6.4.1 Access to Safe and Affordable Drinking Water and Adequate and Equitable Sanitation and Hygiene – Targets 6.1 and 6.2

As noted in Section 6.3, Target 6.1 is somewhat neutral regarding the forest sector. Forests do not, of themselves, provide safe drinking water or sanitation services. Nevertheless, to the extent that stream waters in forested catchments tend to be of higher quality than in agricultural or urban environments, the treatment costs to bring them to a safe potable level may be lower, with a beneficial effect on affordability. It is common to find catchments maintained with a forest cover to form a source of clean water for a nearby city (e.g. Valdivia, Chile, receives water from a 12.7 km² catchment hosting evergreen native forest). Also, direct extraction of drinking water from streams without treatment is also generally safer in forested catchments than elsewhere (e.g. important for Indigenous populations of tropical forests). The performance of Indicator 6.1.1 (Proportion of population using safely managed drinking water services) could therefore be enhanced by a greater forest cover in water supply catchments. In some regions, forest cover is increasing through natural regeneration following abandonment of agricultural land. In other cases it may be worth deliberately afforesting catchments to provide purer water for water treatment plants, as the cost of treating lower-quality water in the absence of forest cover can be high (hundreds of thousands to millions of US dollars per year for individual cities; Ashagre et al. Reference Ashagre, Platts and Njana2018). That cost would have to be compared with the costs of afforestation (including the potential removal of people from the land and the loss of agriculture). Reduction in run-off, and thus water availability, resulting from the afforestation would also have to be considered (e.g. Target 6.4). A more feasible and cheaper option may be the introduction of forested buffer strips along riparian zones, to reduce or interrupt nutrient fluxes to streams in agricultural catchments and sediment fluxes from both agricultural catchments and forested catchments undergoing logging. In the absence of other pressures, Indicator 6.1.1 is likely to favour maintaining existing forest covers. Given the multiple pressures on land resources, however, it would not be surprising if the forests’ (high) worth to drinking-water quality was simply ignored, leaving the successful achievement of Target 6.1 increasingly dependent on artificial water treatment. The target would then be irrelevant to forests. Target 6.2, with its emphasis on sanitation and hygiene for all, should have a positive impact on forest people – that is, those who live in forests and whose lives and livelihoods depend directly on the forest environment and forest resources. Difficulties of accessibility, though, are likely to mean that forest people in remote areas will be among the last to benefit from this target (although perhaps being among those least in need of it).

6.4.2 Improving Water Quality – Target 6.3

Indicator 6.3.1 (Proportion of wastewater safely treated) is not closely linked to forests. However, to the extent that there is an interest in using treated wastewater for forestry, this indicator may drive an increase in, for example, agroforestry schemes. Forests have themselves been proposed as treatment areas for wastewater as part of wider schemes for induced precipitation recycling (Layton and Ellison Reference Layton and Ellison2016). Indicator 6.3.2 (Proportion of bodies of water with good ambient water quality) is more relevant in view of the potential for land use to affect water quality in rivers, reservoirs, estuaries and downstream wetlands. For example, in streams in native forests in Chile, nitrate (NO₃-N) and ammonium concentrations are very low and nitrogen (N) export is very low (0.2–3.5 kg N ha^-1 yr^-1) (Perakis and Hedin Reference Perakis and Hedin2001, Reference Perakis and Hedin2002). The conversion of native forests to other land uses may therefore be expected to lead to increased NO₃-N and total N export. Conversion of native forest to pasture in southern Chile is associated with increased N fluxes and increased dominance of NO₃-N (exports up to 11 kg ha^-1 yr^-1), although some of this increase may be attributable to pasture fertilisation (Oyarzún and Huber Reference Oyarzún and Huber2003). Conversely, afforestation may change soil pH and alter nutrient cycles (Hong et al. Reference Hong, Piao and Chen2018). Nevertheless, because fertiliser use in plantation forestry – in terms of total use – involves a lower application frequency and smaller land areas compared with agricultural systems (May et al. Reference May, Smethurst and Carlyle2009), the conversion of crop or pastureland to plantations (or even reverting to natural forest) is expected to improve water quality.

