Assessing the temporal variance of evapotranspiration considering climate and catchment storage factors

Ruijie Zeng, Ximing Cai

Research output: Contribution to journalArticlepeer-review


Understanding the temporal variance of evapotranspiration (ET) at the catchment scale remains a challenging task, because ET variance results from the complex interactions among climate, soil, vegetation, groundwater and human activities. This study extends the framework for ET variance analysis of Koster and Suarez (1999) by incorporating the water balance and the Budyko hypothesis. ET variance is decomposed into the variance/covariance of precipitation, potential ET, and catchment storage change. The contributions to ET variance from those components are quantified by long-term climate conditions (i.e., precipitation and potential ET) and catchment properties through the Budyko equation. It is found that climate determines ET variance under cool-wet, hot-dry and hot-wet conditions; while both catchment storage change and climate together control ET variance under cool-dry conditions. Thus the major factors of ET variance can be categorized based on the conditions of climate and catchment storage change. To demonstrate the analysis, both the inter- and intra-annul ET variances are assessed in the Murray-Darling Basin, and it is found that the framework corrects the over-estimation of ET variance in the arid basin. This study provides an extended theoretical framework to assess ET temporal variance under the impacts from both climate and storage change at the catchment scale.

Original languageEnglish (US)
Pages (from-to)51-60
Number of pages10
JournalAdvances in Water Resources
StatePublished - May 1 2015


  • Budyko hypothesis
  • Catchment storage change
  • ET variance
  • Temporal variability
  • Water balance

ASJC Scopus subject areas

  • Water Science and Technology


Dive into the research topics of 'Assessing the temporal variance of evapotranspiration considering climate and catchment storage factors'. Together they form a unique fingerprint.

Cite this