bcgov / climr

An R package for downscaling monthly climate data for North America
https://bcgov.github.io/climr/
16 stars 5 forks source link

potential improvements to CMI #260

Open cmahony opened 6 months ago

cmahony commented 6 months ago

vapour pressure deficit (VPD) is saturated vapour pressure minus actual vapor pressure (es-ea). ea represents es at dewpoint temperature (Tdew)

Hogg (1997) uses a regional simplification for ea in the calculation of vapour pressure deficit.

image image

He validated this approximation for his Canadian prairies study area, but i haven't found a validation for extending this approximation to other regions of Canada/North America.

An incremental but still rough improvement might be to estimate ea from relative humidity ea = es*RH/100.

Here is the difference. note that the RH method gives a lower estimate across the board, even in saskatoon. image

the may-september RH in saskatoon using climr RH is 60%, not the 84% cited by hogg (1997) as the mean nightly maximum RH. I'm not sure why Hogg chose this value as the proxy.

there may be added value in terms of spatial differentiation if we can capture different VPD regimes based on better estimates of RH. something for consideration for future enhancements but not a priority for now.