GTR Wo 100

Research and Development | General Technical Report WO-100 | September 2022
Quantifying the Role of National Forest

System and Other Forested Lands in
Providing Surface Drinking Water Supply
for the Conterminous United States
Ning Liu
G. Rebecca Dobbs
Peter V. Caldwell
Chelcy F. Miniat
Ge Sun
Kai Duan
Stacy A.C. Nelson
Paul V. Bolstad
Christopher P. Carlson
Courtesy photos are all by Chris Evans, University of Illinois, Bugwood.org, except for the one above
(USDA Forest Service photo by Your Forests Your Future) and the one on page 4 (USDA Forest Service
photo by Susan Elliott).
September 2022
https://doi.org/10.2737/WO-GTR-100
Published by Southern Research Station

the Southern Research Station 200 W.T. Weaver Blvd.
for the Washington Office Asheville, NC 28804
General Technical Report WO-100 www.srs.fs.usda.gov
Quantifying the Role of National
Forest System and Other Forested
Lands in Providing Surface
Drinking Water Supply for the
Conterminous United States
Ning Liu
G. Rebecca Dobbs
Peter V. Caldwell
Chelcy F. Miniat
Ge Sun
Kai Duan
Stacy A.C. Nelson
Paul V. Bolstad
Christopher P. Carlson
ABSTRACT
The U.S. Department of Agriculture’s Forest In the West, NFS lands comprised 19.2 percent of
Service manages more than 779,000 km2 (193 the total land area but contributed 46.3 percent of
million acres) of national forests and grasslands the 478.7 billion m3 yr-1 surface water supply; in
(collectively, National Forest System [NFS] lands) the East, NFS lands comprised about 2.8 percent
that play a significant role in providing clean, of the total land area and 3.8 percent (66.6 billion
fresh water for local ecosystems and economies. m3 yr-1) of the surface water supply. In total
This water is sometimes transferred hundreds of across the CONUS, NFS and other forested lands
kilometers away to also serve big cities through comprised 28.7 percent of the total land area but
inter-basin transfers (IBTs). contributed 46.0 percent of the surface water
The contribution of NFS lands to surface supply. Approximately 45.8 million people derived
drinking water supplies for public water systems >10 percent of their surface drinking water supply
has not been assessed at the national scale while from NFS lands, and 22.6 million people received
accounting for IBTs. The Forest Service Water >50 percent of their surface drinking water supply
Supply Stress Index (WaSSI) model was modified from NFS lands. Approximately 125.5 million
to provide estimates of 2001–2015 mean annual people, about 39 percent of the total population
surface water supply and the proportion of in the CONUS in 2017, derived >10 percent of
mean surface water supply originating on 172 their surface drinking water supply from NFS
NFS land units and other forested lands at the and other forested lands, with 83.1 million people
12-digit hydrologic unit code scale across the receiving >50 percent of their surface drinking
conterminous United States (CONUS) while water supply from NFS and other forested lands.
accounting for water transfer through IBTs. In addition to those populations receiving surface
Predictions of the proportion of surface water drinking water supply from their local public
supply originating on NFS and other forested surface drinking water intakes, 12.6 million
lands were linked to specific downstream people were served by public water systems that
communities and populations, using surface purchased surface drinking water supply from
drinking water intake information from the U.S. other public water systems deriving >10 percent
Environmental Protection Agency Safe Drinking of their surface drinking water supply from
Water Information System database of public NFS lands. This study provides a systematic
water systems. accounting of NFS and other forested lands for
A new database of 594 IBTs was compiled for this surface drinking water supply. Our results can aid
study, ranging from 0.01 million m3 yr-1 to 8,900 water resource and forest managers in developing
million m3 yr-1, for a total transferred volume integrated watershed management plans at a time
of 116,894 million m yr . Overall, NFS lands
3 -1 when climate change, population growth, and
comprised 9.2 percent of the total CONUS land land development threaten water supplies.
area but contributed 12.8 percent of the surface Keywords: Drinking water, inter-basin transfers,
water supply. National Forest System, WaSSI, water supply,
water yield.
CONTENTS
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
DEFINITIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Extent and Scale of Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Estimating Surface Water Supply From National Forest
System and Other Forested Lands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Linking Water Yield From Forested Lands to Public Surface
Drinking Water Intakes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Inter-Basin Surface Water Transfers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Water Yield and Surface Water Supply From National Forest
System and Other Forested Lands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Water Transferred by IBTs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Population and Communities Served by Water From Forested Lands . . . . . . . . . . . 19
CASE STUDY: PIKE NATIONAL FOREST . . . . . . . . . . . . . . . . . . . . 25
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Forested Lands as the Dominant Drinking Water Source in Arid Areas . . . . . . . . . 28
Role of IBTs in Distributing Water From Forested Lands to People . . . . . . . . . . . . 29
Purchased Surface Drinking Water Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Dependence on Water From Forests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Population Growth, Climate Change, Water Supply, and Water Quality . . . . . . . . . 31
Comparison to Previous Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
ABOUT THE AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
SUPPLEMENTAL MATERIAL . . . https://doi.org /10.2737/ WO-GTR-100-Sup1

DEFINITIONS
Water yield is the amount of excess water leaving a watershed as streamflow
after accounting for losses that include changes in water storage in the soil,
evaporation, and transpiration from vegetation. In this study, water yield is the
depth to which a watershed (HUC12) would be covered if all of the streamflow
were uniformly distributed over it. The unit of water yield is mm yr -1.
Surface water supply is calculated by accumulating the total volume of water

yield generated over a region, or in the entire river system upstream of a location
of interest along the river network (e.g., watershed outlet) and including surface
water supply from inter-basin transfers, both assuming that water losses to
ground water are negligible. The unit of surface water supply is m3 yr -1.
Surface drinking water supply is the surface water supply available to a given
public water system intake at the location of that intake.
THE ROLE OF NATIONAL FOREST SYSTEM AND OTHER FORESTED LANDS IN PROVIDING SURFACE DRINKING WATER SUPPLY 1
INTRODUCTION
For more than a century, U.S. legislation has favorable conditions of flow. Similarly, the
emphasized the importance of protecting forests Sustained Yield Forest Management Act of 1944
and water resources. Early policymakers in the and the National Forest Management Act of 1976
United States recognized this linkage, writing all sought to safeguard our Nation’s forests and
in the Organic Administration Act of 1897, water resources. In all of this enabling legislation,
“No National Forest shall be established, the Forest Service has been charged with
except to improve and protect the forest within sustaining and improving water resources through
the boundaries, or for the purpose of securing protection, restoration, and enhancement of
favorable conditions of water flows….” Initially, forested landscapes. The Forest Service currently
the U.S. Department of the Interior was manages 779,000 km2 (193 million acres) of public
responsible for identifying and managing these lands in the National Forest System (NFS) across
forests, but in 1905 President Theodore Roosevelt the conterminous United States (CONUS), Alaska,
transferred this responsibility to the USDA Forest and Puerto Rico, which includes 155 national
Service. Later, the Weeks Act of 1911, Clarke- forests, 20 national grasslands, 20 national
McNary Act of 1924, and Bankhead-Jones Farm recreation areas, 1 national tallgrass prairie,
Tenant Act of 1937 provided the Forest Service 6 national monuments, and 6 land utilization
with the authority to acquire lands not in the projects. In addition, the Forest Service cooperates
public domain. All three of these acts had water- with States, other Federal agencies, Tribes, and
related objectives and were based on the original private landowners to sustain the Nation’s other
purposes outlined in the Organic Act: securing forests and grasslands.
2 THE ROLE OF NATIONAL FOREST SYSTEM AND OTHER FORESTED LANDS IN PROVIDING SURFACE DRINKING WATER SUPPLY
In addition to generally establishing national Apart from increasing water demand, population
forests to improve the condition of water resources, growth will also affect the quality of water
the Organic Act specified that, “Waters within supply as forests are converted to developed areas
National Forest boundaries may be used for (Tu 2013). Exurban growth (i.e., the increase
domestic, mining, milling or irrigation purposes, in population and associated land development
as governed by state or federal law,” suggesting near the edges of existing developed land) is the
that Congress was concerned about drinking water most persistent and permanent land use change
supply when setting the framework for managing threatening water quality even at low development
NFS lands. Since then, a century of research has densities (Stein and others 2009). This type of
demonstrated unequivocally that forested lands development increased considerably over the past 4
provide the cleanest and most stable water supply decades, with increasing development beyond the
compared to other land types (Brogna and others suburban fringe (Homer and others 2020, Radeloff
2018, Dudley and Stolton 2003, Fiquepron and and others 2005, Theobald 2005). About one-third
others 2013, Giri and Qiu 2016, Jackson and of the land in the United States is now covered
others 2004, Lockaby and others 2013, Nagy by forests, after declining from an estimated
and others 2012). Forested lands have been 4.14 million km2 (46 percent of total land area)
shown to make significant and disproportionate in 1630 to 3.10 million km2 (33 percent of land
contributions to the total water supply for area) in 2012 (Oswalt and others 2019). Although
downstream communities (Brown and others the total amount of forest area in the CONUS
2008, 2016; Caldwell and others 2014; Creed and has stabilized in recent decades (D’Annunzio
others 2019; Liu and others 2020; Sun and others and others 2015), forested lands are predicted to
2015b; Vose 2019). decline as the population grows (Wear and others
Given the many water-related benefits of forested 2013). Consequently, improving water supplies
lands (Brunette and Germain 2003, Seattle Public has been recognized as one of the important goals
Utilities 2011, Taylor 2018), drinking water in forest and water management (Sun and Vose
utilities are increasingly seeking ways to maintain 2016). Managing forests for water resources under
forested lands to protect water quality and sustain land development pressure and climate change is a
water supply (Warziniack and others 2017). Water major challenge in natural resource management
stress is a growing concern in the United States in the 21st century (Haddeland and others 2014,
(Sun and others 2008) and elsewhere (Gosling National Research Council 2008, Vose 2019).
and Arnell 2016, Vörösmarty and others 2000). Although the basic forest and water relationship
Already common in the arid U.S. West, water is well documented, quantifying forest water
stress is predicted to increase even in the water- resources at a national level is rarely done due to the
rich U.S. South with rapid population growth complexity of climate, hydrologic processes, forest
and climate change (Brown and others 2019). structure, land use/land cover, and water use. Water
Annual total water withdrawal in the United States yield from forests is generated when precipitation
increased from about 300 billion m3 in 1950 to is more than sufficient to meet evapotranspiration
580 billion m3 in 2010 (Dieter and Maupin 2017), (ET) needs, resulting in downstream surface water
coinciding with a doubling of the U.S. population supply that supports ecosystems and people (Sun
(U.S. Census Bureau 2010). Even though per and others 2002). Forest removal commonly results
capita and total water consumption have decreased in short-term increased water yield due to reduced
since 1980 in the CONUS (Dieter and Maupin ET, in proportion to the percentage of the watershed
2017), water demand is projected to grow with cut or forest basal area removed (Andréassian 2004,
population, and water withdrawals are expected Bosch and Hewlett 1982, Sun and others 2005),
to continue to increase (Brown and other 2013), while afforestation generally decreases water yield
except in areas where water supply is already (Andréassian 2004, Farley and others 2005, Filoso
overallocated. and others 2017) due to greater ET rates in forests
compared to other vegetation types.
The spatial distribution of forest and land cover imports approximately 90 percent of its water
and land use change is expected to affect surface from the Catskill and Delaware watersheds
water supply for downstream communities (Greene (NYC Environmental Protection 2017), while
and others 2013). Forest conversion to residential, Los Angeles obtains >90 percent of its water
commercial, and agricultural uses reduces water from multiple sources hundreds of kilometers
quality (Mapulanga and Naito 2019, Moore and away (Ashoori and others 2015). Dickson and
others 2005) and increases runoff and flood risk Dzombak (2017) identified 2,161 IBTs crossing
(Li and others 2020). Privately owned forests are 6-digit hydrologic unit code (HUC6) watershed
increasingly vulnerable to urban development boundaries in the United States using the U.S.
(McNulty and others 2013), which will have Department of the Interior, U.S. Geological Survey
implications for both water quantity and water (USGS) National Hydrography Database (NHD)
quality (Li and others 2020, Martinuzzi and (Moore and Dewald 2016), with about 300 IBTs
others 2014). Decades of research has shown that, driven by city water use. However, the volumes of
in general, the greater the forest coverage in a water transferred by each identified IBT were not
watershed, the higher the water quality (Giri and quantified. Without considering these IBTs, the
Qiu 2016, Tu 2013). magnitude and spatial extent of forest influence on
Forest management, which can differ considerably surface drinking water supplies are misrepresented
based upon forest ownership, can affect surface (Emanuel and others 2015). Clearly, a detailed,
water supply via changes in the magnitude and spatially explicit, national-scale IBT database is
timing of water yield as well as the quality of that needed to accurately assess the contribution of
water (Sun and Vose 2016). Forest ownership water originating on forested lands to surface
patterns differ between the eastern and western drinking water supplies.
regions of the United States, with implications for This study aims to quantify the contribution of
surface water supply and watershed management. NFS and other forested lands to surface drinking
While the Federal Government owns 1.7 million water supply systems in the CONUS. As such,
km2 of forested land in the CONUS, most of it we (the researchers and authors of this report)
(89 percent) lies in the 11 Western contiguous estimated the surface water supply and the origin
States. In contrast, most of the forested land in of that water at each public surface drinking
the Eastern United States is privately owned. For water intake using a water balance model while
example, State and private forests—forests owned accounting for natural water drainage throughout
by State and local governments, corporations, the river network as well as water transferred
families, and other private entities—account through IBTs. Our objectives were to: (1) estimate
for about 90 percent of the total forested land how much surface water supply originated from
area across the South. Federally owned forests NFS and other forested lands across the CONUS;
are managed for multiple uses including timber and (2) estimate how many people and which
production, habitat, and other ecosystem services, communities receive this surface drinking water
while corporately owned forests (26 percent of supply, both with and without IBTs. Results
the total forested land in the South) are generally presented here supersede those presented in
managed to maximize timber production. Caldwell and others (2014) for NFS and other
Natural watershed hydrology is not only affected forested lands in the South and are complementary
by forest management but also by water diversion to those in Liu and others (2020) that provide
and human-made hydraulic structures. Water detailed information on surface drinking water
from forested lands can be transferred to other supply from southern forested lands by ownership
regions through inter-basin transfers (IBTs) to type. This study is the first attempt to evaluate
meet demand where supply is scarce or where benefits of NFS and other forested lands to public
raw water quality is paramount (McDonald surface drinking water intakes across the CONUS
and others 2014). New York City, for example, while accounting for IBTs.
METHODS
Extent and Scale of Analysis each NFS unit and other forested lands (fig. 1).
This study focused on surface water supply The boundaries of the 172 units of NFS lands were
originating on 151 national forests, 20 national derived from the Forest Service Basic Ownership
grasslands, and 1 national recreation area dataset (table 1). Similar to Caldwell and others
(hereafter collectively referred to as NFS land) as (2014), other forested lands not part of NFS lands
well as other forested lands, and how that water were defined by the 2006 National Land Cover
contributes to surface drinking water supply at Database (NLCD) (table 1) and include deciduous,
public surface drinking water intakes across the evergreen, and mixed forest.
CONUS. We focused on surface water supplies The spatial resolution of our analysis was the
because we could not be certain of the origin 12-digit, or sixth-level, hydrologic unit code
(i.e., forested land versus nonforested land) of (HUC12) watershed scale. There are approximately
ground water for any given water supply well at 82,000 HUC12s in the CONUS, with a mean
this large scale. Depending on local factors such area of 100 km2. In addition to the HUC12s in
as well depth, elevation gradients, and aquifer the CONUS, watersheds in Canada and Mexico
characteristics, ground water may originate from that drain to the CONUS were included so that
near where it is withdrawn or from some distance the total flow volumes and proportion of flow
away. We quantified the proportion of surface originating on NFS and other forested lands near
drinking water supply serving a given public international borders could be properly estimated.
surface drinking water intake that originated on The NFS lands include approximately 15,352
Main rivers
States
Water Resource Region
■ NFS lands
■ Other forested lands
Figure 1—National Forest System (NFS) lands, other forested lands, and the land area simulated to estimate total surface
water supply and the amount of that water supply originating on NFS and other forested lands serving public surface
drinking water intakes in the conterminous United States. Labels on the map are Water Resource Regions, including: 1 (New
England), 2 (Mid-Atlantic), 3 (South Atlantic-Gulf), 4 (Great Lakes), 5 (Ohio), 6 (Tennessee), 7 (Upper Mississippi), 8 (Lower
Mississippi), 9 (Souris-Red-Rainy), 10 (Missouri), 11 (Arkansas-White-Red), 12 (Texas-Gulf), 13 (Rio Grande), 14 (Upper
Colorado), 15 (Lower Colorado), 16 (Great Basin), 17 (Pacific Northwest), and 18 (California). Source: Liu and others 2022.
HUC12 watersheds, and there are about 16,162 and others 2012, 2015, 2020; Li and others 2020;
watersheds downstream of those NFS lands. The Schwalm and others 2015; Sun and others 2011a,
relative contribution of NFS and other forested 2011b; Sun and others 2015a). It has been used in
lands to the surface water supply was calculated several regional- and national-scale water resource
for every HUC12 through each river network. assessments (Duan and others 2018, 2019; Li and
others 2020; Lockaby and others 2013; Marion and
Estimating Surface Water Supply others 2013; Sun and others 2015b, 2016; Tavernia
From National Forest System and others 2013) in examining the water-energy
and Other Forested Lands nexus at the national scale (Averyt and others 2011),
in quantifying surface water supplied by national
The surface water supply from each land cover type
forests (Caldwell and others 2014) and State and
in a HUC12 was estimated using the Water Supply
private forest lands (Liu and others 2020) in the
Stress Index (WaSSI) hydrologic model (Caldwell
South, and in studying the impacts of historical
and others 2011, 2012; Sun and others 2011a).
drought on national forests and grasslands (Sun and
The WaSSI model has been tested, validated, and
others 2015b) across the CONUS.
compared to other water balance models (Caldwell
Table 1—Data inputs for the Water Supply Stress Index (WaSSI) model
Data/database Source Resolution Time period

