Are Brazilian Catchments Gaining or Losing Water? The  Effective Area of Tropical Catchments

Abstract

Similar to most countries, the Brazilian water resources management considers topographically delineated catchment as a territorial unit for policy implementation. Yet, previous studies have shown that catchments are not hydrologically isolated, and topographic limits often neglect the groundwater boundaries. Thus, studies on effective catchment area are promising for shedding light on inter-catchment groundwater flow. Here, we investigated the deviation between the topographic and effective areas across Brazil. We applied the Effective Catchment Area index (ECI) to 733 Brazilian catchments and identified the most influencing attributes on the ECI by using Principal Component and Random Forest Analyses (PCA and RFA, respectively). Further analysis of consistency was carried out by contrasting the ECI values against the expected range of the Budyko curve considering both topographic and effective catchment areas (classic and adjusted framework). Considering the studied catchments, 15% and 16% of their effective areas were respectively smaller than half (strong losing water condition) and larger than double (strong gaining water condition) of their corresponding topographic areas. The aridity index was the main driving factor and negatively correlated with ECI followed by mean slope, precipitation seasonality, and mean elevation. In general, the more arid biomes in Brazil — the Cerrado and Caatinga — are prone to have smaller effective areas while larger effective areas were mostly found in the Atlantic Forest biome, a humid tropical region with a higher mean elevation. We highlight the potential of adopting a pooling of catchments based on their interconnectivity to minimize management costs while maximizing synergies and lessening trade-offs of water transfer processes. Our results contribute to a better country-scale understanding of hydrological connectivity among catchments and highlight the need to consider the effective catchment area to overcome water-food-energy security challenges on multiple scales.

The Impact of an Open Water Balance Assumption on Understanding the Factors Controlling the Long‐Term Streamflow Components

“…For the no‐correction condition (Figure 5, upper panel), we observed higher $${Q}_{B}$$‐relative bias for the Amazon and Caatinga biomes (see also Figure S4 in Supporting Information ), which were pointed out by Schwamback et al. (2022) as two biomes whose catchments exhibit large differences between $${A}_{\mathrm{t}\mathrm{o}\mathrm{p}\mathrm{o}}$$ and $${A}_{\mathrm{e}\mathrm{f}\mathrm{f}}.$$ This can also be seen, with lower magnitude, for the Atlantic Forest catchments, which also exhibit, in general, $${A}_{\mathrm{e}\mathrm{f}\mathrm{f}}/{A}_{\mathrm{t}\mathrm{o}\mathrm{p}\mathrm{o}}$$ ratio diverging from 1. There is a slight predominance (∼60% of the catchments) of positive deviations for the arid Caatinga biome, whose catchments present, in general, a losing‐water condition ($${A}_{\mathrm{e}\mathrm{f}\mathrm{f}}< \,{A}_{\mathrm{t}\mathrm{o}\mathrm{p}\mathrm{o}}$$).…”

“…Similar results were found for $${Q}_{B}$$. Catchments are prone to gain water through IGF with increasing elevation and lose water with increasing area (Schwamback et al., 2022). Indeed, Brazilian's catchments with larger areas are inclined to present an $${A}_{\mathrm{e}\mathrm{f}\mathrm{f}}$$ lower than $${A}_{\mathrm{t}\mathrm{o}\mathrm{p}\mathrm{o}}$$ (median $${A}_{\mathrm{e}\mathrm{f}\mathrm{f}}/{A}_{\mathrm{t}\mathrm{o}\mathrm{p}\mathrm{o}}=0.82$$ for the 100 largest catchments).…”

“…(2020). This approach introduces the concept of $${A}_{\mathrm{e}\mathrm{f}\mathrm{f}}$$, an equivalent area that accounts for IGF in the water balance (Schwamback et al., 2022). The reasoning of this methodology is described as follows: if we assume the existence of IGF, we can define the continuity equation as $$dS/dt=P-\mathrm{E}\mathrm{T}-Q+\mathrm{I}\mathrm{G}\mathrm{F}$$.…”

“…(2020) that considers the existence of IGFs by evaluating the deviations between recharge and discharge rates—as the basic spatial unit to compute hydrologic fluxes and to define possible mechanisms that may affect the catchment hydrological response (Wagener et al., 2007). Nevertheless, there is a strong evidence that catchments are “leaky” and cannot be considered as closed hydrologic entities, as they may lose or gain water through different mechanisms such as IGF at different spatial scales, transferring water beyond their topographic boundaries (Fan, 2019; Liu et al., 2020; Muñoz et al., 2016; Schwamback et al., 2022). In fact, given that hydrological connections between catchments do exist, these cases might be the rule rather than the exception (Le Moine et al., 2007).…”

“…Schwamback et al. (2022) also reported that nearly 32% of a total of 733 Brazilian catchments showed more than 30% of difference between their $${A}_{\mathrm{t}\mathrm{o}\mathrm{p}\mathrm{o}}$$ and $${A}_{\mathrm{e}\mathrm{f}\mathrm{f}}$$, concluding on the need to consider inter‐catchment connectivity in hydrologic studies. Liu et al.…”