Hello!
I am working with the annual water yield model, and I was reading the user guide which says (regarding precipitation):
“Ideally, meters will have at least 10 years of continuous data, without large gaps…”
Why does the guide recommend these 10 years?
Thank you very much for the information!
Hi @bmnoz_2001 -
My understanding is that the 10 year recommendation is based on typical use of the Budyko Curve method that the model is based on, which is made for long-term annual averages. Now, what “long-term” means might be variable. Some Budyko-related studies use 30 years, others 10, and I think we’ve settled on recommending 10 because data tends to become less and less available further back you go. If you only have 5 years of data, then use that. Some users have applied the AWY model with single-year average data to compare across years.
I’m not deeply familiar with the Budyko Curve method, but if you’re looking for a rationale for the long time period recommendation, I’d suggest reading a paper or two about the method. They mostly say “long-term” and each study defines what that means, based on what data they have.
Of course, using longer time periods (like 10 years) smooths out years that are particularly wet or dry, and helps represent a “typical” period of precipitation, ET and yield. But, that variability might be important to your study, in which case perhaps this model isn’t the best one to use. Also, climate change is really messing with what a “typical” year is, and I don’t know exactly how that affects the applicability of this method.
~ Stacie