In most cases, logging of planted forests may produce elevated sediment export, mainly because of mechanised site preparation, road building and clear-cutting operations in steep terrain, rather than exposure of soil to rainfall (Bathurst and Iroumé Reference Bathurst and Iroumé2014). Research in forest plantations in central Chile found that forest roads produce more sediment than hill slopes; after clear-cutting, the relative contributions increased from 16 per cent to 25 per cent for hill slopes and from 37 per cent to 45 per cent from forest roads (Schuller et al. Reference Schuller, Walling and Iroumé2013). Sediment delivery to streams increases if logging operations take place in rainy periods (with adverse impacts on drinking water and aquatic habitats), but the application of contemporary best management practices, which include guidelines for logging during dry periods, can limit logging-related sedimentation (Bathurst and Iroumé Reference Bathurst and Iroumé2014). Although forest certification has enhanced the adoption of best management practices in Chile, forestry operations are still associated with increased sediment transport and decreases in water quality relative to unmanaged forests.

Because of forests’ generally positive impact on water quality, Indicator 6.3.2 favours maintaining forest cover and increasing cover in protected or degraded catchments. Specifically, achieving good indicator scores for plantation forests will require careful consideration of tree species and the development and implementation of best management practices.

6.4.3 Water Efficiency and Improved Availability – Target 6.4

Indicator 6.4.1 (Change in water-use efficiency over time) tracks the value added (in US dollars) per volume of water withdrawn (cubic metres), by a given economic activity over time (UN Water 2018a). The UN International Standard Industrial Classification of All Economic Activities, Revision 4 code combines forestry with agriculture and fishing. This means that forestry as an economic activity will be considered jointly with agriculture when comparing water use with other sectors. It is already acknowledged that irrigation, as the largest consumer of water by volume, should be one of the big targets for water efficiency (HLPW 2018). There will therefore be increased attention to the efficient use of water in the forestry sector, which may eventually constrain the establishment and continued presence of plantation forests in the water-stressed areas highlighted by Indicator 6.4.2 (Level of water stress: freshwater withdrawal as a proportion of available freshwater resources). This indicator is demand-driven and measures the ratio between withdrawals and the difference between total renewable water resources and the environmental flows (Vanham et al. Reference Vanham, Hoekstra and Wada2018). Strictly, it considers only blue water: the liquid water in rivers, lakes, wetlands and aquifers. However, the amount of blue water is determined by the upstream flows of green water – rainwater held in the unsaturated zone of the soil and available to plants – which is determined by terrestrial ecosystem functions or natural land use (e.g. forests or natural grasslands) and by consumptive water use in rain-fed agriculture and forest plantations. Therefore, analyses of water-stressed environments may lead to closer examinations of the consumptive use of water by forests.

In terms of biomass production per litre of water, trees are considered efficient users of water. Nevertheless, Soto-Schönherr and Iroumé (Reference Soto-Schönherr and Iroumé2014) found in Chile that water-use efficiency (i.e. kilograms of biomass produced per unit of water consumed) differs only a little between forests and grasslands: forests produce 0.1–4 kg of biomass per cubic metre of water, while grasslands produce 0.5–1.3 kg. However, because trees use more water than shorter vegetation, there is a central inconsistency (at least at the catchment scale) between the aims of maintaining forest cover (desirable for many reasons, including lower soil erosion and higher water quality) and of increasing water availability (which implies reducing forest cover). This is less concerning in high-rainfall areas (where there is enough water for all activities) but could be critical in semi-arid areas. This means that forestry as an economic activity will be compared directly with agriculture and other activities when deciding on use of limited water resources. Replacing forest by agriculture could increase annual run-off (and food supply), but at the expense of the forest ecosystem and timber supply. For example, replacement of natural vegetation with agricultural cover in a 175 360 km² catchment in South America produced a significant increase in annual mean discharge and high-flow season discharges because of reduced infiltration and evapotranspiration rates (Costa et al. Reference Costa, Botta and Cardille2003). Conversion of forest cover may ultimately lead to destruction of the land resource itself (Contreras et al. Reference Contreras, Santoni and Jobbágy2013). Thinning of forests considered unnaturally overgrown as a result of fire-suppression programmes has been proposed to increase water supply, e.g. in North America (Poulos Reference Poulos2018), but this may ignore the many other changes produced by forest management (Jones et al. Reference Jones, Achterman and Augustine2009, NRC 2008). For example, forests play important roles in regulating the world’s temperatures and freshwater flows, storing carbon and providing a broad range of important but less recognised benefits (Ellison et al. Reference Ellison, Morris and Locatelli2017). At the subcontinental scale, replacement of forests by shorter vegetation could imply less rainfall in downwind regions (Creed and van Noordwijk Reference Creed and van Noordwijk2018) and possibly therefore less run-off, although the magnitude of this effect remains to be quantified.

