Effects of Eucalyptus plantations on streamflow in Brazil: Moving beyond the water use debate
Introduction
Forest plantations cover 7.8 million hectares of Brazil (less than 0.9%), and Eucalyptus is the most abundant tree genus in these plantations, covering 5.6 million hectares (IBÁ, 2017). In 2011, 273 million m3 of wood was consumed in Brazil, mainly as fuel (48 million m3), of which 210 million m3 (77%) originated from forest plantations and another 23% originated from native forests (Brazilian Forest Service, 2013). For this reason, encouraging forest plantations is considered one of the main strategies for reducing the deforestation of native forests, especially where forests supply wood for energy as well as for the timber industry (Jürgensen et al., 2014, Payn et al., 2015).
In Brazil, since 1966, forest plantations have been established in marginally productive areas (areas of low agricultural potential), encouraged by 20 years of governmental policies that granted tax incentives (Gonçalves et al., 2013). Currently, the tendency of the forest sector to occupy marginal lands (mostly degraded pastures) continues for economic reasons due to the high demand for wood and fiber in Brazil. As with any forest, forest plantations can provide better conditions for soil processes such as infiltration (van Dijk et al., 2007) and reduced surface runoff, thus avoiding soil erosion and water loss from stands (Wichert et al., 2018). However, studies from different locations in Brazil show that the evapotranspiration of Eucalyptus plantations is higher than that of pasture plantations (Salemi et al., 2013, Almeida et al., 2016, Reichert et al., 2017), and this difference can affect the availability of water resources (Farley et al., 2005, Jackson et al., 2005). Additionally, the fast growing and short rotation (6–7 years) of Eucalyptus plantations in Brazil (Gonçalves et al., 2013) demand water that could be reached by trees at deep soil layers, which could reduce soil water recharge starting in the 2nd year of the rotation (Rodríguez-Suárez et al., 2011, Christina et al., 2016).
It has been widely reported that water is the most important limitation for the increased productivity of Eucalyptus plantations in Brazil (Stape et al., 2004b, Stape et al., 2010), and some studies have quantified the high water use by these plantations (Soares and Almeida, 2001, Stape et al., 2004a, Almeida et al., 2007, Cabral et al., 2010, Christina et al., 2016). However, a global synthesis analysis of streamflow responses to forest management shows the absence of paired catchment studies in Brazil (Brown et al., 2005, Jackson et al., 2005, Evaristo and McDonnell, 2019), which indicates a lack of robust information about the effects of Eucalyptus management on water resources. In addition, productivity has been reduced over multiple Eucalyptus rotations (McMahon and Jackson, 2019), showing the need for a better understanding of the relationship among water limitation, wood productivity and water yield.
Foresters should consider climatological zoning of their lands and strategies to include regional water balance and interannual rainfall variability to plan forestry practices to balance water use by trees (evapotranspiration) relative to the amount of water entering the system (precipitation). This is a way to ensure that the quantity of water produced (streamflow) could be sufficient for human and environmental demands (Lima, 2011). However, the relationship between water balance and local water yield on forest plantation sites has not been studied, and thresholds of water availability need to be understood, and recommendations of forest management change should also be discussed. This paper intends to fill this gap.
To understand forest effects on streamflow and balance water use for other users, experimental catchments are commonly studied (Hibbert, 1965, Bosch and Hewlett, 1982, Andreássian, 2004, Jackson et al., 2005, Evaristo and McDonnell, 2019), where the scale and control of land use allow us to understand the management effects. However, experimental catchments demand high installation costs, and the maintenance of the monitoring system is also costly. This could mainly restrict their application in developing countries such as Brazil. For this reason, partnerships between public and private sectors could be an alternative avenue for collecting, monitoring, and interpreting data, generating valuable information for forestry companies, science and society.