Soil properties State Soil Geographic (STATSGO)-based Sacramento Soil Moisture Accounting 1- x 1-km grid N/A
Model (SAC-SMA) soil parameters
National Oceanic and Atmospheric Administration, National Weather Service,
Office of Hydrologic Development, Hydrology Laboratory
Impervious cover for the National Land Cover Database (NLCD) 2006 Percent Developed Imperviousness 30- x 30-m grid 2006
conterminous United States (CONUS) (https://www.mrlc.gov/data/nlcd-2006-percent-developed-
imperviousness-conus)
Impervious cover for areas Global Man-made Impervious Surface (GMIS) Dataset from Landsat, v1 (2010) 30- x 30-m grid 2010
outside the United States (https://sedac.ciesin.columbia.edu/data/set/ulandsat-gmis-v1/data-download)
National Forest System U.S. Department of Agriculture, Forest Service, Automated Lands Program Parcel 2013
Land Status Record System surface ownership parcels (Basic Ownership
dataset) (https://data.fs.usda.gov/geodata/edw/edw_resources/meta/S_USA.
BasicOwnership.xml)
Land cover National Land Cover Database (NLCD) 2006 (CONUS) (https://www.mrlc.gov/ 30- x 30-m grid 2006
data/nlcd-2006-land-cover-conus)
Monthly mean leaf area index Moderate Resolution Imaging Spectroradiometer (MODIS) (https://modis.gsfc. 1- x 1-km grid 2001–2012
(LAI) by land cover nasa.gov/)
Climate (monthly precipitation PRISM Climate Group (http://www.prism.oregonstate.edu/) 4- x 4-km grid 2001–2015
and temperature) for the
conterminous United States
Climate (monthly precipitation Daymet (https://daac.ornl.gov/cgi-bin/dsviewer.pl?ds_id=1345) 1- x 1-km grid 2001–2015
and temperature) for the HUCs
outside the United States
River network National Hydrography Dataset (NHD) (https://www.usgs.gov/core-science- 1:100,000 N/A
systems/ngp/national-hydrography/national-hydrography-dataset)
Watershed boundaries Watershed Boundary Dataset (WBD) (https://www.usgs.gov/core-science- HUC12 (~90 N/A
systems/ngp/national-hydrography/watershed-boundary-dataset) km2)
HUC = hydrologic unit code.
The WaSSI model is parameterized using readily soil water tension greater than field capacity) that
available national-scale soil, land cover, and interact to generate surface runoff, lateral water
climate data (table 1). All input datasets were movement from the upper soil layer to the stream
spatially rescaled using an area-weighted averaging (interflow), percolation from the upper soil layer
scheme to match the scale of analysis (i.e., to the lower soil layer, and lateral water movement
HUC12 watershed scale). In WaSSI, precipitation from the lower soil layer to the stream (baseflow).
is partitioned into rain and snow using an air- Monthly ET is calculated as a function of potential
temperature-based conceptual snow accumulation ET (Hamon 1963), precipitation, and leaf area
and melt model (McCabe and Wolock 1999). index (LAI) using empirical relationships derived
The WaSSI model calculates monthly infiltration, from multisite eddy covariance measurements
surface runoff, soil moisture, and baseflow for (Sun and others 2011a, 2011b). Storage and ET for
each HUC12 watershed land cover type using impervious cover in each HUC12 are assumed to
algorithms of the Sacramento Soil Moisture be negligible; thus, all precipitation falling on the
Accounting Model (SAC-SMA) (Burnash 1995, impervious portion of a watershed is assumed to
Burnash and others 1973). The soil profile is be runoff and is routed directly to the watershed
divided into a relatively thin upper layer and a outlet. While this assumption may overestimate the
much thicker lower layer that supplies moisture effect of impervious cover on water yield in some
to meet ET demands (Koren and others 2003). locations, it was necessary because we could not
Each layer consists of tension water storage (i.e., be certain about the amount of water storage on
between soil water tensions of field capacity and impervious surfaces or their connectivity to surface
the plant wilting point) and free water storage (i.e., water at the national scale. Water yield is calculated
for each land cover type in a given HUC12 as the other forested lands based on the NLCD
the sum of surface runoff from pervious and do not include woody wetlands or areas with
impervious surfaces, interflow, and baseflow after young trees in early succession or trees stunted
accounting for losses that include changes in water by environmental conditions, which would be
storage in the soil, evaporation, and transpiration classified as shrubland in the NLCD (Homer and
from vegetation. Water yield for each HUC12 is others 2015). We revised the WaSSI flow routing
then calculated as the sum of the area-weighted algorithm to track surface water supply from NFS
averages of water yield of each land cover type lands, NFS and other forested lands, and other
present and expressed in mm yr-1. Water yield for lands through the river network (fig.1) at the
each HUC12 is then routed and accumulated from monthly time step from 2001 through 2015 over
upstream to downstream HUC12s along the river all HUC12 watersheds. The years 2001 and 2015
network to estimate the surface water supply at the were selected because they roughly corresponded
outlet of each respective HUC12. The surface water to the drinking water population-served estimates
supply is the sum of the water yield generated in all and IBT database (discussed below). The mean
HUC12s upstream of a given location on the river annual surface water supply and the fraction of
network and expressed in m3 yr-1. mean annual surface water supply originating on
For this analysis, we overlaid the ca. 2013 NFS NFS and other forested lands were quantified for
land ownership parcels on NFS administrative each HUC12 watershed. In addition to a CONUS-
boundaries, the HUC12 boundaries, the wide assessment, we quantified the surface water
NLCD, and the Moderate Resolution Imaging supply originating on each individual national
Spectroradiometer (MODIS) LAI model inputs forest, national grassland, and national recreation
to make a unique land cover category for NFS area in isolation.
lands. The NFS land ownership parcels differ
from NFS administrative forest boundaries in Linking Water Yield From
that NFS land ownership parcels contain only Forested Lands to Public Surface
those parcels owned by the NFS, whereas the land Drinking Water Intakes
in the NFS administrative boundaries includes The surface water supply originating on NFS and
all lands within the boundary regardless of other forested lands was linked to communities
whether the NFS owns the land. For example, and populations served using the 2017 Quarter
the administrative boundary for the Nantahala 3 version of the U.S. Environmental Protection
National Forest includes the town of Franklin, Agency (EPA) Safe Drinking Water Information
NC, but the NFS does not own the land in the System (SDWIS) database of public water systems
town. We also quantified the contribution of (U.S. EPA 2017), which contains information
other (non-NFS) forested lands to surface water on those water systems such as intake locations,
supply. Similar to Caldwell and others (2014), population served, water source, and system type
forested lands in the NLCD including deciduous, (residential or other). Public water system (PWS)
evergreen, and mixed forest were aggregated to surface drinking water intakes in the SDWIS
represent other forests that are not part of NFS database were screened for obvious locational
lands. We present our water supply results for errors, and only those facilities meeting the
three categories: (1) NFS lands alone, (2) NFS following criteria were included in the analysis:
lands and other forested lands, and (3) other lands (1) facilities associated with a PWS serving a
that are neither NFS lands nor forested land as population of at least 25 people, (2) facilities
described by the NLCD (e.g., crop land, developed associated with a PWS whose primary source
land, etc.). It should be noted that the NFS lands was denoted as “surface water” or “ground
evaluated in this study include both national water under the influence of surface water,” and
forests and national grasslands (though the latter (3) facilities whose facility-level water type was
comprise only 2.3 percent of all NFS lands) as denoted as “surface water” or “ground water
well as one national recreation area. In addition, under the influence of surface water.”
Many PWSs that depend on surface water do underrepresent the proportion of the population
not have public surface drinking water intakes of served for a given intake.
their own but rather purchase their surface water The final database used in this analysis included
supply from other PWSs. We also identified those 8,910 public surface drinking water intakes across
PWSs purchasing surface water supply directly 5,041 PWSs (fig. 2) serving a total population
from a selling PWS that obtained some of their of 137.9 million people (43 percent of the
surface water supply from NFS and other forested approximately 323 million people living in the
lands. We could not quantify the proportion CONUS in 2017 [U.S. EPA 2017]). Most of the
of surface water supply from forested lands for remainder of the CONUS population obtains
these purchasing PWSs because the water volume drinking water supplies from ground water
of the purchases is not available at the scale of sources or purchased surface water supplies from
this study. Instead, we identified those PWSs other PWSs. In total, there were 8,412 PWSs
that purchase water from a selling PWS that that purchase surface water supply from other
receives some portion (i.e., ≥0.01 percent) and PWSs through 12,290 consecutive connection or
>10 percent of their surface drinking water supply nonpiped facilities serving 73.2 million people
from NFS and other forested lands. Criteria for in the CONUS. We overlaid the public surface
inclusion of a purchasing PWS were (1) a selling drinking water intakes on the HUC12 watershed
PWS was identified, (2) the purchasing PWS’s boundaries and assumed that the WaSSI-
primary source was denoted as “surface water” estimated proportion of water from NFS and
or “ground water under the influence of surface other forested lands at the outlet of the HUC12
water,” (3) the purchasing PWS’s facility-level watershed in which a given intake was located
water type was denoted as “surface water” or was representative of the intake location. This
“ground water under the influence of surface assumption might not be accurate for those intakes
water,” and (4) the purchasing PWS did not have located on a tributary and not on the HUC main
their own surface water facilities and thus were stem but was necessary because, like other semi-
not already in our database of PWSs. distributed hydrologic models, WaSSI estimates
The population served in the SDWIS database surface water supply at the outlet of each modeling
is attributed to the PWS as opposed to specific unit (in this case, HUC12 watersheds) in the river
intakes within a system. When calculating network but cannot resolve the amount of water
population served by water from NFS and provided by NFS and other forested lands for
other forested lands, we aggregated the surface specific locations within each modeling unit. In
drinking water supply across all intakes for some cases, intakes were located in coves off the
each PWS by calculating the total available main stem of water supply reservoirs; thus, the
surface drinking water supply and the total proportion of surface water supply from NFS and
surface drinking water supply from NFS and other forested lands on the reservoir main stem
other forested lands across all intakes for the was more representative than that of the inundated
PWS. When intakes were displayed on maps tributary in which the intake was located. We
(e.g., fig. 2), we divided the total population assumed that these intakes were receiving source
served by the PWS equally among the intakes water with the same proportion of water from
for that system. As a result, our representations NFS and other forested lands as that of the first
of population served differ spatially from local HUC12 watershed on the main stem downstream
data in some instances and may overrepresent or of the water supply reservoir.
Population
● <25,000
● 25,001–75,000
● >75,000
Figure 2—The 8,910 public surface drinking water intakes in the study area based on the U.S. Environmental Protection
Agency Safe Drinking Water Information System database of public water systems. Intakes are sized and colored according to
the populations that receive water from these intakes. These intakes collectively serve 137.9 million people or 43 percent of the
total population in the conterminous United States.
Inter-Basin Surface Water Transfers crossed HUC6 boundaries but did not estimate
transfer flow volumes for these potential IBTs.
The most comprehensive national-scale database
For the present study, it was necessary to identify
of IBTs in the CONUS was compiled in the 1980s
source and destination watersheds at a much finer
by the USGS (Mooty and Jeffcoat 1986, Petsch
spatial resolution than HUC6 in some areas due
1985). This database was generated from survey
to the numerous transfers from sometimes small
questionnaires at the State level and considered
watersheds that divert surface water supply from
all IBTs that crossed HUC6 boundaries, while
forested watersheds to municipal drinking water
identifying source and destination basins at the
utilities. In addition, more contemporary IBT
8-digit hydrologic unit code (HUC8) scale. In all,
flow volumes were desired that are reflective of
there were 256 IBTs in this inventory with annual
current surface water supply magnitudes and water
transfer flow volumes based on estimates from
uses. To meet this need, we compiled a new IBT
1973 through 1982. While dated, this database
database for the CONUS informed by previous
has been widely used for national-scale water
inventories and the SDWIS but with updated
resource assessments (e.g., Brown and others
transfer flow volumes and added spatial resolution
2019, Duan and others 2019, Emanuel and others
where needed as described below.
2015). More recently, Dickson and Dzombak
(2017) created a CONUS-wide database of 2,161 The process of building the IBT database began
potential IBT locations at the HUC6 level by with researching anthropogenic water movement
identifying artificial connections in the NHD that in a given area in order to understand the spatial
parameters and interconnectivity of transfers. treatment plant (for municipal and industrial uses)
From this, we determined what volumetric data or the apparent end of irrigation infrastructure
we needed to acquire, after which we approached in the case of agricultural uses. As such, our
the numerous agencies managing the relevant data IBT database is not intended for full water
(table 2). After acquiring data, we processed it as budget work. Irrigation transfers were generally
needed for consistency of units and time, and for included only when prominent and/or integrated
fit into the transfer steps we previously identified. within a complex system that included drinking
Units of measurement in the IBT database are water supplies. Because our main interest was
million m3 for transfer volume and an annual time in drinking water supply and the populations
step, covering the 30 years from 1986 through intertwined with it, outside of the Upper Colorado
2015 whenever possible. Region (WRR 14), we filtered potential systems
In the Upper Colorado Region (Water Resource incorporating IBTs by population served, using a
Region [WRR] 14), where water transfers threshold of 200,000 people served based on the
are ubiquitous and vary greatly in spatial and 2017 EPA SDWIS data on public surface drinking
volumetric scope (Mooty and Jeffcoat 1986, water intakes. Using this approach, some IBTs
Petsch 1985), we recognized an IBT when water that transfer water originating on NFS and other
was made to cross the boundaries of HUC12s. forested lands to public surface drinking water
That is, if surface water was transferred from one intakes may not be included in our database.
HUC12 to another, the transfer was considered However, this level of screening for potential IBTs
an IBT and was included in the database. In the was necessary to balance our investigation effort
remainder of the CONUS, we generally did not with providing a reasonable representation of the
consider transfers as IBTs unless they crossed movement of water from NFS lands by IBTs.
HUC8s because IBTs outside of the Upper We modified the WaSSI model to account
Colorado River Basin tend to draw from larger for surface water transfer through IBTs by
water sources; exceptions occur where important incorporating the transfer from the source to
systems cannot be understood without resorting the destination HUC12 for all IBTs in the flow
to a HUC12 scale for defining IBTs, such as the accumulation calculations. In this way, the surface
Atlanta, GA, area and parts of the California water supply from NFS and other forested lands
Region (WRR 18). In all cases, regardless of of all HUCs affected by a given IBT was updated
IBT definition scale, we identified HUC12s as based on the amount of surface water transferred
the origins and destinations of IBTs. We further through the IBT in the source and destination
defined IBTs as ending when water reached its HUC12 as well as those downstream.
Table 2—Data sources for inter-basin transfer volumes
Agency name Level Accessed via