Overall, forest plantations with fast-growing species use more water than native forests, although many of the comparisons are limited to old-growth native forests versus young plantations (Soto-Schönherr and Iroumé Reference Soto-Schönherr and Iroumé2016). Results from southern Chile, for a wide range of latitudes and forest compositions and ages, showed that annual interception accounts for approximately 21 per cent of incoming precipitation in the mean, albeit with some margin of variation (Soto-Schönherr and Iroumé Reference Soto-Schönherr and Iroumé2016). Within the range of variation, broad-leaved forests (including native broad-leaved and eucalyptus forests) generally exhibit higher interception losses than conifers. Indicative of the level of uncertainty, Huber et al. (Reference Huber, Iroumé, Mohr and Frene2010) found that interception is lower, and water use is higher, in eucalyptus compared with pine plantations in southern Chile. Because of the relatively limited difference in canopy interception loss between native forests and forest plantations (Huber and Iroumé Reference Huber and Iroumé2001, Soto-Schönherr and Iroumé Reference Soto-Schönherr and Iroumé2016), the observed differences in water yield between the two must be explained, at least in part, by different transpiration losses.

Despite the above, forest plantations may not use more water than native forests at all stages of the forest rotation. As expected, water use is highest (and yield lowest) in the late stages of plantation growth, especially in short-rotation plantations with a high tree density, but water yield (especially summer water yield) increases just after clear-cutting, in the early phases of replanting (Iroumé et al. Reference Iroumé, Huber and Schulz2005, Reference Iroumé, Mayen and Huber2006). Again, variations between plantation species may exist (e.g. pine versus eucalyptus). Thus, water consumption by forests could be moderated by a small amount through careful choice of tree species (with an eye towards suitability to future climates), maintaining a mix of old and new growth (i.e. avoiding large-scale plantation of new growth), regulating tree density and choosing plantation location carefully, possibly implying longer growth periods or reduced timber yield. Nevertheless, such moderation is likely to be small compared with the effect of forest removal. Overall, the demands of Target 6.4 are likely to be inimical towards forest cover in many parts of the world.

6.4.4 Integrated Water Resources Management – Target 6.5

Integrated water resources management (IWRM) is defined as ‘a process which promotes the coordinated development and management of water, land and related resources in order to maximise economic and social welfare in an equitable manner without compromising the sustainability of vital ecosystems’ (Global Water Partnership 2017). It acknowledges the interconnected nature of hydrological resources and the interdependence of different water uses. Within the wider requirements of the definition, IWRM implementation implies the integrated management of water supply, water quality, flood management, navigation, hydroelectric power generation and other water provisions and services within a river basin. Within this balancing act, forest cover reduces water supply relative to other vegetation covers (at the annual scale) but has the potential to improve water quality. From points 8 to 10 in Section 6.2 it can be concluded that forest cover could probably be reduced without necessarily increasing damaging flood magnitudes, but flood frequencies might increase, as would soil erosion. These points refer to the catchment scale. At the subcontinental scale, forest evapotranspiration is not lost from the system but may be recycled as precipitation downwind. Water resources management needs to be integrated with forest management, an end likely encouraged by Indicator 6.5.1 (Degree of integrated water resources management implementation). Such integration is most required in regions where there are multiple and conflicting demands for water, where water supplies are restricted and forests account for significant water use (e.g. semi-arid regions) and where soil erosion is a significant problem (e.g. semi-arid regions, degraded lands and areas of unregulated agricultural expansion). The exact impact of Indicator 6.5.1 on forest cover will vary from basin to basin depending on economic, social, political and other circumstances. It seems more likely, though, that IWRM is implemented as a function of whatever level of forest cover happens to exist, and therefore reacts to rather than drives forest cover change.

