 1390 Pa.
I calculated water vapour pressure from dewpoint temperature using a standard saturation vapour pressure formula applied to d2m, assuming that the actual vapour pressure can be derived from dewpoint temperature. However, the resulting values are substantially higher than those in the BIOCLIMATE product, especially in low-latitude regions.
Could you please clarify how the monthly water vapour pressure indicator in this dataset was calculated? In particular, I would like to know:
Was the BIOCLIMATE water vapour pressure indicator calculated from ERA5 d2m, or from another variable such as 2 m specific humidity and surface pressure?
What exact formula and temporal aggregation procedure were used to calculate the monthly water vapour pressure indicator?
Is there any additional scaling, bias correction, masking, spatial aggregation, or other post-processing applied to this variable?
Are the published values intended to represent actual near-surface water vapour pressure in Pa?
I would appreciate any guidance on what I may be doing incorrectly or what additional processing steps are required to reproduce the published BIOCLIMATE water vapour pressure product from ERA5 data.
Thank you very much for your help.