Journal of Hydrology 542 (2016) 744755

Contents lists available at ScienceDirect

Journal of Hydrology
journal homepage: www.elsevier.com/locate/jhydrol

Research papers

Potential of green infrastructure to restore predevelopment water

budget of a semi-arid urban catchment
Youcan Feng , Steven Burian, Christine Pomeroy
Department of Civil and Environmental Engineering, University of Utah, 110 Central Campus Drive, Suite 2000, Salt Lake City, UT 84112, USA

a r t i c l e

i n f o

Article history:
Received 22 April 2016
Received in revised form 15 September
2016
Accepted 18 September 2016
Available online 19 September 2016
This manuscript was handled by Tim R.
McVicar, Editor-in-Chief, with the assistance
of Dawen Yang, Associate Editor
Keywords:
Water budget
Green infrastructure
Evapotranspiration
Stormwater
SWMM

a b s t r a c t
This paper presents a study of the potential for green infrastructure (GI) to restore the predevelopment
hydrologic cycle in a semi-arid urban catchment. Simulations of stormwater runoff from a 0.11-km2
urban catchment in Salt Lake City, Utah, USA for predeveloped (Natural Hydrology, NH), developed
(Baseline, BL), and developed with GI (Green Infrastructure, GI) conditions were executed for a oneyear period. The study was repeated for a relatively dry year, wet year, and an average year based on precipitation amounts in the year. Bioretention and green roofs were chosen for the GI plan. Results showed
that the water budget of the catchment with the GI plan implemented more closely matches the NH
water budget compared to the BL scenario, for all three years (dry, wet, average). The BL and GI scenarios
showed more significant modifications to the water budget than what has been found by studies in
humid climates. Compared to the BL condition, GI annually reduces surface runoff by 35%, 45%, and
43% and restores evapotranspiration by 18%, 19%, and 25% for the dry, average, wet years, respectively.
Based on the introduced water budget restoration coefficient (WBRC), the water budget of the study
catchment was restored by the GI plan to 90%, 90%, and 82% of the predevelopment state in the dry, average, and wet years, respectively. By comparing the WBRC estimated for other studies, it is further inferred
that the water budget is more significantly affected by development and GI restoration in semi-arid than
humid climates, but the differences lessen as the precipitation amount increases.
2016 Elsevier B.V. All rights reserved.

1. Introduction
Urbanization alters the water budget due to the removal of
native vegetation, alteration and compaction of soils, building of
impervious surfaces, changes in water use, and introduction of
water diversions (Whitford et al., 2001; Pauleit et al., 2005;
Shuster et al., 2005; Claessens et al., 2006; Powell et al., 2008;
Scalenghe and Marsan, 2009; Jacobson, 2011; Guan et al., 2016;
Yao et al., 2016). Such changes lead to a complicated mixture of
modifications to the hydrologic cycle across a range of spatial
scales. Surface runoff in most watersheds is observed to increase
with urbanization (Rose and Peters, 2001; Weng, 2001; Lee and
Heaney, 2003; Haase, 2009; Boggs and Sun, 2011; Zhang et al.,
2013; Wu, 2015), while changes to other water budget components have been reported to typically be reduced, such as precipitation (Rosenfeld, 2000; Shepherd, 2006; Kaufmann et al., 2007;
Hand and Shepherd, 2009), groundwater recharge (Lerner, 1990,
2002; Foster et al., 1994; Rose and Peters, 2001; Zhang and
Kennedy, 2006; He et al., 2009; Jeppesen et al., 2011; He and
Corresponding author.
E-mail address: youcan.feng@gmail.com (Y. Feng).
http://dx.doi.org/10.1016/j.jhydrol.2016.09.044
0022-1694/ 2016 Elsevier B.V. All rights reserved.