In South America conversion of forest to agriculture is the major reason for changes in water availability to rivers and streams (Jones et al. Reference Jones, Almeida and Cisneros2017). The concept of virtual water (Yang and Zehnder Reference Yang and Zehnder2007) assesses water-use efficiency based on water used to grow products traded globally. From 1986 to 2007, South America increased its annual use of water from 42 km³ to 178 km³ and became the continent using the greatest amount of water in food products traded globally, with significant increases in soy exports to China – which in turn has contributed to deforestation in Amazonia (Dalin et al. Reference Dalin, Konar, Hanasaki, Rinaldo and Rodriguez-Iturbe2012).

In the Federal District of Brazil, river basins with substantial expansion of agriculture since the end of the 1970s show a dramatic decrease of base flow discharge by 40–70 per cent – presumably the effect of irrigation extractions more than compensating for the increase in run-off otherwise expected from reduced forest evapotranspiration. Additionally, the run-off ratio is significantly positively related to the cover of natural vegetation (Lorz et al. Reference Lorz, Abbt-Braun and Bakker2012). In south-central Chile, the run-off ratios in four large catchments were positively related to the area of native forest and negatively related to the area of eucalyptus and pine plantations (Lara et al. Reference Lara, Little and Urrutia2009).

Careful choice of tree species and plantation characteristics could moderate water consumption by forests. Reducing the area of forest plantations can potentially increase water availability at the catchment scale, as might replacing exotic fast-growing trees with native forest species, although it has yet to be proven that this would allow adequate timber yields to be produced with less water use. Native forest riparian buffers may increase water yield and improve water quality in forest plantations of eucalyptus in south-central Chile (Little et al. Reference Little, Cuevas, Lara, Pino and Schoenholtz2015) and along rivers in degraded native forest in southeast Brazil (de Souza et al. Reference de Souza, Fonseca, Liborio and Tanaka2013). Forest thinning (Poulos Reference Poulos2018) should be approached with caution: thinned plantations may initially increase water yield (Forrester et al. Reference Forrester, Collopy, Beadle, Warren and Baker2012), but subsequent forest growth might take up the additional water, depending on the timing and degree of thinning (Perry and Jones Reference Perry and Jones2017).

At the international level, Indicator 6.5.2 (Proportion of transboundary basin area with an operational arrangement for water cooperation) may drive a more integrated consideration of the forest environment, especially where a downstream or downwind state feels adversely affected by the forestry activities of an upstream or upwind state. At the subcontinental scales typical of many politically sensitive transboundary basins, the downstream impacts of upstream forestry activities may not always be significant. For example, there is no clear evidence for the so-called Himalayan environmental degradation theory, which proposes that loss of forest cover caused by a rapidly growing population in the Himalayan headwaters of the Ganges and Brahmaputra Rivers increases soil erosion and flood run-off, thereby generating increased flooding and siltation in the delta regions of Bangladesh (Hofer Reference Hofer1993). Downwind impacts at subcontinental scales are increasingly thought to be important but have yet to be proven conclusively (Creed and van Noordwijk Reference Creed and van Noordwijk2018). Large amounts of moisture from evapotranspiration may be recycled within Amazonia and transferred to other parts of South America. However, the extent to which forest conversion has affected this precipitation recycling so far is yet to be clearly quantified (Bagley et al. Reference Bagley, Desai, Harding, Snyder and Foley2014, Khanna et al. Reference Khanna, Medvigy, Fueglistaler and Walko2017, Spera et al. Reference Spera, Galford, Coe, Macedo and Mustard2016, Swann et al. Reference Swann, Longo, Knox, Lee and Moorcroft2015). The impact of Target 6.5 is most likely to be a more integrated consideration of forest management, especially with respect to water resources management, both within and among river basins.

6.4.5 Protecting and Restoring Water-Related Ecosystems – Target 6.6

Indicator 6.6.1 (Change in the extent of water-related ecosystems over time) will drive careful consideration of the relationship between forest cover and the health and sustainability of water-related ecosystems. Although the title of Target 6.6 includes forests along with mountains, wetlands, rivers, aquifers and lakes, Indicator 6.6.1 considers only mangroves. The absence of other types of forests severely limits this indicator’s degree of protection.