Alabama Office of Water Resources State Request
Arizona Department of Water Resources State Online report documents
Aurora (CO) Water Municipal Request
California State Water Project State Online report documents
Central Arizona Project Regional Request
Central Arkansas Water (in/serving Little Rock, AR) Municipal Request
Charlotte (NC) Water Municipal Request
City of Sacramento (CA) Municipal Request
Coachella Valley (CA) Water District Regional Online report documents
Colorado Decision Support Systems State Online public data server
Colorado Springs (CO) Utilities Municipal Request
Denver (CO) Water Municipal Online report documents; request
Des Moines (IA) Water Works Municipal Request
East Bay Metropolitan Utilities Department (CA) Regional Request
Georgia Environmental Protection Division State Request
Greater Cincinnati (OH) Water Works Municipal Request
Los Angeles (CA) Department of Water and Power Municipal Request
Massachusetts Water Resources State Request
Metropolitan District Commission (CT) Regional Request
Metropolitan Utility District of Omaha (NE) Municipal Request
Metropolitan Water District of Southern California Regional Request
(in/serving parts of Greater Los Angeles, CA)
New Mexico Office of State Engineer State Online public data server
New York City (NY) Open Records a
Municipal Request
Northern Water (CO) Regional Request
Oklahoma Water Resources Board State Request
Sabine River Authority of Louisiana Regional Request
San Francisco (CA) Public Utilities Commission Municipal Request
South Carolina Department of Health State Request
and Environmental Control
Texas Water Development Board State Request
U.S. Department of the Interior, Bureau of Online public data server; online
Federal
Reclamationb report documents; request
Online public data server; online
U.S. Department of the Interior, U.S. Geological Survey Federal report documents
Utah Division of Water Rights State Online public data server
Virginia Department of Environmental Quality State Request
Wyoming State Engineer’s Office State Online report documents
a
Some New York City data are also from U.S. Geological Survey, including historical reports.
b
Multiple Bureau of Reclamation offices and projects.
RESULTS
Water Yield and Surface Water In comparison with other land cover types, NFS
Supply From National Forest and other forested lands produce disproportionate
System and Other Forested Lands contributions to the surface water supply across
Mean annual (2001–2015) water yield showed the CONUS (table 3 and figs. 3 and 4). National
clear spatial patterns between the 11 Western Forest System and other forested lands comprised
States (West) and the 37 Eastern States (East) 28.7 percent of the total CONUS land area but
(fig. 3A). The predicted mean annual water yield contributed 46.0 percent of the surface water
in the East ranged from <200 mm yr-1 in the Great supply. Alone, NFS lands comprised 9.2 percent
Plains to >1,000 mm yr-1 in the high-elevation of the total CONUS land area and provided 12.8
Appalachian Mountains; in the West, it was percent of the surface water supply.
generally <200 mm yr-1, except in the Northwest National Forest System lands are the main water
Coast and a few other high-precipitation areas resource in the West, while other forested lands
(fig. 3A). The annual surface water supply ranged dominate the water supply in the East. In the West,
from zero in watersheds in arid and semi-arid NFS lands accounted for 19.2 percent of the total
areas to about 28.9 billion m3 yr-1 at the outlet of land area and contributed 46.3 percent of the 478.7
the Colorado River in the West and to about 801.4 billion m3 yr-1 surface water supply generated there.
billion m3 yr-1 at the outlet of the Mississippi River In contrast, in the East, >90 percent of forested
in the East (fig. 3B). lands are other forested lands, which provided 35.0
(A)
Water yield
(mm yr-1)
■ 0–200
■ 201–400
■ 401–600
■ 601–800 States
■ 801–1,000 Regions
■ >1,000 Water Resource Region
(B)
Water supply
(million m3 yr -1)
■ 0–250
■ 251–500
■ 501–1,000
■ 1,001–1,500
■ 1,501–2,000
■ 2,001–824,872 Water Resource Region
Figure 3—(A) Estimated 2001–2015 mean annual water yield in mm yr-1 and (B) surface water
supply in millions of m3 yr-1 by 12-digit (sixth-level) hydrologic unit code (HUC12) watersheds.
The HUCs in (B) are colored by the magnitude of available water supply at the watershed
outlet.
(A)
Percentage of water from NFS lands

■
>0–10 51–75
■
11–25 76–100
■
26–50 Water Resource Region

(B)
Percentage of water from NFS and

other forested lands
■
>0–10 ■
51–75
■
11–25 ■
76–100
■
26–50 Water Resource Region
Figure 4—Percentage of the total 2001–2015 mean annual surface water supply that originated
on (A) National Forest System (NFS) lands and (B) NFS and other forested lands by hydrologic
unit code (HUC) watershed streamlines. Surface water supply is the total amount of surface
water available at the outlet of each HUC watershed, including flow accumulated from HUCs
upstream after accounting for inter-basin transfers. Streamlines of HUC12 (12-digit [sixth-
level]) watersheds are colored according to the fraction of total water supply at the watershed
outlet that originated on (A) NFS lands and (B) NFS and other forested lands. Source for (B):
Liu and others 2022.
Table 3—Summary of water supply from National Forest System (NFS) and other
forested lands in the conterminous United States (CONUS)
West a East b CONUS

Total land area (million km2) c 3.1 4.7 7.8
Total land area in NFS lands (million km2) (percentage of the total) 0.6 (19.2%) 0.1 (2.8%) 0.7 (9.2%)
Total land area in NFS and other forested lands (million km2) (percentage of the 0.9 (29.7%) 1.3 (28.1%) 2.2 (28.7%)
total)
Mean total annual water supply (billion m3 yr -1) c 478.7 1,769.4 2,248.1
Mean total annual water supply originating on NFS lands (billion m3 yr -1) 221.4 (46.3%) 66.6 (3.8%) 288.1 (12.8%)
(percentage of the total)
Mean total annual water supply originating on NFS and other forested lands (billion 347.2 (72.5%) 687.3 (38.8%) 1,034.6
m3 yr -1) (percentage of the total) (46.0%)
a
Includes 11 Western U.S. States (Arizona, California, Colorado, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming).
b
Includes 37 Eastern U.S. States (Alabama, Arkansas, Connecticut, Delaware, Florida, Georgia, Illinois, Indiana, Iowa, Kansas, Kentucky, Louisiana, Maine,
Maryland, Massachusetts, Michigan, Minnesota, Mississippi, Missouri, Nebraska, New Hampshire, New Jersey, New York, North Carolina, North Dakota, Ohio,
Oklahoma, Pennsylvania, Rhode Island, South Carolina, South Dakota, Tennessee, Texas, Vermont, Virginia, West Virginia, and Wisconsin).
c
Including water bodies.
percent of the 1,769.4 billion m3 yr-1 surface water billion m3 yr-1). Overall, in the central and western
supply generated in the East. WRRs (9–18), NFS and other forested lands
Forest area distribution and surface water supply contributed a much higher percentage of surface
varied by WRR within the CONUS (WRRs water supply than other lands, while in the eastern
shown in fig. 1). The South Atlantic-Gulf Region WRRs (1–8), NFS land area and the proportion
(WRR 3) (fig. 5A) had the greatest surface water of surface water supply that originated on NFS
supply from all lands with about 333.1 billion m3 lands were closely linked (<5-percent difference
yr-1, followed by the Pacific Northwest (WRR in percentage of NFS forest coverage and surface
17) and Ohio (WRR 5) Regions. However, the water supply).
contribution of NFS and other forested lands
to surface water supply was much higher in the Water Transferred by IBTs
Pacific Northwest Region (WRR 17) than in the There are 594 IBTs in the database we compiled
South Atlantic-Gulf Region (WRR 3) (fig. 5B). for this study, transferring from 0.01 million m3
The Upper Colorado Region (WRR 14) had yr-1 to 8,900 million m3 yr-1, based on average
the highest percentage (83.2 percent) of surface transfer volumes from 2001 through 2015
water supply from NFS and other forested lands, (fig. 6). The IBTs transferred a total water volume
followed by the Pacific Northwest (WRR 17) and of 116,894 million m3 yr-1 over a total distance of
California (WRR 18) Regions. Areas with a high 36,339 km. More than half of those IBTs (386 of
proportion (>40 percent) of surface water supply 594) are transferring water from HUC12s where
from NFS lands are predominantly distributed in >50 percent of their water originated on NFS
the western WRRs (13–18). The Pacific Northwest and other forested lands, with most of those IBTs
Region (WRR 17) had the greatest surface water located in the Western United States (Colorado and
supply from NFS lands with about 133.7 billion California Regions [WRRs 14, 15, and 18]). Those
m3 yr-1, followed by the California (WRR 18) and IBTs generally related to supplying urban areas,
Missouri (WRR 10) Regions. In contrast, the such as Los Angeles, CA; San Francisco, CA;
Upper Colorado Region (WRR 14) had the highest Denver, CO; Las Vegas, NV; and Phoenix, AZ; in
percentage (69.7 percent) of its surface water the West, and New York City, NY; and Atlanta,
supply from NFS lands, which accounts for 18.9 GA; in the East.
percent of the total land area. In the East, NFS
lands comprised about 2.8 percent of the total land Inter-basin transfers moved surface water
area and 3.8 percent of surface water supply (66.6 supply both within and across WRRs. Notably,
1,000
(A) ■ NFS lands
■ NFS and other forested lands 333
Water supply (billion m3 yr -1)
291
257
219 ■ All lands 167
146 168 151 157
134 134 124 133 118
93 106
100 67
81 75
66 60
44 48
39 41 40
30
24 21
20 20
16 17
13 12
10 10
10 8 8 8 9
6 7
5
4 5 4
4 4 4 4
2
1 2
1
18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
90
(B)
80
Percentage of the total
70
60
50
40
30
20
10
0
18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
WEST EAST
■ Land area of NFS lands
■ Water supply from NFS lands
Land area of NFS and other forested lands
Water supply from NFS and other forested lands
Figure 5—(A) Mean annual water supply originating on National Forest System (NFS) lands, NFS and other
forested lands, and all lands in billions of m3 yr-1; and (B) percentage of water supply originating on and land area
of NFS and other forested lands for each Water Resource Region (WRR). The WRRs are shown in figure 1.
multiple IBTs transferred a total of 873.5 million Pass Tunnel, which transfers water from the
m3 yr-1 of surface water supply from the Upper Blue River headwaters into the Missouri Region
Colorado Region (WRR 14) to adjacent WRRs (WRR 10), where it enters the Blue River Pipeline
10, 11, 13, and 16 (fig. 7). Based on 2001–2015 owned by and serving Colorado Springs, CO;
average values, 459.6 million m3 of surface water (b) headwater collection systems on the Blue and
supply was transferred from the Upper Colorado Fraser Rivers, channeling water to the Roberts
Region (WRR 14) to the Missouri Region (WRR and Moffat Tunnels (City of Denver), respectively,
10), with 38.3 percent of that coming from the thence into various South Platte River tributaries
Arapaho National Forest, 18.8 percent from for use downstream in metro Denver, CO; (c)
the White River National Forest, and another the U.S. Department of the Interior, Bureau
21.6 percent from other forested lands. Four of Reclamation’s Windy Gap and Colorado-
major (and several smaller) IBT projects carry Big Thompson Projects, which gather surface
this surface water supply, including: (a) Hoosier water supply from the uppermost reaches of the
Inter-basin surface water transfers by origin and destination HUC12s

(A)
(B)
(C)
Average
(B) transfer (C)
volume
million m3 yr -1
2001–2015
▲
FROM TO
>0–520
521–1,200
1,201–1,800
1,801–8,900
Figure 6—Inter-basin transfers (IBTs) in the (A) conterminous United States, (B) Upper Colorado Region (Water Resource
Region [WRR] 14) and (C) California Region (WRR 18) from 2001 through 2015. Red diamond and green circle symbols
represent mean annual transfer volume (million m3 yr-1) and direction of IBTs between 12-digit (sixth-level) hydrologic unit
code (HUC12) watersheds (red = transferred from HUC12; green = transferred to HUC12). The green symbols are spatially
offset to make red symbols visible where there is spatial coincidence. Source: Liu and others 2022.
NFS and other surface water transferred out of Water Resource Region 14
Average volume, 2001–2015
■ Arapaho NF ■ San Juan NF ■ Other forested lands

■ Ashley NF ■ Uinta NF ■ Other lands
■ Gunnison NF ■ Wasatch NF Volume transferred
■ Medicine Bow NF ■ White River NF to WRR (million m3 yr -1)
Figure 7—Mean annual surface water (million m3 yr-1) transferred out of the Upper Colorado Region (Water Resource Region
[WRR] 14) from National Forest System (NFS) lands, other forested lands, and other lands through inter-basin transfers from
2001 through 2015. Source: Liu and others 2022.
Colorado River and pump it through the Adams (78.3 percent from NFS land), and downstream
Tunnel, after which it is distributed throughout the communities on the Rio Grande.
northern Colorado Front Range, from Boulder to While there are several small water transfers
Fort Collins; and (d) water from the headwaters between the Upper Colorado (WRR 14) and Great
of Little Snake River that is transferred into the Basin (WRR 16) Regions, the two largest transfers
North Platte watershed, as replacement for water are part of two different Bureau of Reclamation
that Cheyenne, WY, extracts from other parts of projects. The first diverts surface water from
the North Platte watershed. the upper reaches of the Duchesne River in the
Most of the surface water supply transferred Wasatch National Forest and channels it through
from the Upper Colorado Region (WRR 14) to the Duchesne Tunnel as part of the Provo River
the Arkansas-White-Red Region (WRR 11) (95.5 Project. The second, called the Strawberry Valley
percent) originated from the White River National Project, involves a complex collection system
Forest, gathered from headwaters of Colorado within the Duchesne watershed, with surface water
River tributary systems on the Eagle River where originating in the Wasatch, Uinta, and Ashley
the collection system in the Homestake Creek National Forests. Collected water is channeled
watershed (the Homestake Project) is co-owned through the Strawberry Tunnel, after which it
by Aurora, CO, and Colorado Springs, while the follows natural flow toward the broader Salt Lake
Bureau of Reclamation’s Fryingpan-Arkansas City, UT, urban area (61.6 percent from NFS land
Project collects from headwaters of Fryingpan in total).
and Roaring Fork Rivers. Homestake Project
water travels through Homestake Tunnel and Population and Communities Served
then flows into the upper reaches of the Arkansas by Water From Forested Lands
River while Fryingpan-Arkansas Project water
From National Forest System lands only—Many
travels through several different tunnels to
people receive surface drinking water supply from
eventually reach the Arkansas River. Water
NFS lands through intakes managed by their PWS.
partly made up of these transfers is extracted
Including surface water supply from IBTs, we
from the Arkansas River downstream by two
found that:
major (and some smaller) systems. At the Otero
Pump Station, water from the river is diverted • Approximately 79.6 million people obtained
for transfer to the Missouri Region (WRR 10), some portion (≥0.01 percent) of their surface
some into South Platte tributaries, later extracted drinking water supply from NFS lands through
by Aurora, and some into Homestake Pipeline to intakes managed by their PWS.
Colorado Springs (back in the Arkansas-White- • Approximately 45.8 million people obtained >10
Red Region [WRR 11]). The other main Arkansas percent of their surface drinking water supply
River extraction tied into these IBT systems takes from NFS lands through intakes managed by
water from the on-river Pueblo Reservoir further their PWS. This represents 14.2 percent of the
downstream and pipes it to Colorado Springs and total CONUS population (323 million people)
nearby communities without entering WRR 10. and 33 percent of the 138 million people served
The main water transferred from the Upper by public surface drinking water intakes in 2017.
Colorado Region (WRR 14) to the Rio Grande • Approximately 22.6 million people received >50
Region (WRR 13) is via the San Juan-Chama percent of surface drinking water supply from
Project (Bureau of Reclamation). Surface water is NFS lands (figs. 8A and 9).
collected from headwaters of the San Juan River,
Without incorporating IBTs, we found that:
some originating in the San Juan National Forest,
and is tunneled to a small Rio Grande tributary. • Approximately 43.8 million people would have
The transferred water serves the Jicarilla Apache derived >10 percent of their surface drinking
near the tunnel outfall, the cities of Albuquerque water supply from NFS lands.
(78.9 percent from NFS land) and Santa Fe, NM
(A)
Percentage of water
from NFS lands
>0–10 Population
11–25 ■
150,000
26–50 ■
100,000
51–75 ■
50,000
76–100 ■
5,000
(B)
Percentage of water from NFS