Hogue, 2012; Hibbs and Sharp, 2012; Barron et al., 2013), baseflow
(Brun and Band, 2000; White and Greer, 2006; Jacobson, 2011; Nie
et al., 2011), and evapotranspiration (ET) (Oke, 1979; Grimmond
and Oke, 1986; Balling and Brazel, 1987; Dow and DeWalle,
2000a; Rose and Peters, 2001; Dimoudi and Nikolopoulou, 2003;
Gober et al., 2009; Haase, 2009; Jeppesen et al., 2011; Ramier
et al., 2011; Shields and Tague, 2012; Wijesekara et al., 2012;
Barron et al., 2013; Bijoor et al., 2014; Gwenzi and Nyamadzawo,
2014). However, the magnitude and direction of the water budget
component modifications are difficult to predict given the complexities of the urban system (Burian and Pomeroy, 2010).
Such alterations to the hydrologic cycle can negatively impact
the urban ecosystem and downstream areas. Increased runoff, for
example, is directly connected to a wide array of environmental
stressors (Hasse and Lathrop, 2003), such as flood risk (Liu et al.,
2006; Haase, 2009; Du et al., 2012; Rutland and Dukes, 2012;
Wijesekara et al., 2012), sediment erosion and transport (Nie
et al., 2011), stream quality degradation (Interlandi and Crockett,
2003; Foley et al., 2005; Astaraie-Imani et al., 2012; Zgheib et al.,
2012), aquifer pollution (Lerner and Barrett, 1996; Chisala and
Lerner, 2008; Hibbs and Sharp, 2012), waterborne diseases
(Vrsmarty et al., 2000; Narain, 2012), acidification of water

Y. Feng et al. / Journal of Hydrology 542 (2016) 744755

bodies (Kelly et al., 2011; Xiao et al., 2012), and aquatic species loss
(Gillies et al., 2003). To respond to these changes and uncertainties,
quantifying urban impacts on spatiotemporal water budget
responses remains an area of great need, especially in the planning
and design that guide the configuration and operation of stormwater management systems in cities.
To develop plans to mitigate the adverse ecosystem impacts
due to urbanization, a new international trend of pursuing the goal
of near-natural stormwater management has emerged (Gbel
et al., 2004; Gbel and Coldewey, 2013). The concept of nearnatural aims to replicate the quasi-natural local water balance so
as to preserve the local ecosystems integrity (Kebler et al., 2012;
Walsh et al., 2016). This trend is consistent with the efforts of using
green infrastructure (GI) to restore the predevelopment hydrologic
cycle, promoted by the United States Environmental Protection
Agency (EPA) (USEPA, 2000), and aligns with sustainable design
goals incorporated into the EnvisionTM sustainable infrastructure
rating system (Envision) (http://sustainableinfrastructure.org/, last
accessed on July 2nd, 2016) and the Leadership in Energy and Environmental Design (LEED) rating system (http://www.usgbc.org/
leed, last accessed on July 2nd, 2016).
For stormwater management, GI is designed to reduce the
quantity and improve the quality of runoff by adding storage (often
pervious) with the capacity to capture, evapotranspire, and infiltrate stormwater. Compared to predevelopment landscapes, the
vertical storage capacity of GI in cities compensates for the lost
area of natural surface storage. By expanding storage in the vertical
direction and incorporating water conservation, GI seeks to efficiently (in terms of land area) achieve stormwater runoff management and environmental benefits of natural landscapes. Most GIrelated studies and applications have focused on runoff (Booth
et al., 2004; Culbertson and Hutchinson, 2004; Simpson, 2007;
Brown et al., 2009; Li et al., 2009; Alfredo et al., 2010; Burian
and Pomeroy, 2010; Fassman and Blackbourn, 2010; Voyde et al.,
2010a; DeBusk et al., 2011; Petrucci et al., 2013; Trinh and Chui,
2013; Ellis and Viavattene, 2014; Loperfido et al., 2014;
Zahmatkesh et al., 2014; Ambrose and Winfrey, 2015; Jarden
et al., 2016; Guan et al., 2015a, 2015b; Wella-Hewage et al.,
2016) and groundwater recharge (Shuster et al., 2007; Moglia
et al., 2010; Kidmose et al., 2015). A critical, yet often overlooked,
water budget component addressed by GI is ET, because (1) ET controls the amount of available water for percolation (Ellis, 2013),
and therefore affects runoff volumes and peak rates (Boggs and
Sun, 2011; Sun et al., 2013; Walsh et al., 2016; Wong and Jim,
2015; Yang et al., 2016); (2) ET affects the urban heat island
(UHI) intensity (Sailor, 1995; Alexandri and Jones, 2008; USEPA,
2008; Gober et al., 2009, 2012; Shashua-Bar et al., 2009;
Krayenhoff and Voogt, 2010), and in turn the cooling costs and
related energy consumption (Barrio, 1998; Kumar and Kaushik,
2005; Lazzarin et al., 2005; Levallius, 2005; Getter and Rowe,
2006; Takebayashi and Moriyama, 2007; Alexandri and Jones,
2008; Mitchell et al., 2008; USEPA, 2008; Fioretti et al., 2010;
Gartland, 2010; Ouldboukhitine et al., 2011; Saadatian et al.,
2013); (3) ET from green roofs generates cool air, which may give
rise to strengthened street canyon flow and improved air quality
near roads (Baik et al., 2012); (4) green spaces (increasing ET) provide space for plants and increase carbon sinks, especially in arid
regions (Sun et al., 2011), and improves biodiversity (Currie,
1991); and (5) ET enhances atmospheric moisture, which may lead
to enhanced precipitation under certain circumstances (e.g., semiarid climates) (Eltahir, 1998; Schr et al., 1999; Shepherd and
Burian, 2003; Koster et al., 2004; Burian and Shepherd, 2005;
Jung et al., 2010; Seneviratne et al., 2010; Aragao, 2012;
Spracklen et al., 2012; Taylor et al., 2012).
In accordance with the concept of integrated ecosystem management/stewardship (Falkenmark and Rockstrm, 2004; Chapin