Despite the restricted nature of the indicator, forest-related water ecosystems services form an important underlying concept linked to the essence of Target 6.6. They are indeed one of the most proffered links between SDG 6 and forests, but they are not always properly understood and are difficult to quantify. For instance, not all forested areas safeguard water quality: rapid and aggressive afforestation and reforestation with exotic species reduces water availability, affecting its quality, especially in dry seasons (Filoso et al. Reference Filoso, Bezerra, Weiss and Palmer2017). Achieving good indicator scores requires maintenance of water flows – and of good water quality – to wetlands. This could imply reforesting agricultural areas, replacing plantations of exotic tree species and monocultures with a wider range of native trees (which may be slower growing), removing existing plantations or avoiding new plantations (e.g. in peatlands, páramo soils or semi-arid regions) and other scenarios. Such land management could have adverse consequences for livelihoods based on the existing agricultural and plantation activity (and for the production of the associated crops) but might open new employment prospects in forestry activities. Also, there is a high potential for payments for ecosystem services if those services are clearly proven. In the Hindu Kush Himalaya region, research is attempting to show the tangible value of water-related ecosystem services; for example, purification of the downstream water supply via upstream forests is an important ecosystem function that payments for ecosystem services schemes have attempted to quantify and that can contribute to achieving Target 6.3 by natural means (Ashagre et al. Reference Ashagre, Platts and Njana2018). However, there is still a gap in our understanding of how individual attributes (such as changes in land-use patterns) impact ecosystem service flows, including water (Polasky et al. Reference Polasky, Nelson, Pennington and Johnson2011, Nelson et al. Reference Nelson, Kareiva and Ruckelshaus2013, Su and Fu Reference Su and Fu2013). In particular, it is difficult to correlate change in a land unit with change in the volume of ecosystem services that this unit provides (Bhatta et al. Reference Bhatta, Khadgi and Rai2017).

Despite these uncertainties, payments are already being made to promote ecosystem services. Recently, the Chilean government, acting according to the UN Framework Convention on Climate Change (UNFCCC), gave a commitment to afforest (mainly with native tree species) 100 000 ha of degraded soils as a contribution to reducing greenhouse gases (CONAF 2016). To this is added some 470 000 ha that were burnt during the 2016–2017 (southern) summer season. Of this, the government will finance the afforestation of 100 000 ha on lands belonging to small- and medium-size owners, while the remaining burnt lands owned by larger forest companies are already being afforested. The aims of these afforestations are mainly to restore and improve the ecosystem services of the degraded areas, except for the afforestation of lands owned by larger private forest companies whose purpose remains timber production. Not all those forested areas would safeguard water quantity; potentially, there could be resistance to forestry plantations in water-stressed regions, as highlighted by Indicator 6.4.2.

In conclusion, SDG 6 seems unlikely to exert a major influence on forest cover, and indeed is unlikely to be pursued with forests at the forefront of consideration (Table 6.2). Possible exceptions are that Target 6.4 may restrict the spread of new plantations in semi-arid areas and Target 6.5 may drive a more integrated view of catchments and their management. Plantations may be developed with more careful consideration of tree species and plantation characteristics. It seems much more likely, however, that native forest cover in much of the world will continue to decline in the face of pressures greater than SDG 6: (1) to convert forest for food production, driven by population growth and increasing aspirations for living standards worldwide; and (2) to exploit timber and other forest resources, driven first by those seeking a profit but ultimately responding to individual demand globally, with little consideration for the resulting impacts. The decline is exacerbated by the inability or unwillingness of governments in many countries to control such developments, and possibly by climate change reducing or shifting the areas of the world suitable for sustaining the current forests (Guardian 2019, WWF 2019). The extent of monoculture plantations, on the other hand, could increase or decrease: demand for plantation products (e.g. palm oil and timber) is likely to increase, but water efficiency considerations may curtail the spread of plantations in water-stressed areas.

6.5.1 Contextual Factors for SDG 6: The Hindu Kush Himalayas

Trying to understand the real impacts that SDG 6 may have on forests and forest people requires a careful evaluation of the context of water–forest interactions, in particular physical and social settings and the interactions with other SDGs. While context is always affected by local conditions, some situations do recur. The Hindu Kush Himalayas case illustrates some of the contextual factors that must be considered in implementing SDG 6 and highlights the upstream/downstream relationships, inherent where forested mountain areas feed major river systems, which can be found on all continents.