and other forested lands
>0–10 Population
11–25 ■
150,000
26–50 ■
100,000
51–75 ■
50,000
76–100 ■
5,000
Figure 8—Public surface drinking water intakes where some amount of source water originated on
(A) National Forest System (NFS) lands and (B) NFS and other forested lands. Circles representing
intakes are colored by the percentage of surface drinking water from (A) NFS lands and (B) NFS and
other forested lands after accounting for inter-basin transfers and sized by the population served.
Source for (B): Liu and others 2022.
140
Origin of surface drinking water
■ NFS lands
■ NFS lands including inter-basin tranfers (IBTs)
120
NFS and other forested lands
NFS and other forested lands including IBTs
100
Population served (millions)
80
60
40
20
>90 >80 >70 >60 >50 >40 >30 >20 >10

Percentage of surface drinking water
Figure 9—Cumulative frequency of population served according to percentage of surface drinking
water originating on National Forest System (NFS) lands and NFS and other forested lands.
• Approximately 19.6 million people would have water supply from NFS lands. For example, the
derived >50 percent of their surface drinking City of Calexico, CA, serving 40,211 people,
water supply from NFS lands. received 65.3 percent of its surface drinking water
Of the 2,725 public surface drinking water intakes supply from NFS lands, 100 percent of which is
receiving >10 percent of water from NFS lands, received through IBTs (via the All-American Canal
whether as natural downstream flow or via IBTs, from the Lower Colorado River).
1,531 intakes (56.2 percent) received >50 percent Many PWSs receive water from NFS lands through
of their surface drinking water supply from a mixture of natural downstream flow and IBTs.
NFS lands (fig. 8A). Of the 2,653 public surface Examples include Colorado Springs Utilities,
drinking water intakes located downstream of which serves 424,171 people with 62.5 percent of
NFS lands and receiving >10 percent of surface the surface drinking water supply coming from
drinking water supply from those NFS lands NFS lands overall, including 4.6 percent from
through natural flow, 1,466 (55.3 percent) received several IBTs; and the City of San Diego, CA, which
>50 percent of their surface drinking water supply serves 1.3 million people with 53.3 percent of the
from NFS lands. The City of Asheville, NC, PWS, water coming from NFS lands overall, including
for example, serves a population of 124,300; 29.7 percent from IBTs, notably through the
46.7 percent of this surface water originates on Metropolitan Water District of Southern California
NFS lands upstream of Asheville, with 0 percent and Bureau of Reclamation. In some cases, the
received through IBTs. IBT influx decreases instead of increases the
There were 62 public surface drinking water percentage of water from NFS lands, either because
intakes that obtained all of their NFS-origin the transferred water has a lower portion of NFS
surface drinking water supply through IBTs; 23 surface drinking water supply than the downstream
of these received >50 percent of surface drinking water, or because an outgoing IBT upstream of
the public surface drinking water intakes removes a large portion of their surface drinking water
NFS water and delivers it elsewhere. An example of supply from NFS and other forested lands that
this situation is found in Palmdale Water District, lie in other regions (fig. 10). For example, the Los
CA, which serves 116,183 people with 47.7 percent Angeles-Long Beach-Riverside-San Bernardino,
of the surface drinking water supply coming from CA, area, serving 7.1 million people, received 68.7
NFS lands. However, before IBT water is mixed in, percent of its surface drinking water supply from
the portion of surface drinking water supply from NFS and other forested lands in the Colorado
NFS lands is 83.1 percent. and California Regions (WRRs 14, 15, and 18)
From National Forest System and other forested through several IBTs: the Los Angeles Aqueduct
lands —Many more people receive surface (and Second Los Angeles Aqueduct) owned by
drinking water supply from NFS and other the Los Angeles Department of Water and Power
forested lands through intakes managed by their and carrying water from Mono Lake and Owens
PWS. Including surface water supply from IBTs, River; the Colorado River Aqueduct, owned by the
we found that: Metropolitan Water District of Southern California
and carrying lower Colorado River content; and
• Approximately 136.7 million people obtained both the West Branch and the East Branch of
some portion (≥0.01 percent) of their surface the California Aqueduct (California State Water
drinking water supply from NFS and other Project). Similarly, the Las Vegas-Henderson, NV,
forested lands through intakes managed by area, with 1.6 million people served, received
4,994 PWSs. about 81.5 percent of its surface drinking water
• Approximately 125.5 million people, around 39 supply from NFS and other forested lands in the
percent of the total population in the CONUS Upper Colorado Region (WRR 14) through the
in 2017, derived >10 percent of their surface Griffith Project, drawing water from Lake Mead,
drinking water supply from NFS and other formed by construction of the Hoover Dam (both
forested lands (fig. 9). Bureau of Reclamation projects). The Greater
• Approximately 83.1 million people received >50 Phoenix, AZ, area, with 3.4 million people served,
percent of their surface drinking water supply received 82.0 percent of its surface drinking water
from NFS and other forested lands. supply from NFS and other forested lands in the
Colorado Regions (WRRs 14 and 15) through the
Without incorporating IBTs, we found that: Central Arizona Project and the Salt River Project,
• Approximately 123.7 million people would have both currently private, though the Bureau of
derived >10 percent of their surface drinking Reclamation built some of their infrastructure.
water supply from NFS and other forested lands. Water purchased from other public water
• Approximately 78.0 million people would have systems —In addition to those PWSs and
received >50 percent of their surface drinking populations receiving surface drinking water
water supply from NFS and other forested lands. supply from forested lands through intakes
Of the 8,910 public surface drinking water intakes managed by their PWS, there were numerous
in 5,041 PWSs in the CONUS, 7,891 intakes (88.6 PWSs that purchased surface drinking water
percent) serving 4,621 PWSs received >10 percent supply from another PWS, including:
of their surface drinking water supply from NFS • 1,566 PWSs serving 12.6 million people that
and other forested lands, with 5,073 intakes in purchased surface drinking water supply from
3,107 PWSs receiving >50 percent of their surface another PWS that received >10 percent of their
drinking water supply from NFS and other surface drinking water supply from NFS lands
forested lands (fig. 8B). • 2,660 PWSs serving 24.6 million people that
Inter-basin transfers help redistribute water from purchased surface drinking water supply directly
NFS and other forested lands to people in other from another PWS receiving some portion (≥0.01
locations, especially in dry urban areas in the percent) of their surface drinking water supply
West. With the help of IBTs, many dry cities get from NFS lands after accounting for IBTs
NFS and other surface water supplying key arid cities

Average volume, 2001–2015
Other NFs (<1% each):
Arapaho NF Manti-La Sal NF
Ashley NF Medicine Bow NF Other NFs (<1% each):
Bridger NF Mendocino NF
Cleveland NF Modoc NF Carson NF
Coconino NF Prescott NF Fishlake NF
Dixie NF Routt NF Humboldt NF
Fishlake NF Sequoia NF Kaibab NF
Grand Mesa NF Sitgreaves NF Sitgreaves NF
Gunnison NF Trinity NF Toiyabe NF
Kaibab NF Uinta NF Uinta NF
Los Padres NF Uncompahgre NF
Wasatch NF
Other NFs (<1% each):

Arapaho NF
Ashley NF
Dixie NF
Fishlake NF
Grand Mesa NF
Manti-La Sal NF
Medicine Bow NF
Sitgreaves NF
Uinta NF
Uncompahgre NF
Wasatch NF
■ Angeles NF ■ Eldorado NF ■ Plumas NF ■ Tahoe NF

■ Apache NF ■ Grand Mesa NF ■ Prescott NF ■ Toiyabe NF
■ Arapaho NF ■ Gunnison NF ■ Routt NF ■ Tonto NF
■ Ashley NF ■ Inyo NF ■ San Bernardino NF ■ Uncompahgre NF
■ Bridger NF ■ Kaibab NF ■ San Juan NF ■ Wasatch NF
■ Carson NF ■ Lassen NF ■ Shasta NF ■ White River NF
■ Coconino NF ■ Manti-La Sal NF ■ Sierra NF ■ Other forested lands
■ Dixie NF ■ Medicine Bow NF ■ Stanislaus NF ■ Other lands
Source of surface Major inter-basin ■ Multiple NFs, <1%
water supply (by %) transfers to cities
Figure 10—The contribution of National Forest System (NFS) lands, other forested lands, and other lands to surface water
supply in key cities in the arid Southwestern United States. WRR = Water Resource Region.
• 4,683 PWSs serving 38.3 million people that lands collectively, we evaluated the individual
purchased surface drinking water supply from contribution for each of 151 national forests, 20
another PWS that received >10 percent of their national grasslands, and 1 national recreation
surface drinking water supply from NFS and area. To accomplish this objective, we created
other forested lands unique model input databases, performed the
• 5,329 PWSs serving 41.9 million people that surface water supply simulation with the WaSSI
purchased surface drinking water supply from model including IBTs, and linked the surface
another PWS receiving some portion (≥0.01 water supply outputs to the EPA SDWIS database
percent) of their surface water supply from NFS of public surface drinking water intakes to
and other forested lands after accounting for IBTs estimate the population and communities served
by each individual NFS unit. Results of the
These estimates of the population served by analyses for each national forest, grassland, and
PWSs purchasing surface drinking water supply recreation area are provided in a supplemental
from another PWS receiving >10 percent of their information document that accompanies this
surface drinking water supply from forested lands report (https://doi.org/10.2737/WO-GTR-
may not reflect the actual proportion of surface 100-Sup1) and an online database (https://doi.
drinking water supply from forested lands for these org/10.2737/RDS-2021-0098). In a case study
purchasing PWSs because the water volume of the presented here, we highlight one of the national
purchases is not available at the scale of this study. forests—the Pike National Forest—which, among
From individual National Forest System units—In all national forests, served the largest population
addition to evaluating the contribution of surface receiving >50 percent of its surface drinking water
water supply from all NFS and other forested supply from NFS lands.
CASE STUDY:
PIKE NATIONAL FOREST
We selected the Pike National Forest in Colorado of water originated from the Pike National Forest
to demonstrate how NFS lands contribute to the with 424,171 people served. In addition to those
surface drinking water supply for downstream PWSs receiving surface water supply from the Pike
communities as well as those in other river basins National Forest through their own intakes, there
through IBTs. In total, the Pike National Forest were 206 PWSs serving 1.6 million people that
served the largest population (1.2 million people) purchased surface water from PWSs receiving some
receiving >50 percent of its surface drinking portion of their surface water supply from Pike
water supply from NFS lands among all national National Forest and 47 PWSs serving 0.6 million
forests. The Pike National Forest encompasses people that purchase surface water from PWSs
approximately 4,435 km2 (1.1 million acres) in receiving >10 percent of their surface water from
the Front Range of Colorado (just east of the Pike National Forest. For example, the Denver
Continental Divide) and includes the headwaters Water Board, CO, sells surface water to 37 other
of the Arkansas and South Platte Rivers (in PWSs across the greater Denver area serving an
WRRs 11 and 10, respectively). Water from additional 0.57 million people.
the Pike National Forest makes its way east to In addition, 25 PWSs received some of their
communities in Nebraska, Missouri, Kansas, surface drinking water supply from the Pike
Oklahoma, Tennessee, and Arkansas, and down National Forest through IBTs. For example, one
the Mississippi River to Louisiana (fig. 11). of four public surface drinking water intakes of
Downstream of the Pike National Forest, we the Denver Water Board received 15.3 percent of
estimated that 53 PWSs receive >10 percent of its surface drinking water supply from the Pike
their surface drinking water supply from the Pike National Forest through IBTs. Two of 23 intakes
National Forest and serve approximately 1.74 of the City of Aurora, CO, together received
million people. Around 1.2 million people receive an average of 45.9 percent of their surface
>50 percent of their surface drinking water supply drinking water supply from the Pike National
from the Pike National Forest, mainly from public Forest through IBTs, with 325,000 people served
surface drinking water intakes located in close over the entire PWS (fig. 11). In both of these
proximity to the Forest. For example, two of four examples, the transferred water is withdrawn
intakes of the Denver Water Board, CO, each from Strontia Springs Reservoir, which received
received about 67 percent of their surface drinking both natural downstream flow of South Platte
water supply from the Pike National Forest; over River tributaries within Pike National Forest
the entire PWS, 56.0 percent of water originated and water transferred into the South Platte
from the Pike National Forest with 1 million people system from the Arkansas-White-Red and Upper
served (figs. 11 and 12). Thirteen of 26 intakes of Colorado Regions (WRRs 11 and 14), much of
Colorado Springs Utilities, CO, received >50 percent it from the White River National Forest that
of their surface drinking water supply from Pike lies across the Continental Divide from the Pike
National Forest; over the entire PWS, 8.9 percent National Forest.
Pike National Forest in Colorado

Public water system intakes receiving surface drinking water from Pike National Forest
One intake of Denver Water Board, CO,

obtained 15% of water from Pike NF
through IBTs
Two intakes of Denver Water Board, CO,
obtained 67% of water from Pike NF
Two intakes of City of Aurora, CO,

obtained 46% of water
from Pike NF through IBTs
Streamlines Intakes Population

■
>0–10 ● >0–10 <5,000
■
11–25 ● 11–25 5,000–50,000
■
26–50 ● 26–50
■
51–75 ● 51–75 50,000–150,000
■
76–101 ● 76–100 >150,000
Others
Cities ■ Pike National Forest (NF)
■
Streamlines Water Resource Region

IBT connections Intakes affected by IBTs
Figure 11—Case study of public surface drinking water intakes receiving surface water from the Pike National Forest, CO.
IBT = inter-basin transfer.
SUMMARY OF PUBLIC WATER SYSTEMS AND POPULATIONS RECEIVING SURFACE DRINKING WATER SUPPLY FROM NATIONAL FOREST SYSTEM LANDS
Supplemental information for General Technical Report WO-100 (https://doi.org/10.2737/WO-GTR-100)
38 REGION 2: ROCKY MOUNTAIN REGION

Pike and San Isabel National Forests
Pike National Forest
~1.1 million acres (4,435 km2) in Colorado
BETWEEN 2001 AND 2015:

● Average surface water supply from Pike National Forest was 604.8 million gallons per day (835.6
million m3 yr -1).
● Fifty-three public water systems (PWSs) serving a total population of approximately 1.7 million
people received >10 percent of their surface drinking water supply from Pike National Forest
whether inter-basin transfers were included or not.
● Twenty-three PWSs serving 1.2 million people received >50 percent of their surface drinking water
supply from Pike National Forest.
● Forty-seven additional PWSs serving 586,580 people purchased surface water from other PWSs
that received >10 percent of their surface water from Pike National Forest.
Cumulative frequency of population served according to the percentage of water coming from
Pike National Forest (not including those served through surface water purchases)
1.5
■ Without including inter-basin transfers (IBTs)
■ Including IBTs
Population served (millions)
1.0
0.5
1.65 million
1.65 million
1.74 million
1.74 million
1.74 million
1.72 million
1.18 million
1.18 million
1.17 million
1.11 million
1.17 million
1,402
1,159
1,159
289
532
289
740
0.0
>90 >80 >70 >60 >50 >40 >30 >20 >10
Surface water supply (percent)
Example public water systems (PWSs) getting surface drinking water

supply from Pike National Forest accounting for inter-basin transfers
Number of Population Percentage of total
PWS ID PWS name intakes served in 2017 surface water
CO0116001 DENVER WATER BOARD 4 1,000,000 56.0
CO0103005 AURORA CITY OF 23 325,000 36.8
CO0101150 THORNTON CITY OF 3 136,977 36.2
CO0118015 CENTENNIAL WSD 7 96,394 57.2
CO0118010 CASTLE ROCK TOWN OF 9 59,362 25.0
CO0103045 ENGLEWOOD CITY OF 3 46,541 50.4
CO0138045 STERLING CITY OF 1 15,100 16.7
CO0118055 ROXBOROUGH PARK WSD 1 10,622 57.5
CO0121950 WOODMOOR WSD 3 8,741 51.3
CO0160900 WOODLAND PARK CITY OF 2 8,500 29.9
Source: Liu, N.; Dobbs, G.R.; Caldwell, P.V. [and others]. 2022. Public water systems and populations receiving surface drinking water supply from National
Forest System lands. 2d ed. Fort Collins, CO: Forest Service Research Data Archive. https://doi.org/10.2737/RDS-2021-0098-2.
Figure 12—Summary of public water systems and populations receiving surface drinking water supply from the
Pike National Forest. This summary and others for all National Forest System units can be found in a supplemental
information document that accompanies this report.
DISCUSSION
This is the first national-scale study to account Forested Lands as the Dominant
for IBTs in the linkage between water originating Drinking Water Source in Arid Areas
on NFS and other forested lands and the PWSs National Forest System and other forested lands
and populations they serve, both through the in the western arid areas tend to provide a much
intakes managed by their local PWSs and through higher proportion of surface water supply than
purchases from other PWSs. We combined a what would be expected from their percentage
hydrologic model, a database of public surface of total land (fig. 5). Although NFS and other
drinking water intakes for PWSs, and an IBT forested lands make up only 28.7 percent of
database to determine the role of NFS and other total land in the CONUS, they provide almost
forested lands in providing surface drinking water half of the total surface water supply (46.0
supply to public surface drinking water intakes percent). This result is very close to previous
in the CONUS. We found that NFS and other studies regarding the amount of water supply
forested lands provide almost half (46.0 percent) originating on forested lands (Brown and other
of the surface water supply, providing some 2016, Luce and others 2017). The higher water
surface drinking water supply for 136.7 million yield in forested lands than other lands resulted
people—nearly half of the total population. More in this disproportionate contribution of forested
importantly, many people get the majority of their lands on surface water supply in the arid areas.
surface drinking water supply from NFS and other However, this does not mean that forests use less
forested lands with the help of IBTs. water than nonforests; on the contrary, water use
by forests through evapotranspiration is generally Role of IBTs in Distributing Water

much higher than on nonforested lands (Zhang From Forested Lands to People
and others 2016). The large difference in water For people who obtain the majority (>50 percent)
yield between forested and other lands in arid of their surface drinking water supply from NFS
areas (fig. 3A) was a result of the difference in and other forested lands across the CONUS,
precipitation amounts across the regions (Daly 19.4 million, or 6.0 percent of total population,
and others 1997). The annual precipitation was obtained some of this water through IBTs. Inter-
972 ± 678 mm yr-1 and 358 ± 275 mm yr-1 for basin transfers are critical in the arid West where
forested lands and nonforested lands in the West, 15.2 million people obtained some of their surface
respectively. On the one hand, forests can only drinking water supply through IBTs, which
exist in relatively high-precipitation areas; on the represents 52 percent of the total population in
other hand, forested lands enhance precipitation the West who obtain the majority of their water
by maintaining atmospheric moisture (Spracklen from NFS and other forested lands. Despite the
and others 2012). In contrast, in the relatively wet recognition that forests are critical for reliable
Eastern United States, the contribution of NFS and high-quality water supplies for downstream
and other forested lands to surface water supply communities (Creed and van Noordwijk 2018),
closely reflected their land area (fig. 4) (Liu and this study demonstrates the role of forested lands
others 2020) because precipitation and water in surface drinking water supply for people
yield were similar across land cover types living outside of forested basins. Although the
(fig. 3A), and most of this region is not water- contributions of surface drinking water supply
limited (Renner and Bernhofer 2012). from forests to urban areas through certain
A much higher percentage of people living in water resources management programs have
the West get the majority (>50 percent) of their been reported previously, such as by New York
surface drinking water supply from NFS and City (NYC Environmental Protection 2017), Los
other forested lands than do people living in Angeles (Ashoori and others 2015), and Seattle
the East (fig. 8). Our results showed that about (Seattle Public Utilities 2011), no previous national-
29.5 million people, or 39.3 percent of the total scale studies were able to link each NFS unit to
population in the West, get >50 percent of their the population served and accurately represent
surface drinking water supply from NFS and the water transfers from these NFS lands across
other forested lands, while about 53.6 million WRRs. Therefore, the importance of forested
people, or 21.6 percent of the total population lands to surface drinking water supply would have
in the East, get >50 percent of their surface been underestimated by previous studies (Barnes
drinking water supply from NFS and other and others 2009, Caldwell and others 2014, Liu
forested lands. Previously, Liu and others (2020) and others 2020). Our IBT database can help fill
reported that around 25.3 percent of people in this gap in future hydrologic modelling to more
the 13 Southern States get >50 percent of their accurately account for the human-mediated water
surface drinking water supply from non-Federal transfers between basins, which have generally
forested lands, which make up 90 percent of the been ignored in past studies because of data
total forested land area. Similarly, Caldwell and limitations (Dickson and others 2020).
others (2014) found that around 20 percent of
people get >50 percent of their surface drinking Purchased Surface Drinking
water supply from non-Federal forested lands Water Supply
in the South. While there are fewer IBTs in the
Eastern United States, neglecting their role in Many PWSs that depend on surface water do not
water delivery to water utilities could result have public surface drinking water intakes of their
in an underestimation of the contribution of own but rather purchase their surface drinking
forested lands to surface drinking water supply water supply from other PWSs. Connections
in those studies. between purchasing and selling PWSs can be
complex. For example, some PWSs that sell
surface drinking water supply to other PWSs this assumption may be uncertain and could lead
also purchase water from one or more other to an overestimation of the population served in
PWSs themselves. Further, the volumes of surface some cases; the estimated proportion of the total
drinking water purchases between sellers and surface water supply originating on forested land
purchasers are not available at the national scale. that is available to these purchasing PWSs may not
Accurately quantifying the proportion of surface reflect the actual proportion because the volume of
drinking water supply from forested lands for these water purchased from various PWSs is not known.
purchasing PWSs requires knowledge of purchase On the other hand, we could reasonably estimate
volumes, their other sources of surface drinking the total population receiving some portion (≥0.01
water supply and the corresponding volumes, and percent) of their surface drinking water supply
all of the connections between PWSs. This level of from forested lands under this assumption because
analysis was beyond the scope of the current study; this approach only requires knowledge of which
however, we did identify those PWSs that purchase PWSs are connected through these water purchases
surface drinking water supply from selling PWSs but does not require knowledge of the volume of
that obtained some portion (≥0.01 percent) and water purchased. Under this assumption, a total of
>10 percent of their surface drinking water supply 104.2 million people would obtain some of their
from forested lands, along with populations surface drinking water supply from NFS lands,
served. In generating these estimates, we included and 178.6 million people would obtain some of
the first water transaction among purchasing and their surface drinking water supply from NFS and
selling PWSs but did not account for additional other forested lands.
transactions when selling PWSs also purchase
water from other PWSs. As a result, our estimates Dependence on Water From Forests
of the PWSs and populations served by surface
water originating on forested lands through water Various estimates of populations that use surface
purchases may be conservative. water supply from forested lands have been
developed over the years and are used widely in
If it were assumed that purchasing PWSs serve publicly available literature today. For example,
surface drinking water supply from forested lands a 2007 Forest Service briefing states, “180
at the same proportion as that obtained by the million people…rely on forested lands to capture
selling PWSs through their intakes, the population and filter their drinking water” (USDA Forest
served by these purchasing PWSs could be added Service 2007). This estimate can be traced back
to the population served by PWSs through the to Stein and others (2005), which references a
intakes they manage that obtain a given proportion personal communication with Dr. James Sedell
of surface drinking water supply from forested in 2005, stating, “According to Forest Service
land. For example, under this assumption, with estimates, some 180 million people depend on
45.8 million people receiving >10 percent of their forests for their drinking water.” A report by
surface drinking water supply from NFS lands Sedell and others (2000) is sometimes referenced
through their PWS intakes and 12.6 million people for information about the population dependent
receiving surface drinking water supply from a on surface drinking water supply originating
PWS that purchases water from another PWS that on NFS lands; this report mentions a 1999 EPA
receives >10 percent of their surface drinking water study that showed about 60 million people live
supply from NFS land, a total of 58.4 million in communities located in watersheds containing
people would obtain >10 percent of their surface NFS lands. These earlier, coarse estimates are
drinking water supply from NFS lands. Similarly, consistent with our estimates of populations
a total of 163.8 million people would obtain >10 receiving any portion (≥0.01 percent) of their
percent of their surface drinking water supply from surface drinking water supply from forests either
NFS and other forested lands. Estimating the total through their public surface drinking water
population getting >10 percent of their surface intakes or through their PWS purchasing from
drinking water supply from forested lands under another PWS, i.e., 104.2 million people receiving
surface drinking water supply from NFS lands that protecting forested lands from development
and approximately 178.6 million people receiving (Brown and others 2019) and maintaining healthy
surface drinking water supply from NFS and other forests can improve water resource sustainability.
forested lands. Clearly, many of these communities Although per capita water use in U.S. cities
receive substantial quantities of surface drinking has been declining steadily over the past few
water supply from other forested and nonforested decades (Dieter and Maupin 2017, Rockaway
lands as well. and others 2011), population growth is expected
In this study, we present results for communities to increase total water demand, especially in
and populations that receive >10 percent of their urban areas (Brown and others 2019, Yigzaw
surface drinking water supply from forested and Hossain 2016). Population growth will also
lands, but we show all public surface drinking indirectly further increase water demand across
water intakes that receive any portion of their the United States by raising the water demand of
water from forested lands to illustrate upstream/ agricultural and landscape irrigation, which may
downstream relationships between forests and be exacerbated by climate change (Brown and
PWSs. Fifty percent of surface drinking water others 2019, Creed and others 2014). Drought
supply originating on forested lands is an upper and warming have already resulted in a dramatic
threshold that highlights communities which reduction of water available to ecosystems and
rely on forested land for most of their water. The to the public across the United States (Sun and
relative importance of any quantity depends on others 2015b). With continued climate change,
several local factors including the total available larger deficits between water supply and demand
raw surface water supply at an intake location, will likely occur in the central and southern
frequency and severity of drought, population Great Plains States, the Southwestern States and
and per capita domestic water use, water use by Intermountain and Rocky Mountain States,
other sectors, and downstream water rights and and California, and even in the relatively wet
environmental flow requirements. Therefore, it Southeastern States (Brown and others 2019,
would be misleading to suggest that amounts Naumann and others 2018, Sun and Vose 2016),
above or below an arbitrary threshold are suggesting more people may be subject to water
equivalent in terms of public value across all stress (Duan and others 2019, Gosling and
locations. Displaying the population served Arnell 2016). The increase in population and
information as a function of percentage of surface urbanization in some parts of the country will
drinking water supply from forested land (e.g., increase demand for clean water while putting
fig. 9) allows readers to assess importance using more emphasis on minimizing development of
use their own judgement and experience. We existing forested lands (Brown and others 2019).
suggest readers carefully consider what threshold is While some areas located at climate margins that
appropriate for their watershed and PWS. currently support forests may not be able to do
so under climate change (Guo and others 2018),
Population Growth, Climate Change, maintaining natural land cover in these areas
Water Supply, and Water Quality (e.g., grassland, shrubland) could help maintain
downstream water quality.
Our study has shown the importance of NFS and
other forested lands in providing surface drinking Apart from providing a disproportionate amount
water supply for communities during the period of of the Nation’s public surface water supply, larger
2001–2015. However, climate change is expected areas of forest cover also result in higher water
to increase surface temperature and increase quality, thereby lowering the raw water treatment
the frequency and severity of droughts, both of costs for public drinking water and providing
which will likely reduce surface water supply numerous other ecosystem services such as habitat
(Creed and others 2014, Duan and others 2017, for aquatic biota (Abildtrup and others 2013,
Sun and others 2015b). The high contribution of Lopes and others 2019, Warziniack and others
surface water supply from forested lands suggests 2017). Watersheds with more forest cover tend
Table 4—Comparison among studies that quantified the contribution of National Forest
System (NFS) and other forested lands to water supply in the conterminous United States
Brown Brown Luce

and others 2008 and others 2016 a and others 2017 This study
Hydrologic model Advection-Aridity model VIC VIC WaSSI
(Brutsaert and Stricker 1979) (Mahat and others 2017) (Livneh and others 2013) (Caldwell and others 2012,
Sun and others 2011b)
Zhang model
(Zhang and others 2001)
Model spatial resolution 5 x 5 km ~12 x 12 km ~6 x 6 km HUC12 (~100 km2)
Model temporal resolution Monthly/mean annual Daily Daily Monthly
Product temporal resolution Mean annual Mean annual Mean annual + mean Mean annual
summer
Time period 1953–1994 1981–2010 1915–2011 2001–2015
NFS boundaries Proclamation Ownership and Proclamation Ownership