A successful example of cooperation on monitoring the effects of forest plantations was implemented 20 years ago by a partnership among the University of São Paulo (USP), the Forest Science and Research Institute (IPEF) and forestry companies in Brazil. The program for catchment environmental monitoring and modeling (PROMAB), which was built over the last two decades, now encompasses a network of 24 experimental catchments located in different regions of Brazil, monitoring the hydrologic effects of Eucalyptus, Pinus and Tectona plantations, as well as those of native vegetation and grasslands.
In this context, for the first time, we used data from the PROMAB experimental catchments and other studies to synthesize the effects of Eucalyptus forest plantations on water quantity at the catchment scale in Brazil. In the last 30 years, the main discussion was driven by questions such as “How much water does Eucalyptus plantation use?” or “Does Eucalyptus plantation use more water than native forest?” (Lima, 2011), but now, based on a robust dataset, in this paper, we were able to move beyond these questions to further understand the effects of Eucalyptus plantations on streamflow, to identify their climatic limitations and to discuss how to adapt forest management to balance wood production with water provision and related ecosystem services for other users.
Section snippets
Materials and methods
This study used annual data from 120 observations from 19 experimental catchments, 104 observations from 14 experimental catchments of PROMAB, and 16 observations from 6 previous published studies conducted in Brazil (Soares and Almeida, 2001, Facco et al., 2012, Salemi et al., 2013, Almeida et al., 2007, Almeida et al., 2013, Almeida et al., 2016) (Fig. 1, Table 1). The average catchment occupation by Eucalyptus plantation is 71%, ranging from 42% to 96.7% (Table 1); the remainder is occupied
Results
Among 120 observations, the mean annual precipitation was 1412 mm, varying from 643 mm to 3593 mm, and the mean annual evapotranspiration was 1194 mm, varying from 563 mm to 3345 mm (Fig. 2). Considering all annual observations, 50% (60 observations) presented values of ET:P higher than 0.90, and 81% showed evapotranspiration above ET values estimated by the Zhang et al. (2001) curve (Fig. 2), probably because they represent commercial plantations managed for high productivity. On average, the
Water use by plantations
The evapotranspiration found at the 19 catchments covered by Eucalyptus plantation is expected to be higher than that of the catchments covered by pasture, as observed in Brazil by Salemi et al., 2013, Almeida et al., 2016, Reichert et al., 2017. Regarding native vegetation, Salemi et al. (2013) found similar relative annual water yield between paired catchments covered by Eucalyptus plantations and dense montane rainforest (P > 2000 mm). In this study, we do not have data from native
Acknowledgements
Most of the results in this paper are based on work conducted as part of the Cooperative Research Program between IPEF and Brazilian forest companies (PROMAB). This project depends on the participation IPEF and of numerous employees and assistants of the participating companies; we would like to thank them all for their valuable contributions.
The authors thank FAPESP (2013/22679-5 and 2013/13243-9) and CNPq (337395/2015-4 and 307395/2015-4). They are also grateful to the entire staff from the
References (86)
- et al.
Growth and water balance of Eucalyptus grandis hybrid plantations in Brazil during a rotation for pulp production
For. Ecol. Manage.
(2007) Waters and forests: from historical controversy to scientific debate
J. Hydrol.
(2004)The use of the aridity index to assess climate change effect on annual runoff
J. Hydrol.
(2002)- et al.
Landscape context of plantation forests in the conservation of tropical mammals
J. Nat. Conserv.
(2018) - et al.
A review of catchment experiments to determine the effect of vegetation changes on water yield and evapotranspiration
J. Hydrol.
(1982) - et al.
Role of eucalypt and other planted forests in biodiversity conservation and the provision of biodiversity-related ecosystem services
For. Ecol. Manage.
(2013) - et al.
A review of paired catchment studies for determining changes in water yield resulting from alterations in vegetation
J. Hydrol.
(2005) - et al.
The energy and water balance of a Eucalyptus plantation in southeast Brazil
J. Hydrol.
(2010) Forests and water-ensuring forest benefits outweigh water costs
For. Ecol. Manage.
(2007)- et al.
Forest age-induced changes in evapotranspiration and water yield in a eucalypt forest
J. Hydrol.
(2001)