The Hindu Kush Himalayas harbour major river systems providing services, particularly in the form of recharge, to a mountain population of 240 million and a downstream population of 1.9 billion. Indirectly, 3 billion people are dependent on numerous ecosystem services, including climate and hydrological services provided at regional and global scales, and harvested commodities traded at multiple economic scales (Kotru et al. Reference Kotru, Rathore and Pradhan2015). The observed overall increase in forest cover in India does not mean that forest degradation is controlled. An alarming rate of deforestation in parts of the Himalayas, primarily for agricultural land and fuel supply, threatens the sustained flow of forest ecosystem services. As is generally the case in South Asia, multiple sectors and actors influence forests and forest management; it is not only through forest management that the forest–water relationship can be improved for sustained water yield.

Multiple water-related objectives across a portfolio of SDGs present new challenges for policymakers and managers of forests and landscapes with partial tree cover. Hence, SDG 6 cannot be seen in isolation from other key challenges in the Himalayas, such as SDGs 1 (No Poverty), 2 (Zero Hunger) and 5 (Gender Equality). Thus, investments (e.g. in mass tourism) made upstream for addressing other SDGs are not necessarily environmentally friendly or complemented by good governance, making it potentially more difficult to achieve the SDG 6 targets (6.1, 6.3, 6.4 and 6.6). On the other hand, the policies, practices and investments necessary to achieve SDG 6 may not be coordinated with those for other SDGs, so positive outcomes for people are not ensured (Singh and Kotru Reference Singh and Kotru2018). The transboundary nature of hydrological resources – overlaying local, regional and national boundaries – make the challenges to safe water access more complex. New institutional responses are needed to tackle multiple water-related objectives across the full portfolio of SDGs, taking a multiple benefits approach (Creed and van Noordwijk Reference Creed and van Noordwijk2018). A distinction may be made between a first group of SDGs (SDGs 1, 2, 6, 7) implying an increased demand for clean, reliably flowing water, and a second group of SDGs (5, 10, 12, 16) that stresses a change in power-sharing that allows multi-stakeholder involvement, thus increasing the need for transparency and equity in decision-making.

Several socio-economic and governance realities challenge forest regimes in fulfilling their socio-ecological role (as envisaged under SDG 6) in the Hindu Kush Himalayas (Kotru et al. Reference Kotru, Sharma, Sharma and Hofer2017):

• poverty and inequity are still prevalent in South Asia, a water-deficit area;

• institutional capacities and existing policies are inadequate to meet the future challenge of forest management for sustained water yields;

• research on the forest–water relationship is essentially very limited, with no long-term monitoring data or studies available;

• there are very limited cross-sector policy interfaces (e.g. water policy and forest policy interface) that focus on a forest or landscape approach aimed at sustaining water services;

• sustainable forest management is seriously disadvantaged by a lack of proactive management, itself arising from policy deficits;

• data deficits and a lack of harmonised methodologies and data sharing mean that the planning and application of conservation and development strategies contributing to SDG 6 have only a limited foundation on firm data.

Future progress towards achieving Targets 6.3–6.6 in the Hindu Kush Himalayas will require improved upstream–downstream integration, improved transboundary cooperation and greater coordination and simultaneous progress in the implementation of different SDGs: for example, SDGs 1, 2 and 5 (already mentioned) and SDGs 13 (Climate Action), 15 (Life on Land) and 17 (Partnerships for the Goals). Adoption of a landscape approach would allow stakeholder priority interventions to be matched with public and private investments but, equally, there is a need for an improved understanding of the role of forests in influencing ecosystems services at the larger landscape scale. Greater efforts are required to make the communities struggling on the frontline of sustainable forest management more climate resilient.