proclamation
Total land area 7,691 7,700 NR 7,776
(thousand km2)
NFS land area (thousand km2) 846 (11%) 693 (9%) 846 (11%) 719 (9%)
(percentage of total)
All forested land area 2,230 (29%) 1,987 (26%) NR 2,235 (29%)
(thousand km2)
Total water supply (billion m3) 1,769 1,922 NR 2,248
NFS water supply (billion m3) 326 (18%) 280 (15%) NR (18%) 288 (13%)
All forested land water supply 931 (53%) 884 (46%) NR 1,035 (46%)
(billion m3) (percentage of total)
All forested land representationb 1992 NLCD classes 41, 42, 43 2006 NLCD classes 41, NR NFS ownership parcels
42, 43 and remaining 2006 NLCD
classes 41, 42, 43
HUC = hydrologic unit code; NLCD = National Land Cover Database; NR = not reported; VIC = Variable Infiltration Capacity; WaSSI = Water Supply Stress Index.
a
NFS land area and water yield data shown are based on estimates for NFS ownership parcels.
b
NLCD land cover classes 41, 42, and 43 include deciduous, evergreen, and mixed forest, respectively.
to have lower concentrations of nutrients and 2001). By 2060, it is predicted that the population
sediment than watersheds with less forest cover of the United States will be around 500 million
(Swank and others 2001, Tu 2013, Warziniack people (under Shared Socioeconomic Pathway
and others 2017). However, these forest-based Scenarios 5; Riahi and others 2017), which could
ecosystem services are increasingly threatened result in a >35-percent increase of the developed
by land use change (Curtis and others 2018) and land in comparison with 2010 (U.S. EPA
climate change and its cascading direct (e.g., 2020). Moreover, private forests would be most
increase in evaporative potential and drought vulnerable to the development (Liu and others
severity, increased stream temperature) (Hoegh- 2021). Forest conversion to urban use in some
Guldberg and others 2018, Isaak and others 2017) areas might relieve water stress conditions locally
and indirect (tree species shift, wildland fires, by increasing water yield (Suttles and others 2018);
outbreak of insects and diseases, altered flood however, dispersed development on private forested
magnitudes) impacts (Hallema and others 2018, lands might elevate stormflow and flood risk and
Hultine and others 2010). Forest loss could lead degrade water quality through increased sediment
to an increase in sediment and nutrient loads in delivery associated both with development and
streams (Arthur and others 1998, Goode and with a densification of road networks (Stein
others 2012, Riekerk 1985, Swank and others and others 2009). Therefore, protecting existing
forests from fragmentation and addressing other evergreen, and mixed forest, respectively), and
environmental threats become even more critical Brown and others (2016) used the 2006 NLCD
for surface water supplies that depends on privatelyclasses 41, 42, and 43. Our study used the NFS
owned forests. ownership parcels and 2006 NLCD classes 41, 42,
and 43 for remaining lands to represent all forested
Comparison to Previous Studies lands. Due in part to the differences in time period
of land cover data, Brown and others (2016)
Several studies have estimated water supply and reported lower forested land area (26 percent)
the proportion of water supply originating on than Brown and others (2008) (29 percent), which
NFS and other forested lands across the CONUS contributed to a lower proportion of water supply
(table 4). These studies differ in their modeling from forested land (46 percent versus 53 percent,
approaches, spatial and temporal resolution, respectively). Similarly, because our study used
modeling time period, and representation of NFS NFS lands, including both national forests and
and other forested lands. For example, Brown and national grasslands, as part of all forested land,
others (2008) used a multimodeling approach to we report higher forested land area (29 percent)
predict mean annual water supply over the 1953– than Brown and others (2016) (26 percent) yet
1994 time period, land cover based on the 1992 similar water supply from all forested lands (46
NLCD, and NFS lands based on proclamation percent). In addition to differences in forested
boundaries. Our study used the monthly WaSSI area, differences in modeling time period will
hydrologic model over the 2001–2015 time period, affect water supply predictions due to differences
land cover based on the 2006 NLCD, and NFS in climate over those time periods. For example,
lands based on ownership parcels. The NFS precipitation increased approximately 4 percent
ownership parcels differ from NFS proclamation from 1901–2015 across the CONUS and was
boundaries in that NFS ownership parcels contain generally greater in the East and lower in the West
only those parcels owned by the NFS, whereas from 1986 through 2015 compared to 1901–1960
the NFS proclamation boundaries include all (Easterling and others 2017). These differences in
lands regardless of whether NFS owns the land. precipitation over time may partly explain why
As a result of differences in NFS boundaries used, our study and Brown and others (2016) predicted
Brown and others (2008) and Luce and others greater total water supply than Brown and others
(2017) report larger percentages of NFS land in (2008). Despite differences in predicted magnitudes
the CONUS (11 percent) than in our study and of water supply, the proportions of water supply
Brown and others (2016) (9 percent), as well as from NFS lands and other forested lands are very
larger proportions of water supply from NFS lands consistent across studies when considering the
(18 percent) compared to Brown and others (2016) differences in representation of forested land area
(15 percent) and our study (13 percent). Similarly, discussed above. In all, results of this study are
differences in the representation of all forested consistent with previous work when considering
lands resulted in differences in reported land area differences in modeling approaches, spatial and
and water supply from all forests. To represent all temporal resolution, modeling time period, and
forests, Brown and others (2008) used 1992 NLCD representation of NFS and other forested lands.
land cover classes 41, 42, and 43 (deciduous,
SUMMARY
Overall, NFS lands comprise 9.2 percent of the total CONUS land area but
contributed 12.8 percent of the total surface water supply. When incorporating
IBTs, approximately 45.8 million people obtained >10 percent of their surface
drinking water supply from NFS lands, and 22.6 million people received >50
percent of surface drinking water supply from NFS lands. In addition to those
populations receiving surface drinking water supply from their local public surface
drinking water intakes, 12.6 million people were served by PWSs that purchased
surface drinking water from other PWSs deriving >10 percent of their surface
drinking water supply from NFS lands.
Although NFS and other forested lands make up 28.7 percent of the total
CONUS land area, they provide 46.0 percent of the surface water supply. About
125.5 million people, or 38.9 percent of the total population in the CONUS in
2017, derived >10 percent of their surface drinking water supply from NFS and
other forested lands from the intakes managed by their PWSs, including those
receiving water through IBTs. Around 83.1 million people, or 25.7 percent of
the total population in the CONUS, receive the majority (>50 percent) of their
surface drinking water supply from NFS and other forested lands. In addition to
those populations receiving surface drinking water supply from their local public
surface drinking water intakes, 38.3 million people were served by PWSs that
purchased surface drinking water from other PWSs deriving >10 percent of their
surface drinking water supply from NFS and other forested lands.
NFS and other forested lands are the dominant surface water supply source in
the West. Inter-basin transfers played a critical role in providing surface drinking
water supply from NFS and other forested lands to urban areas, especially in
the arid West. Our study developed benchmark high-resolution data for water
supply, identified surface water sources and withdrawal locations for public
surface drinking water supplies, and highlights the water-related benefits of NFS
and other forested lands to downstream communities and people living in other
areas through IBTs.
ACKNOWLEDGMENTS
This study was supported by the U.S. Department of Agriculture (USDA), Forest
Service, Washington Office, Biological and Physical Resources by agreement
number 17-CS-11330140-028 to Dr. Paul Bolstad at the University of Minnesota,
the Southern Group of State Foresters by agreement number 18-JV-11330140-
007 to Dr. Stacy A.C. Nelson at North Carolina State University, and the
Forest Service Southern Research Station. Any opinions, findings, conclusions,
or recommendations expressed in the material are those of the authors and do
not necessarily reflect the views of the USDA, North Carolina State University,
or University of Minnesota. We also want to thank Dr. Travis Warziniack and
Brian Staab for helpful comments on previous versions of this manuscript.
ABOUT THE AUTHORS

