Large‐Scale Controls of the Surface Water Balance Over Land: Insights From a Systematic Review and Meta‐Analysis

Abstract

The long‐term surface water balance over land is described by the partitioning of precipitation (P) into runoff and evapotranspiration (ET), and is commonly characterized by the ratio ET/P. The ratio between potential evapotranspiration (PET) and P is explicitly considered to be the primary control of ET/P within the Budyko framework, whereas all other controls are often integrated into a single parameter, ω. Although the joint effect of these additional controlling factors of ET/P can be significant, a detailed understanding of them is yet to be achieved. This study therefore introduces a new global data set for the long‐term mean partitioning of P into ET and runoff in 2,733 catchments, which is based on in situ observations and assembled from a systematic examination of peer‐reviewed studies. A total of 26 controls of ET/P that are proposed in the literature are assessed using the new data set. Results reveal that: (i) factors controlling ET/P vary between regions with different climate types; (ii) controls other than PET/P explain at least 35% of the ET/P variance in all regions, and up to ∼90% in arid climates; (iii) among these, climate factors and catchment slope dominate over other landscape characteristics; and (iv) despite the high attention that vegetation‐related indices receive as controls of ET/P, they are found to play a minor and often nonsignificant role. Overall, this study provides a comprehensive picture on factors controlling the partitioning of P, with valuable insights for model development, watershed management, and the assessment of water resources around the globe.

Global Dynamics of Stored Precipitation Water in the Topsoil Layer From Satellite and Reanalysis Data

“…In addition, the Budyko curve can be represented with evapotranspiration ET/P as a function of ϕ (Figure a). In the present research, we used an analytical solution to the Budyko framework following previous research by Padrón et al () to calculate ω parameter. This analytical equation is derived by combining dimensional analysis and mathematical reasoning as follows: …”

“…The ω can be interpreted thus: the higher the ω is, the higher the favorability of the ET condition is. In other words, pixels with the same value of ϕ but with higher ω parameter have land cover conditions that favor ET over runoff (Padrón et al, ). The aridity index ( ϕ ) values and the ω parameters were calculated from GLDAS V2.0 using 63 years of data sets (from 1948 to 2010) on a global scale (Figure b).…”

Opposite Effects of Climate and Land Use Changes on the Annual Water Balance in the Amazon Arc of Deforestation

“…Nevertheless, the sensitivity of annual runoff to forest cover changes tends to attenuate as the watershed size increases in large watersheds, and water‐limited watersheds are more sensitive than energy‐limited watersheds (M. Zhang et al, ). However, Padrón et al () found less influence of land cover and direct human interventions on the spatial variability of the partitioning of precipitation into evapotranspiration than frequently hypothesized in the literature.…”

A Ranking of Hydrological Signatures Based on Their Predictability in Space

“…Aridity is commonly regarded as the main driver of water partitioning at the land surface (Budyko, 1974;Hrachowitz et al, 2013). The influence of climate on hydrological regimes (Berghuijs et al, 2014) and the water balance (Padrón et al, 2017) is well acknowledged, yet it is debated whether this influence is direct, via the water balance, or indirect, via the long-term influence of climate on the landscape (Harman & Troch, 2014). Importantly, climatic variables do not only drive current but probably also trends induced by climate change (Rice et al, 2016).…”