6.5.2 Implementation of SDG 6

The final impact of SDG 6 on forests and forest people will be determined by the extent of its implementation. There are four important considerations here: finances, institutions, data and other SDGs. First, implementation costs are increasing: the estimated cost of achieving the WASH targets is USD 1.7 trillion (Hutton and Varughese Reference Hutton and Varughese2016). While there are no reliable estimates of the whole cost of achieving SDG 6, it is clear that the required threefold increase for Targets 6.1 and 6.2 alone indicates a huge increase in water targets expenditure. As aid is decreasing, it is not at all clear where this money will come from. The UN calls for more technology transfer and new financing mechanisms, with some based on the recognition of the economic value of water and freshwater ecosystems (UN 2018). Forests may eventually benefit from the growing interest in nature-based solutions, which use or mimic natural processes to enhance water availability and water quality and to reduce risks associated with water-related disasters and climate change. The UN High Level Panel on Water specifically mentions that natural capital solutions, including the ‘water-retaining abilities of forest’, can be used at a fraction of the cost of engineering solutions (HLPW 2018). It labels forests as ‘natural infrastructure’ required to assure future supplies of water, calling for a better alignment of incentives to recognise the value of these services. It is of the utmost importance that natural capital solutions recognise the nuanced role of forests currently accepted as best practice and take into account local conditions. They should be particularly considered in the context of deforestation and forest degradation while recognising the need to understand water quantity effects at catchment, regional and continental scales. It is important to acknowledge that not all water-poor locations have forests to use as improvement tools.

Second, the success of SDG 6 depends on the existence of national and global institutions able and willing to implement the goal. While the WASH sector has spent nearly 20 years trying to achieve global targets, the level of institutional readiness for the new water resources targets is frequently low or non-existent at the country level. Even with the apparently successful MDG 7.C drinking water target, 53 countries were seriously off-target and 19 could not produce data. Good water governance depends on strong formal and informal institutions and the accompanying human resources. There is an acute lack of capacity across most developing countries, particularly in sub-Saharan Africa and South and Southeast Asia (UN 2018). With low institutional capacity, we can expect a slow articulation of the new targets within SDG 6 and, subsequently, low impact on the forest sector. However, since the creation of UN Water in 2003 as a focal point for coordinating efforts of UN entities and international organisations working on water and sanitation issues, the alignment of global water initiatives has increased (UN Water 2018b). Eventually this will lead to actions on the ground. The existence of SDG 6 in itself is a clear proof of the strength of these efforts. The HLPW (High Level Panel on Water 2018) has identified a number of initiatives especially relevant for states trying to implement SDG 6: the World Water Data Initiative; the OECD Water Governance Initiative; the Delta Coalition; High-level Experts and Leaders Panel on Water and Disasters, including an Alliance of Alliances on disaster risk-reduction researches; the initiative on Financing Water Infrastructure convened by the OECD; and the Water Innovation Engine. Whether these will accelerate SDG 6 implementation is yet to be seen, but they may support natural infrastructure projects with a role for forests.

The third consideration is the challenge of having enough good-quality data for monitoring SDG 6. It took a serious global effort from 1990 onwards to develop the WHO/UNICEF Joint Monitoring Programme, now the custodian of global WASH data for Targets 6.1 and 6.2. The situation is more complicated with the other targets as many countries lack the financial, institutional and human resources to acquire and analyse the required data. Fewer than half of UN member states have comparable data available on progress towards meeting the SDG 6 targets (UN 2018). This is important because countries will focus on being able to report to the Integrated Monitoring Initiative (UN Water 2018a). Accordingly, the factors monitored for each target are likely to become the focus of public policy. It is therefore important to understand how forests relate to this monitoring programme.

The final consideration is the dynamic interdependence between SDG 6 and the other SDGs (UN Water 2018c). The majority of these interlinkages are positive and mutually reinforcing. Since the MDG programme, the WASH targets have been identified as critical to: reducing poverty (SDG 1), malnutrition (SDG 2) and diseases (SDG 3); supporting education (SDG 4); and addressing gender (SDG 5) and other inequalities (SDG 10). Moreover, SDG 6 highlights how water of sufficient quality and quantity is required for food production (Target 2.4) and sustainable consumption and production (SDG 12). However, there are some targets – such as doubled agricultural productivity (Target 2.3), energy for all (Target 7.1) and sustained economic growth (Target 8.1) – that potentially could impact negatively on water resources and water ecosystems and, as such, on forests and forest people. IWRM (Target 6.5) is the appropriate framework to balance all these competing needs, for water and forests.