Ning Liu was formerly a Postdoctoral Research Fellow, Chelcy F. Miniat was formerly a Project Leader and Research
U.S. Department of Agriculture, Forest Service, Southern Ecologist, U.S. Department of Agriculture, Forest Service,
Research Station, Coweeta Hydrologic Lab, Otto, NC Southern Research Station, Center for Forest Watershed
28763, and University of Minnesota, College of Natural Research, Otto, NC 28763; currently a Program Manager,
Resources, Department of Forest Resources, St. Paul, MN U.S. Department of Agriculture, Forest Service, Rocky
55108; currently an Oak Ridge Institute for Science and Mountain Research Station, Maintaining Resilient Dryland
Education (ORISE) Postdoctoral Fellow, U.S. Department Ecosystems, Albuquerque, NM 87102.
of Agriculture, Forest Service, Southern Research Station, Ge Sun is a Research Hydrologist and Project Leader,
Eastern Forest Environmental Threat Assessment Center, U.S. Department of Agriculture, Forest Service, Southern
Research Triangle Park, NC 27709. Research Station, Eastern Forest Environmental Threat
G. Rebecca Dobbs was formerly a Geospatial Research Assessment Center, Forestry Sciences Laboratory, Research
Associate, U.S. Department of Agriculture, Forest Service, Triangle Park, NC 27709.
Southern Research Station, Coweeta Hydrologic Lab, Kai Duan is an Associate Professor, School of Civil
Otto, NC 28763, and North Carolina State University, Engineering, Sun Yat-Sen University, Guangzhou, China.
College of Natural Resources, Department of Forestry and
Environmental Resources, Center for Geospatial Analytics, Stacy A.C. Nelson is a Professor, North Carolina State
Raleigh, NC 27695; currently an Oak Ridge Institute University, College of Natural Resources, Department
for Science and Education (ORISE) Postdoctoral Fellow, of Forestry and Environmental Resources, Center for
U.S. Department of Agriculture, Forest Service, Southern Geospatial Analytics, Raleigh, NC 27695.
Research Station, Integrating Human and Natural Systems, Paul V. Bolstad is a Professor, University of Minnesota,
Forestry Sciences Laboratory, Athens, GA 30602. College of Natural Resources, Department of Forest
Peter V. Caldwell is a Research Hydrologist, U.S. Department Resources, St. Paul, MN 55108.
of Agriculture, Forest Service, Southern Research Station, Christopher P. Carlson is a National Program Leader, U.S.
Center for Integrated Forest Science, Otto, NC 28763. Department of Agriculture, Forest Service, Watershed and
Aquatic Ecology Research, Washington, DC 20250.
REFERENCES
Abildtrup, J.; Garcia, S.; Stenger, A. 2013. The effect of forest Brown, T.C.; Mahat, V.; Ramirez, J.A. 2019. Adaptation
land use on the cost of drinking water supply: a spatial to future water shortages in the United States caused by
econometric analysis. Ecological Economics. 92: 126–136. population growth and climate change. Earth’s Future.
https://doi.org/10.1016/j.ecolecon.2013.01.004. 7(3): 219–234. https://doi.org/10.1029/2018EF001091.
Andréassian, V. 2004. Waters and forests: from Brunette, V.; Germain, R.H. 2003. Forest management in
historical controversy to scientific debate. Journal of the New York City watershed. Paper submitted to the
Hydrology. 291(1–2): 1–27. https://doi.org/10.1016/j. XII World Forestry Congress. https://www.fao.org/3/
jhydrol.2003.12.015. xii/0649-b3.htm. [Date last accessed: October 27, 2021].
Arthur, M.A.; Coltharp, G.B.; Brown, D.L. 1998. Effects Brutsaert, W.; Stricker, H. 1979. An advection-aridity
of best management practices on forest streamwater approach to estimate actual regional evapotranspiration.
quality in eastern Kentucky. Journal of the American Water Resources Research. 15(2): 443–450. https://doi.
Water Resources Association. 34(3): 481–495. https://doi. org/10.1029/WR015I002P00443.
org/10.1111/j.1752-1688.1998.tb00948.x. Burnash, R.J.C. 1995. The NWS river forecast system -
Ashoori, N.; Dzombak, D.A.; Small, M.J. 2015. catchment modelling. In: Singh, V.P., ed. Computer models
Sustainability review of water-supply options in the Los of watershed hydrology. Littleton, CO: Water Resources
Angeles region. Journal of Water Resources Planning Publications: 311–366.
and Management. 141(12). https://doi.org/10.1061/(asce) Burnash, R.J.C.; Ferral, R.L.; McGuire, R.A. 1973. A
wr.1943-5452.0000541. generalized streamflow simulation system: conceptual
Averyt, K.; Fisher, J.; Huber-Lee, A. [and others]. 2011. modeling for digital computers. [Place of publication
Freshwater use by U.S. power plants: electricity’s thirst for unknown]: U.S. Department of Commerce, National
a precious resource. Cambridge, MA: Union of Concerned Weather Service, and State of California, Department
Scientists. 52 p. of Water Resources. 204 p. http://books.google.com/
books?id=aQJDAAAAIAAJ. [Date last accessed: October
Barnes, M.C.; Todd, A.H.; Lilja, R.W.; Barten, P.K. 2009.
27, 2021].
Forests, water and people: drinking water supply and
forest lands in the Northeast and Midwest United States. Caldwell, P.V.; Kennen, J.G.; Hain, E.F. [and others]. 2020.
NA-FR-01-08. Newtown Square, PA: U.S. Department of Hydrologic modeling for flow-ecology science in the
Agriculture, Forest Service, Northeastern Area State and Southeastern United States and Puerto Rico. Gen. Tech.
Private Forestry. 71 p. https://www.fs.fed.us/nrs/pubs/na/ Rep. SRS-246. U.S. Department of Agriculture, Forest
NA-FR-01-08.pdf. [Date last accessed: October 27, 2021]. Service, Southern Research Station. 77 p. https://doi.
org/10.2737/SRS-GTR-246.
Bosch, J.M.; Hewlett, J.D. 1982. A review of catchment
experiments to determine the effect of vegetation Caldwell, P.V.; Kennen, J.G.; Sun, G. [and others]. 2015. A
changes on water yield and evapotranspiration. comparison of hydrologic models for ecological flows and
Journal of Hydrology. 55(1–4): 3–23. https://doi. water availability. Ecohydrology. 8(8): 1525–1546. https://
org/10.1016/0022-1694(82)90117-2. doi.org/10.1002/eco.1602.
Brogna, D.; Dufrêne, M.; Michez, A. [and others]. 2018. Caldwell, P.V.; Muldoon, C.; Miniat, C.F. [and others]. 2014.
Forest cover correlates with good biological water quality. Quantifying the role of National Forest System lands in
Insights from a regional study (Wallonia, Belgium). Journal providing surface drinking water supply for the Southern
of Environmental Management. 211: 9–21. https://doi. United States. Gen. Tech. Rep.SRS-197. Asheville,
org/10.1016/j.jenvman.2018.01.017. NC: U.S. Department of Agriculture, Forest Service,
Southern Research Station. 135 p. https://doi.org/10.2737/
Brown, A.E.; Western, A.W.; McMahon, T.A.; Zhang, L.
SRS-GTR-197.
2013. Impact of forest cover changes on annual streamflow
and flow duration curves. Journal of Hydrology. 483(0): Caldwell, P.V.; Sun, G.; McNulty, S.G. [and others]. 2012.
39–50. https://doi.org/10.1016/j.jhydrol.2012.12.031. Impacts of impervious cover, water withdrawals, and
climate change on river flows in the conterminous US.
Brown, T.C.; Froemke, P.; Mahat, W.; Ramirez, J.A. 2016.
Hydrology and Earth System Sciences. 16(8): 2839–2857.
Mean annual renewable water supply of the contiguous
https://doi.org/10.5194/hess-16-2839-2012.
United States. Briefing paper. Fort Collins, CO: U.S.
Departent of Agriculture, Forest Service, Rocky Mountain Caldwell, P.; Sun, G.; McNulty, S. [and others]. 2011.
Research Station. 55 p. http://www.fs.fed.us/rmrs/ Modeling impacts of environmental change on ecosystem
documents-and-media/really-mean-annual-renewable- services across the conterminous United States. In:
water-supply-contiguous-united-states. [Date last accessed: Medley, C.N.; Patterson, G.; Parker, M.J., eds. Observing,
October 27, 2021]. studying, and managing for change—proceedings of
the fourth interagency conference on research in the
Brown, T.C.; Hobbins, M.T.; Ramirez, J.A. 2008. Spatial
watersheds. Scientific Investigations Report 2011–5169.
distribution of water supply in the conterminous
Reston, VA: U.S. Department of the Interior, U.S.
United States. Journal of the American Water
Geological Survey: 63–69.
Resources Association. 44(6): 1474–1487. https://doi.
org/10.1111/j.1752-1688.2008.00252.x.
Creed, I.F.; Jones, J.A.; Archer, E. [and others]. 2019. Easterling, D.R.; Kunkel, K.E.; Arnold, J.R. [and others].
Managing forests for both downstream and downwind 2017. Precipitation change in the United States. In:
water. Frontiers in Forests and Global Change. 2: 64. Wuebbles, D.J.; Fahey, D.W.; Hibbard, K.A. [and others],
https://doi.org/10.3389/ffgc.2019.00064. eds. Climate science special report: Fourth National
Climate Assessment. Vol. I. Washington, DC: U.S.
Creed, I.F.; Spargo, A.T.; Jones, J.A. [and others]. 2014.
Global Change Research Program: 207–230. https://doi.
Changing forest water yields in response to climate
org/10.7930/J0H993CC.
warming: results from long-term experimental watershed
sites across North America. Global Change Biology. Emanuel, R.E.; Buckley, J.J.; Caldwell, P.V. [and
20(10): 3191–3208. https://doi.org/10.1111/gcb.12615. others]. 2015. Influence of basin characteristics on
the effectiveness and downstream reach of interbasin
Creed, I.F.; van Noordwijk, M. 2018. Forest and water on a
water transfers: displacing a problem. Environmental
changing planet: vulnerability, adaptation and governance
Research Letters. 10(12): 124005. https://doi.
opportunities. World Series vol. 38. Vienna: International
org/10.1088/1748-9326/10/12/124005.
Union of Forest Research Organizations (IUFRO). 192 p.
Farley, K.A.; Jobbágy, E.G.; Jackson, R.B. 2005.
Curtis, P.G.; Slay, C.M.; Harris, N.L. [and others]. 2018.
Effects of afforestation on water yield: a global
Classifying drivers of global forest loss. Science. 361(6407):
synthesis with implications for policy. Global
1108–1111. https://doi.org/10.1126/science.aau3445.
Change Biology. 11(10): 1565–1576. https://doi.
D’Annunzio, R.; Sandker, M.; Finegold, Y.; Min, Z. 2015. org/10.1111/j.1365-2486.2005.01011.x.
Projecting global forest area towards 2030. Forest Ecology
Filoso, S.; Bezerra, M.O.; Weiss, K.C.B.; Palmer, M.A.
and Management. 352: 124–133. https://doi.org/10.1016/j.
2017. Impacts of forest restoration on water yield: a
foreco.2015.03.014.
systematic review. PLOS ONE. 12(8): e0183210. https://
Daly, C.; Taylor, G.; Gibson, W. 1997. The PRISM approach doi.org/10.1371/journal.pone.0183210.
to mapping precipitation and temperature. In: Proceedings,
10th conference on applied climatology, Reno, NV. Boston, Fiquepron, J.; Garcia, S.; Stenger, A. 2013. Land use
impact on water quality: valuing forest services in terms
MA: American Meteorological Society: 10–12.
of the water supply sector. Journal of Environmental
Dickson, K.E.; Dzombak, D.A. 2017. Inventory of interbasin Management. 126: 113–121. https://doi.org/10.1016/j.
transfers in the United States. Journal of the American jenvman.2013.04.002.
Water Resources Association. 53(5): 1121–1132. https://doi.
Giri, S.; Qiu, Z. 2016. Understanding the relationship of land
org/10.1111/1752-1688.12561.
uses and water quality in twenty first century: a review.
Dickson, K.E.; Marston, L.T.; Dzombak, D.A. 2020. Journal of Environmental Management. 173: 41–48.
Editorial perspectives: the need for a comprehensive, https://doi.org/10.1016/j.jenvman.2016.02.029.
centralized database of interbasin water transfers in the
Goode, J.R.; Luce, C.H.; Buffington, J.M. 2012. Enhanced
United States. Environmental Science: Water Research
sediment delivery in a changing climate in semi-arid
& Technology. 6(3): 420–422. https://doi.org/10.1039/
mountain basins: implications for water resource
D0EW90005B.
management and aquatic habitat in the northern Rocky
Dieter, C.A.; Maupin, M.A. 2017. Public supply and Mountains. Geomorphology. 139: 1–15.
domestic water use in the United States, 2015. Open-
Gosling, S.N.; Arnell, N.W. 2016. A global assessment of
File Report 2017–1131. Reston, VA: U.S. Department
the impact of climate change on water scarcity. Climatic
of the Interior, U.S. Geological Survey. 6 p. https://doi.
Change. 134(3): 371–385. https://doi.org/10.1007/
org/10.3133/ofr20171131.
s10584-013-0853-x.
Duan, K.; Caldwell, P.V.; Sun, G. [and others]. 2019.
Greene, J.L.; Straka, T.J.; Cushing, T.L. 2013. The Southern
Understanding the role of regional water connectivity in
Forest Futures Project: technical report. Gen. Tech. Rep.
mitigating climate change impacts on surface water supply
SRS-178. Asheville, NC: U.S. Department of Agriculture,
stress in the United States. Journal of Hydrology. 570:
Forest Service, Southern Research Station. 542 p.
80–95. https://doi.org/10.1016/j.jhydrol.2019.01.011.
Guo, F.; Lenoir, J.; Bonebrake, T.C. 2018 Land-use change
Duan, K.; Sun, G.; Caldwell, P.V. [and others]. 2018.
interacts with climate to determine elevational species
Implications of upstream flow availability for watershed
redistribution. Nature Communications. 9: 1315. https://
surface water supply across the conterminous
doi.org/10.1038/s41467-018-03786-9.
United States. Journal of the American Water
Resources Association. 54(3): 694–707. https://doi. Haddeland, I.; Heinke, J.; Biemans, H. [and others]. 2014.
org/10.1111/1752-1688.12644. Global water resources affected by human interventions
and climate change. Proceedings of the National Academy
Duan, K.; Sun, G.; McNulty, S.G. [and others]. 2017. Future
of Sciences of the United States of America. 111(9): 3251–
shift of the relative roles of precipitation and temperature
3256. https://doi.org/10.1073/pnas.1222475110.
in controlling annual runoff in the conterminous United
States. Hydrology and Earth System Sciences. 21(11): Hallema, D.W.; Sun, G.; Caldwell, P.V. [and others].
5517–5529. https://doi.org/10.5194/hess-21-5517-2017. 2018. Burned forests impact water supplies. Nature
Communications. 9: 1307. https://doi.org/10.1038/
Dudley, N.; Stolton, S. 2003. Running pure: the
s41467-018-03735-6.
importance of forest protected areas to drinking
water. [Place of publication unknown]: World Bank/ Hamon, W.R. 1963. Computation of direct runoff amounts
WWF Alliance. https://openknowledge.worldbank.org/ from storm rainfall. International Association of Scientific
handle/10986/15006. [Date last accessed: October 28, Hydrology Publication. 63: 52–62.
2021].
Hoegh-Guldberg, O.; Jacob, D.; Taylor, M. [and others]. Liu, N.; Dobbs, R.; Caldwell, P.V. [and others]. 2022. Inter-
2018. Impacts of 1.5°C global warming on natural and basin transfers extend the benefits of water from forests
human systems. In: Masson-Delmotte, V.; Zhai, P.; to population centers across the conterminous U.S. Water
Pörtner, H.-O. [and others], eds. Global warming of Resources Research. 58(5): e2021WR031537. https://doi.
1.5°C. An IPCC special report on the impacts of global org/10.1029/2021WR031537.
warming of 1.5°C above pre-industrial levels and related Livneh, B.; Rosenberg, E.A.; Lin, C. [and others]. 2013. A
global greenhouse gas emission pathways, in the context of long-term hydrologically based dataset of land surface
strengthening the global response to the threat of climate fluxes and states for the conterminous United States:
change. [Place of publication unknown]: Intergovernmental update and extensions. Journal of Climate. 26(23): 9384–
Panel on Climate Change: 175–311. https://www.ipcc.ch/ 9392. https://doi.org/10.1175/JCLI-D-12-00508.1.
sr15. [Date last accessed: October 28, 2021].
Lockaby, G.; Nagy, C.; Vose, J.M. [and others]. 2013. Forests
Homer, C.; Dewitz, J.; Jin, S. [and others]. 2020. and water. In: Wear, D.N.; Greis, J.G., eds. 2013. The
Conterminous United States land cover change patterns Southern Forest Futures Project: technical report. Gen.
2001–2016 from the 2016 National Land Cover Tech. Rep. SRS-178. Asheville, NC: U.S. Department of
Database. ISPRS Journal of Photogrammetry and Agriculture, Forest Service, Southern Research Station:
Remote Sensing. 162: 184–199. https://doi.org/10.1016/j. 309–339.
isprsjprs.2020.02.019.
Lopes, A.F.; Macdonald, J.L.; Quinteiro, P. [and others].
Homer, C.; Dewitz, J.; Yang, L. [and others]. 2015. 2019. Surface vs. groundwater: the effect of forest
Completion of the 2011 National Land Cover Database cover on the costs of drinking water. Water Resources
for the conterminous United States – representing a decade and Economics. 28: 100123. https://doi.org/10.1016/j.
of land cover change information. Photogrammetric wre.2018.06.002.
Engineering and Remote Sensing. 81: 345–354.
Luce, C.H.; Lute, A.; Kormos, P.; Livneh, B. 2017. Modeled
Hultine, K.R.; Nagler, P.L.; Morino, K. [and others]. 2010. historical streamflow metrics for the contiguous United
Sap flux-scaled transpiration by tamarisk (Tamarix States and national forest lands. Fort Collins, CO: U.S.
spp.) before, during and after episodic defoliation by the Department of Agriculture, Forest Service, Research Data
saltcedar leaf beetle (Diorhabda carinulata). Agricultural Archive. https://doi.org/10.2737/RDS-2017-0046.
and Forest Meteorology. 150(11): 1467–1475. https://doi.
org/10.1016/j.agrformet.2010.07.009. Mahat, V.; Ramírez, J.A.; Brown, T.C. 2017. Twenty-first-
century climate in CMIP5 simulations: implications for
Isaak, D.J.; Wenger, S.J.; Peterson, E.E. [and others]. 2017. snow and water yield across the contiguous United States.
The NorWeST summer stream temperature model and Journal of Hydrometeorology. 18(8): 2079–2099. https://
scenarios for the Western US: a crowd-sourced database doi.org/10.1175/JHM-D-16-0098.1.
and new geospatial tools foster a user community and
predict broad climate warming of rivers and streams. Mapulanga, A.M.; Naito, H. 2019. Effect of deforestation on
Water Resources Research. 53(11): 9181–9205. https://doi. access to clean drinking water. Proceedings of the National
org/10.1002/2017WR020969. Academy of Sciences. 116(17): 8249–8254. https://doi.
org/10.1073/pnas.1814970116.
Jackson, C.R.; Sun, G.; Amatya, D. [and others]. 2004. Fifty
years of forest hydrology in the Southeast. In: Ice, G.G.; Marion, D.; Sun, G.; Caldwell, P. [and others]. 2013.
Stednick, J.D., eds. A century of forested and wildland Managing forest water quantity and quality under climate
watershed lessons. Bethesda, MD: Society of American change. In: Vose, J.M.; Klepzig, K.D., eds. Climate change
Foresters: 33–112. Chapter 3. adaptation and mitigation management options. Boca
Raton, FL: CRC Press: 249–306. https://doi.org/10.1201/
Koren, V.; Smith, M.; Duan, Q. 2003. Use of a priori b15613-10.
parameter estimates in the derivation of spatially consistent
parameter sets of rainfall-runoff models. In: Duan, Q.; Martinuzzi, S.; Januchowski-Hartley, S.R.; Pracheil,
Gupta, H.V.; Sorooshian, S. [and others], eds. Calibration B.M. [and others]. 2014. Threats and opportunities for
of watershed models. Vol. 6. Washington, DC: American freshwater conservation under future land use change
Geophysical Union: 239–254. https://doi.org/10.1029/ scenarios in the United States. Global Change Biology.
WS006p0239. 20(1): 113–124. https://doi.org/10.1111/gcb.12383.
Li, C.; Sun, G.; Caldwell, P.V. [and others]. 2020. Impacts McCabe, G.J.; Wolock, D.M. 1999. General-circulation-
of urbanization on watershed water balances across the model simulations of future snowpack in the Western
conterminous United States. Water Resources Research. United States. Journal of the American Water
56: e2019WR026574. https://doi.org/10.1029/2019WR026574. Resources Association. 35(6): 1473–1484. https://doi.
org/10.1111/j.1752-1688.1999.tb04231.x.
Liu, N.; Caldwell, P.V.; Dobbs, G.R. [and others]. 2021.
Forested lands dominate drinking water supply in the McDonald, R.I.; Weber, K.; Padowski, J. [and others]. 2014.
conterminous United States. Environmental Research Water on an urban planet: urbanization and the reach
Letters. 16: 084008. https://doi.org/10.1088/1748-9326/ of urban water infrastructure. Global Environmental
ac09b0. Change. 27(1): 96–105. https://doi.org/10.1016/j.
gloenvcha.2014.04.022.
Liu, N.; Dobbs, G.R.; Caldwell, P.V. [and others]. 2020.
Quantifying the role of State and private forest lands in McNulty, S.; Moore Myers, J.; Caldwell, P.; Sun, G. 2013.
providing surface drinking water supply for the Southern Climate change summary. In: Wear, D.N.; Greis, J.G.,
United States. Gen. Tech. Rep. SRS-248. Asheville, eds. 2013. The Southern Forest Futures Project: technical
NC: U.S. Department of Agriculture, Forest Service, report. Gen. Tech. Rep. SRS-178. Asheville, NC: U.S.
Southern Research Station. 405 p. https://doi.org/10.2737/ Department of Agriculture, Forest Service, Southern
SRS-GTR-248. Research Station: 27–43.
Moore, R.B.; Dewald, T.G. 2016. The road to NHDPlus— Rockaway, T.D.; Coomes, P.A.; Rivard, J.; Kornstein,
advancements in digital stream networks and B. 2011. Residential water use trends in North
associated catchments. Journal of the American Water America. Journal AWWA. 103(2): 76–89. https://doi.
Resources Association. 52(4): 890–900. https://doi. org/10.1002/j.1551-8833.2011.tb11403.x.
org/10.1111/1752-1688.12389. Schwalm, C.R.; Huntzinger, D.N.; Cook, R.B. [and
Moore, R.D.; Spittlehouse, D.L.; Story, A. 2005. Riparian others]. 2015. How well do terrestrial biosphere models
microclimate and stream temperature response to forest simulate coarse-scale runoff in the contiguous United
harvesting: a review. Journal of the American Water States? Ecological Modelling. 303: 87–96. https://doi.
Resources Association. 41(4): 813–834. https://doi. org/10.1016/j.ecolmodel.2015.02.006.
org/10.1111/j.1752-1688.2005.tb03772.x. Seattle Public Utilities. 2011. South Fork Tolt watershed
Mooty, W.S.; Jeffcoat, H.H. 1986. Inventory of interbasin management plan. Seattle, WA. https://www.seattle.gov/
transfer of water in the Eastern United States. Open- Documents/Departments/SPU/EnvironmentConservation/
File Report 86-148. Tuscaloosa, AL: U.S. Department SouthForkToltMunicipalWatershedManagementPlan.pdf.
of the Interior, U.S. Geological Survey. 47 p. https://doi. [Date last accessed: October 28, 2021].
org/10.3133/ofr86148. Sedell, J.; Sharpe, M.; Apple, D. [and others]. 2000. Water
Nagy, R.C.; Lockaby, G.B.; Kalin, L.; Anderson, C. 2012. and the Forest Service. FS-660. Washington, DC: U.S.
Effects of urbanization on stream hydrology and water Department of Agriculture, Forest Service. 40 p.
quality: the Florida Gulf Coast. Hydrological Processes. Spracklen, D.V.; Arnold, S.R.; Taylor, C.M. 2012.
26(13): 2019–2030. https://doi.org/10.1002/hyp.8336. Observations of increased tropical rainfall preceded by air
National Research Council. 2008. Hydrologic effects of a passage over forests. Nature. 489: 282–285. https://doi.
changing forest landscape. Washington, DC: The National org/10.1038/nature11390.
Academies Press. 167 p. https://doi.org/10.17226/12223. Stein, S.M.; McRoberts, R.E.; Alig, R.J. [and others]. 2005.
Naumann, G.; Alfieri, L.; Wyser, K. [and others]. 2018. Forests on the edge: housing development on America’s
Global changes in drought conditions under different levels private forests. Gen. Tech. Rep. PNW-636. Portland, OR:
of warming. Geophysical Research Letters. 45(7): 3285– U.S. Department of Agriculture, Forest Service, Pacific
3296. https://doi.org/10.1002/2017GL076521. Northwest Research Station. 16 p. https://doi.org/10.2737/
PNW-GTR-636.
NYC Environmental Protection. 2017. New York City 2017
drinking water supply and quality report. New York. 20 Stein, S.M.; McRoberts, R.E.; Mahal, L.G. [and others].
p. https://www1.nyc.gov/assets/nyw/downloads/pdf/nyw- 2009. Private forests, public benefits: increased
2017-dep-water-report.pdf. [Date last accessed: October housing density and other pressures on private forest
28, 2021]. contributions. Gen. Tech. Rep. PNW-795. Portland, OR:
U.S. Department of Agriculture, Forest Service, Pacific
Oswalt, S.N.; Smith, W.B.; Miles, P.D. [and others]. 2019.
Northwest Research Station. 74 p. https://doi.org/10.2737/
Forest resources of the United States, 2017: a technical
PNW-GTR-795.
document supporting the Forest Service 2020 RPA
Assessment. Gen. Tech. Rep. WO-97. Washington, Sun, G.; Alstad, K.; Chen, J. [and others]. 2011a. A general
DC: U.S. Department of Agriculture, Forest Service, predictive model for estimating monthly ecosystem
Washington Office. 223 p. https://doi.org/10.2737/ evapotranspiration. Ecohydrology. 4(2): 245–255. https://
WO-GTR-97. doi.org/10.1002/eco.194.
Petsch, H.E. 1985. Inventory of interbasin transfers of Sun, G.; Caldwell, P.; Noormets, A. [and others].
water in the western conterminous United States. Open- 2011b. Upscaling key ecosystem functions across the
File Report 85-166. Lakewood, CO: U.S. Department conterminous United States by a water-centric ecosystem
of the Interior, U.S. Geological Survey. 45 p. https://doi. model. Journal of Geophysical Research. 116: G00J05.
org/10.3133/OFR85166. https://doi.org/10.1029/2010jg001573.
Radeloff, V.C.; Hammer, R.B.; Stewart, S.I. [and others]. Sun, G.; McNulty, S.G.; Amatya, D.M. [and others]. 2002. A
2005. The wildland-urban interface in the United States. comparison of the watershed hydrology of coastal forested
Ecological Applications. 15(3): 799–805. https://doi. wetlands and the mountainous uplands in the Southern
org/10.1890/04-1413. US. Journal of Hydrology. 263(1–4): 92–104. https://doi.
org/10.1016/S0022-1694(02)00064-1.
Renner, M.; Bernhofer, C. 2012. Applying simple water-
energy balance frameworks to predict the climate Sun, G.; McNulty, S.G.; Lu, J. [and others]. 2005. Regional
sensitivity of streamflow over the continental United States. annual water yield from forest lands and its response to
Hydrology and Earth System Sciences. 16(8): 2531–2546. potential deforestation across the Southeastern United
https://doi.org/10.5194/hess-16-2531-2012. States. Journal of Hydrology. 308(1–4): 258–268. https://
doi.org/10.1016/j.jhydrol.2004.11.021.
Riahi, K.; van Vuuren, D.P.; Kriegler, E. [and others]. 2017.
The Shared Socioeconomic Pathways and their energy, Sun, G.; McNulty, S.G.; Moore Myers, J.A.; Cohen, E.C.
land use, and greenhouse gas emissions implications: an 2008. Impacts of multiple stresses on water demand and
overview. Global Environmental Change. 42: 153–168. supply across the Southeastern United States. Journal of
https://doi.org/10.1016/j.gloenvcha.2016.05.009. the American Water Resources Association: 44(6): 1441–
1457. https://doi.org/10.1111/j.1752-1688.2008.00250.x.
Riekerk, H. 1985. Water-quality effects of pine flatwoods
silviculture. Journal of Soil and Water Conservation. 40(3): Sun, G.; Vose, J.M. 2016. Forest management challenges for
306–309. sustaining water resources in the Anthropocene. Forests.
7(3): 1–13. https://doi.org/10.3390/f7030068.
Sun, S.; Sun, G.; Caldwell, P. [and others]. 2015a. Drought U.S. Department of Agriculture (USDA), Forest Service.
impacts on ecosystem functions of the U.S. national forests 2007. Todays challenges and opportunities: abundant clean
and grasslands: part I evaluation of a water and carbon water [Briefing paper]. Washington, DC: U.S. Department
balance model. Forest Ecology and Management. 353: of Agriculture, Forest Service, Office of Communication. 2
260–268. https://doi.org/10.1016/j.foreco.2015.03.054. p. https://www.fs.fed.us/biology/resources/pubs/watershed/
abundant-clean-water.pdf. [Date last accessed: December
Sun, S.; Sun, G.; Caldwell, P. [and others]. 2015b. Drought
1, 2021].
impacts on ecosystem functions of the U.S. national forests
and grasslands: part II assessment results and management U.S. Environmental Protection Agency (EPA). 2017.
implications. Forest Ecology and Management. 353: 269– Safe Drinking Water Information System (SDWIS)
279. https://doi.org/10.1016/j.foreco.2015.04.002. Federal reporting services. https://www.epa.gov/
ground-water-and-drinking-water/safe-drinking-water-
Sun, Y.; Piao, S.; Huang, M. [and others]. 2016. Global
information-system-sdwis-federal-reporting. [Date last
patterns and climate drivers of water-use efficiency
accessed: October 29, 2021].
in terrestrial ecosystems deduced from satellite-based
datasets and carbon cycle models. Global Ecology and U.S. Environmental Protection Agency (EPA). 2020.
Biogeography. 25(3): 311–323. https://doi.org/10.1111/ Integrated Climate and Land-Use Scenarios (ICLUS
geb.12411. version 2.1) for the Fourth National Climate Assessment.
https://www.epa.gov/gcx/about-gcx-iclus-tool. [Date last
Suttles, K.M.; Singh, N.K.; Vose, J.M. [and others]. 2018.
Assessment of hydrologic vulnerability to urbanization accessed: October 28, 2021].
and climate change in a rapidly changing watershed in Vörösmarty, C.J.; Green, P.; Salisbury, J.; Lammers, R.B.
the Southeast U.S. Science of the Total Environment. 645: 2000. Global water resources: vulnerability from climate
806–816. https://doi.org/10.1016/j.scitotenv.2018.06.287. change and population growth. Science. 289(5477): 284–
288. https://doi.org/10.1126/science.289.5477.284.
Swank, W.T.; Vose, J.M.; Elliott, K.J. 2001. Long-term
hydrologic and water quality responses following Vose, J.M. 2019. Forest and water in the 21st century: a
commercial clearcutting of mixed hardwoods on a global perspective. Journal of Forestry. 117(1): 80–85.
Southern Appalachian catchment. Forest Ecology and https://doi.org/10.1093/jofore/fvy054.
Management. 143(1–3): 163–178. https://doi.org/10.1016/ Warziniack, T.; Sham, C.H.; Morgan, R.; Feferholtz, Y.
S0378-1127(00)00515-6. 2017. Effect of forest cover on water treatment costs.
Tavernia, B.G.; Nelson, M.D.; Caldwell, P.; Sun, G. 2013. Water Economics and Policy. 3(4): 1750006. https://doi.
Water stress projections for the Northeastern and org/10.1142/S2382624X17500060.
Midwestern United States in 2060: anthropogenic and Wear, D.N.; Huggett, R.; Li, R. [and others]. 2013. Forecasts
ecological consequences. Journal of the American Water of forest conditions in regions of the United States under
Resources Association. 49(4): 938–952. https://doi. future scenarios: a technical document supporting the
org/10.1111/jawr.12075. Forest Service 2012 RPA Assessment. Gen. Tech. Rep.
Taylor, M. 2018. Improving California’s forest and SRS-170. Asheville, NC: U.S. Department of Agriculture,
watershed management. Sacramento, CA: Legislative Forest Service, Southern Research Station. 101 p. https://
Analyst’s Office. 42 p. doi.org/10.2737/SRS-GTR-170.
Theobald, D.M. 2005. Landscape patterns of exurban Yigzaw, W.; Hossain, F. 2016. Water sustainability of
growth in the USA from 1980 to 2020. Ecology large cities in the United States from the perspectives
and Society. 10(1): 32. https://doi.org/10.5751/ of population increase, anthropogenic activities, and
ES-01390-100132. climate change. Earth’s Future. 4(12): 603–617. https://doi.
org/10.1002/2016EF000393.
Tu, J. 2013. Spatial variations in the relationships between
land use and water quality across an urbanization Zhang, L.; Dawes, W.R.; Walker, G.R. 2001. Response of
gradient in the watersheds of northern Georgia, USA. mean annual evapotranspiration to vegetation changes at
Environmental Management. 51(1): 1–17. https://doi. catchment scale. Water Resources Research. 37(3): 701–
org/10.1007/s00267-011-9738-9. 708. https://doi.org/10.1029/2000WR900325.
U.S. Census Bureau. 2010. State and county quick facts [data Zhang, Y.; Song, C.; Sun, G. [and others]. 2016.
derived from population estimates, American Community Development of a coupled carbon and water model
Survey, Census of Population and Housing, County for estimating global gross primary productivity and
Business Patterns, Economic Census, Survey of Business evapotranspiration based on eddy flux and remote sensing
Owners, building permits]. https://www.census.gov/ data. Agricultural and Forest Meteorology. 223: 116–131.
quickfacts/. [Date last accessed: October 29, 2021]. https://doi.org/10.1016/j.agrformet.2016.04.003.
Liu, Ning; Dobbs, G. Rebecca; Caldwell, Peter V.; Miniat, Chelcy F.; Sun, Ge; Duan,
Kai; Nelson, Stacy A.C.; Bolstad, Paul V.; Carlson, Christopher P. 2022. Quantifying
the role of National Forest System and other forested lands in providing surface
drinking water supply for the conterminous United States. Gen. Tech. Rep. WO-100.
Washington, DC: U.S. Department of Agriculture, Forest Service, Washington
Office. 40 p. [plus supplement]. https://doi.org/10.2737/WO-GTR-100.
The U.S. Department of Agriculture’s Forest Service manages more than 779,000 km2
(193 million acres) of national forests and grasslands (collectively, National Forest
System [NFS] lands) that play a significant role in providing clean, fresh water for local
ecosystems and economies. This water is sometimes transferred hundreds of kilometers
away to also serve big cities through inter-basin transfers (IBTs). The contribution of
NFS lands to surface drinking water supplies for public water systems has not been
assessed at the national scale while accounting for IBTs. The Forest Service Water
Supply Stress Index (WaSSI) model was modified to provide estimates of 2001–2015
mean annual surface water supply and the proportion of mean surface water supply
originating on 172 NFS land units and other forested lands at the 12-digit hydrologic
unit code scale across the conterminous United States (CONUS) while accounting for
water transfer through IBTs. Predictions of the proportion of surface water supply
originating on NFS and other forested lands were linked to specific downstream
communities and populations, using surface drinking water intake information from
the U.S. Environmental Protection Agency Safe Drinking Water Information System
database of public water systems. A new database of 594 IBTs was compiled for this
study, ranging from 0.01 million m3 yr-1 to 8,900 million m3 yr-1, for a total transferred
volume of 116,894 million m3 yr-1. Overall, NFS lands comprised 9.2 percent of the total
CONUS land area but contributed 12.8 percent of the surface water supply. In the West,
NFS lands comprised 19.2 percent of the total land area but contributed 46.3 percent of
the 478.7 billion m3 yr-1 surface water supply; in the East, NFS lands comprised about
2.8 percent of the total land area and 3.8 percent (66.6 billion m3 yr-1) of the surface
water supply. In total across the CONUS, NFS and other forested lands comprised 28.7
percent of the total land area but contributed 46.0 percent of the surface water supply.
Approximately 45.8 million people derived >10 percent of their surface drinking water
supply from NFS lands, and 22.6 million people received >50 percent of their surface
drinking water supply from NFS lands. Approximately 125.5 million people, about
39 percent of the total population in the CONUS in 2017, derived >10 percent of their
surface drinking water supply from NFS and other forested lands, with 83.1 million
people receiving >50 percent of their surface drinking water supply from NFS and other
forested lands. In addition to those populations receiving surface drinking water supply
from their local public surface drinking water intakes, 12.6 million people were served
by public water systems that purchased surface drinking water supply from other public
water systems deriving >10 percent of their surface drinking water supply from NFS
lands. This study provides a systematic accounting of NFS and other forested lands for
surface drinking water supply. Our results can aid water resource and forest managers
in developing integrated watershed management plans at a time when climate change,
population growth, and land development threaten water supplies.
Keywords: Drinking water, inter-basin transfers, National Forest System, WaSSI, water
supply, water yield.
In accordance with Federal civil rights law and U.S. Department of
Agriculture (USDA) civil rights regulations and policies, the USDA, its
Agencies, offices, and employees, and institutions participating in or
administering USDA programs are prohibited from discriminating based
on race, color, national origin, religion, sex, gender identity (including
gender expression), sexual orientation, disability, age, marital status,
family/parental status, income derived from a public assistance program,
political beliefs, or reprisal or retaliation for prior civil rights activity,
in any program or activity conducted or funded by USDA (not all bases
apply to all programs). Remedies and complaint filing deadlines vary by
program or incident.
Persons with disabilities who require alternative means of
communication for program information (e.g., Braille, large print,
audiotape, American Sign Language, etc.) should contact the responsible
Agency or USDA’s TARGET Center at (202) 720-2600 (voice and TTY)
or contact USDA through the Federal Relay Service at (800) 877-8339.
Additionally, program information may be made available in languages
other than English.
To file a program discrimination complaint, complete the USDA
Program Discrimination Complaint Form, AD-3027, found online at
http://www. ascr.usda.gov/complaint_filing_cust.html and at any USDA
office or write a letter addressed to USDA and provide in the letter
all of the information requested in the form. To request a copy of the
complaint form, call (866) 632-9992. Submit your completed form or
letter to USDA by: (1) mail: U.S. Department of Agriculture, Office of
the Assistant Secretary for Civil Rights, 1400 Independence Avenue, SW,
Washington, D.C. 20250-9410; (2) fax: (202) 690-7442; or (3) email:
[email protected].
USDA is an equal opportunity provider, employer, and lender.

