camiloramirezgo / NWSAS-paper-IOP

This is the repository of the NWSAS paper using the IOP template
1 stars 0 forks source link

Comment 2.1 #59

Closed camiloramirezgo closed 3 years ago

camiloramirezgo commented 3 years ago

It is not clear how this analysis could be done for an average year between 2015 and 2050, since all the input datasets in Table S1 are historical. The authors are accounting for growth in population and irrigated area but ignoring all other possible changes such as climate change. The weather variables are calculated between 1970-2000 and there is no use of future climate projections to account for changes in the weather variables that the authors use. These variables are used to estimate evapotranspiration and we can expect evapotranspiration to increase with temperature, so this is currently an important blind spot of this study. Further, the authors are also making an implicit assumption of a constant recharge rate of the aquifer between 1970-2000 and between 2015-2050. Recharge rate is partially determined by precipitation, which is another variable that is being affected by climate change. I see at least two options:

  1. The experimental setup with the current 2015-2050 time period can remain if the authors convincingly prove that their results are not significantly affected by increases in evapotranspiration driven by higher temperatures and to changes in recharge driven by changes in precipitation expected over the next 30 years.
  2. Otherwise, I would recommend that the authors repeat their analyses for an average year between 1981-2015 (also 35 years) for which they have historical data. Since the science questions that the authors are seeking to answer are not centered around projecting dynamics over the next 30 years, I believe it would be more appropriate to shift this manuscript to a historical analysis that focuses on evaluating the different scenarios under recent climatic and demographic conditions. This historical analysis would arguably have lower uncertainties since there would be no assumptions of if and how each parameter is expected to change in the future.