6.5.3 Integration of SDG 6 and Forests

The discussion of the contextual challenges in South Asia clearly shows both the difficulty and the necessity of aligning SDG 6 and forest policies. For this alignment to succeed in an IWRM framework, we need to follow a landscape approach at all levels. As larger forms of vegetation, trees use great amounts of water to produce biomass and for the process of evapotranspiration, more so than many other vegetation types, including crops and grasses. However, when considering Target 6.4 on increasing water-use efficiency and ensuring sustainable withdrawals and supply of freshwater and how it may impact forests and forest management, several issues need to be considered.

Firstly, most forests or tree-based landscapes are naturally occurring, rain-fed systems. Globally, only 7 per cent of forests are planted forests, predominantly found in temperate zones (FAO 2015). Moreover, most of these planted forests are native species: only 20 per cent of planted forests – 1.4 per cent of forests globally – are exotic, and these are located mainly in the southern hemisphere (FAO 2015). It is assumed that natural forests will not be removed for the purposes of achieving Target 6.4 as they provide a wealth of other goods and services, including water-related ecosystem services. Consequently, discussion of water-use efficiency will be limited to planted forests, despite their representing only a small proportion of global forests.

Trees are highly resilient and adaptive organisms that optimise their water use. In other words, they drastically reduce their water consumption in periods of drought and use what they can when water is available (Chaves et al. Reference Chaves, Pereira and Maroco2002). This means that during periods without rain trees can use water stored in the soil; they generally have higher annual rates of water use than shallow-rooted, annual cropping and pasture systems. In high-rainfall areas (> 1500 mm per annum), planted forests can use up to 200 mm more water than pastures, but only if the water use is not energy limited. In low rainfall areas (< 600 mm per annum), forest plantations use an amount of water similar to annual crops and pastures. In intermediate rainfall areas, planted forests potentially use more water than annual crops and pastures. If this is in conflict with other demands for water, policies are required to regulate allocation of water to plantations among other uses. Policy instruments to regulate plantation water can be direct (e.g. a moratorium on land concessions) or indirect (e.g. a market for allocable water). For example, South Africa, Australia and India have implemented policies to regulate or limit plantation establishment (Brown et al. Reference Brown, Zhang, McMahon, Western and Vertessy2005, Dye and Versfeld Reference Dye and Versfeld2007, Farley et al. Reference Farley, Jobbágy and Jackson2005, van Dijk and Keenan Reference van Dijk and Keenan2007, Whitehead and Beadle Reference Whitehead and Beadle2004).

Even if trees are able to optimise their water use, it is important to note that management can be improved in order to further optimise water-use efficiency – including tailoring species selection, as well as thinning and harvesting techniques – to the environmental conditions such as slope, soil type and condition. More importantly, as planted forests are managed at the stand scale and water is managed at a catchment or basin scale, it is critical when planning to look at planted forests, and even tree-based systems more broadly, at the landscape level, taking into consideration the mosaic of land uses and their effects on water. This requires a cross-sectoral approach to land and water planning and management.

Integrated planning and management may require reframing our approach, taking into account integrated solutions such as agroforestry and the use of recycled wastewater in planted forests. Studies show that agroforestry increases water-use efficiency (Bai et al. Reference Bai, Sun and Zheng2016, Droppelmann et al. Reference Droppelmann, Lehmann, Ephrath and Berliner2000). The recycling of treated wastewater for planted forests can reduce competition for water use (particularly in semi-arid and arid areas where water is scarce), reduce the costs associated with water treatment and reduce downstream contamination (FAO 2018a). Planted forests irrigated with treated wastewater in turn improve soil water-storage capacity, reduce soil degradation and erosion, combat desertification in arid areas and provide essential goods that support livelihoods, such as timber, pulpwood and fuelwood (FAO 2018a). According to FAO’s Aquastat database (FAO 2018b), only 52 per cent of the municipal wastewater produced globally is recycled, so there is ample opportunity to explore such options. Egypt, Jordan, Mexico and Spain, among others, are exploring the use of treated wastewater for agroforestry and planted forests. In Jacksonville, North Carolina, USA, upstream forests are being irrigated with treated wastewater, with the forests acting as the final stage of the filtering process and returning water back into the catchment for use downstream (Tew Reference Tew2016).

Much can be gained from a deeper integration of SDG 6 and forest policies. However, this integration must be guided by a shared understanding of the complex relationships between water and forests and their impacts on both forest people and the communities downstream, and possibly downwind.