GTR Wo 100

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

GTR Wo 100

Uploaded by

Copyright:

Available Formats

Research and Development | General Technical Report WO-100 | September 2022

Quantifying the Role of National Forest

Published by Southern Research Station

CASE STUDY: PIKE NATIONAL FOREST . . . . . . . . . . . . . . . . . . . . 25

ABOUT THE AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

SUPPLEMENTAL MATERIAL . . . https://doi.org /10.2737/ WO-GTR-100-Sup1

Surface water supply is calculated by accumulating the total volume of water

Data/database Source Resolution Time period

Table 2—Data sources for inter-basin transfer volumes

Agency name Level Accessed via

Percentage of water from NFS lands

26–50 Water Resource Region

Percentage of water from NFS and

26–50 Water Resource Region

West a East b CONUS

Inter-basin surface water transfers by origin and destination HUC12s

Water Resource Region

■ Arapaho NF ■ San Juan NF ■ Other forested lands

Percentage of water from NFS

>90 >80 >70 >60 >50 >40 >30 >20 >10

NFS and other surface water supplying key arid cities

Other NFs (<1% each):

■ Angeles NF ■ Eldorado NF ■ Plumas NF ■ Tahoe NF

Pike National Forest in Colorado

One intake of Denver Water Board, CO,

Two intakes of City of Aurora, CO,

Streamlines Intakes Population

Streamlines Water Resource Region

Supplemental information for General Technical Report WO-100 (https://doi.org/10.2737/WO-GTR-100)

38 REGION 2: ROCKY MOUNTAIN REGION

BETWEEN 2001 AND 2015:

Example public water systems (PWSs) getting surface drinking water

by forests through evapotranspiration is generally Role of IBTs in Distributing Water

Brown Brown Luce

Model temporal resolution Monthly/mean annual Daily Daily Monthly

NFS boundaries Proclamation Ownership and Proclamation Ownership

ABOUT THE AUTHORS

You might also like