745

et al., 2009), there remains a need to evaluate the effect of GI in

terms of recreating the near-natural water budget (Burns et al.,
2012; Fletcher et al., 2013; Olszewski and Davis, 2013). Restoring
ET and infiltration to predevelopment levels is a critical part of this
goal, and is now represented in stormwater management design
criteria specified in Envision and LEED. Therefore, the objective of
this paper is to evaluate the potential of GI to restore the water
budget of a developed area to an estimated predevelopment level
in a semi-arid region.

2. Methods
2.1. Modeling framework
EPA SWMM 5.0.022 was selected as the modeling platform for
this study, as it is able to simulate a water budget for both natural
and urban environments, and it is one of the few models with the
flexibility to simulate multiple types of GI (Elliott and Trowsdale,
2007). The bioretention unit in EPA SWMM 5.0.022 was used to
model both bioretention systems and green roofs for this study.
The bioretention model in SWMM is composed of surface, soil,
storage, and drainage layers. The storage layer was assumed to represent the drainage mat layer for green roofs. For both bioretention
and green roofs, the layers were parameterized with appropriate
hydraulic properties following guidance in the SWMM Users Manual (https://www.epa.gov/water-research/storm-water-management-model-swmm#downloads, last accessed on July 2nd,
2016). Compared to GI, landscape elements have surface and soil
layers, with the latter represented by the unsaturated layer of
the aquifer component of SWMM. The outflows from GI are specified in SWMM to drain onto landscapes or into storm drains.
The Penman-Monteith equation (Monteith, 1965) was used to
estimate potential ET (PET) rates, following the standard practice
(Kingston et al., 2009; Sherwood and Fu, 2014; Thompson et al.,
2014). Parameters like albedo and surface resistances were set to
represent the different GI and land surface covers (Feng and
Burian, 2016). The water stress coefficient was set to convert PET
rates to actual ET (ETa) rates using the equation from the Food
and Agriculture Organization of the United Nations, Paper 56
(FAO-56) (Allen et al., 1998; DiGiovanni et al., 2013). The moisture
balance simulated by SWMM was used to calculate the water
stress coefficient (Feng and Burian, 2016). Hourly PET and ETa rates
of six types of land covers including ponding water, bioretention,
green roofs, turf landscapes, deciduous trees, and coniferous trees
were estimated separately.

2.2. Study site

A small urban catchment (0.11 km2) located on the campus of
the University of Utah in northeast Salt Lake City (SLC), Utah, U.S.
A. was chosen for this study (Fig. 1). SLC has a semi-arid climate
(Bailey, 1979; Eubank and Brough, 1979; Bair, 1992; Russell and
Cohn, 2012). From 1981 to 2010, the SLC average annual precipitation is 409 mm and the average annual air temperature is 11.5 C
(NOAA, 2013). From the Web Soil Survey (http://websoilsurvey.
sc.egov.usda.gov/App/HomePage.htm, accessed 03/17/2015) operated by U.S. Department of Agriculture (USDA)s Natural Resources
Conservation Service (NRCS), the primary soil type of the catchment is Bingham gravelly loam. Its hydraulic conductivity is
approximately 0.899 cm/h; and its porosity is 0.459, while its wilting point and field capacity are 0.148 and 0.288, respectively
(Merrell, 2013). The water table was measured as 38.26 m below
the land surface by a U.S. Geological Survey (USGS) groundwater
station near the study site (U.S. Geological Survey, 2015). The aver-

746

Y. Feng et al. / Journal of Hydrology 542 (2016) 744755

Fig. 1. Delineated study catchment.

age thickness of the local valley fill aquifer was estimated as 823 m
(Arnow and Mattick, 1968).
ET contribution from the deep groundwater was not considered
in this study. To confirm this, a simulation was executed with deep
percolation in SWMM and it was shown that this would generate a
very small (1.07 mm) difference in ETa for the predevelopment
condition for the year 2014 (simulated by EPA SWMM 5.1.007).
This is equivalent to 0.22% of the total annual precipitation; thus,
it was assumed negligible in the study.
Meteorological data collected at 5-min intervals from two
weather stations operated by the Department of Atmospheric
Science at the University of Utah were downloaded for this study
from the Mesowest website (http://mesowest.utah.edu/, accessed
on 03/17/2015). The Mountain Met (MTMET) weather station
(40450 6000 N, 111490 4200 W) located within the study catchment
was used to represent meteorological conditions from July 3,
2012, to December 31, 2014. Meteorological data before that period (starting in 2011) was obtained from the nearby William
Browning Building (WBB) weather station (40450 5800 N,
111500 5100 W), which is 1.66 km from the MTMET station. Except
the precipitation, other raw data were summed up to hourly
amounts.
Spatial distribution and fractions of current land cover within
the watershed were determined by manually interpreting 1-footresolution orthophotography images downloaded from Utah Automated Geographic Reference Center (AGRC, http://gis.utah.gov/,
accessed 06/08/2015) and verified by site visits. The average building height in the catchment was estimated at 10.66 m based on
1 m and 1.25 m horizontal spatial resolution airborne Lidar data
(collected in 2006) acquired from Utah AGRC (http://gis.utah.gov/
, accessed 06/08/2015). Similarly, the average heights of the deciduous and coniferous trees were estimated as 12.70 m and 14.35 m,
respectively. The height information was used to convert wind
speed measurements to the canopy levels corresponding to green
roofs and trees, based on a logarithmic profile (Allen et al., 1998).
A storm drainage system serves the catchment and directs runoff into Red Butte Creek. The drainage catchment was delineated
and subdivided based on terrain, locations of storm drain inlets,
and other local features (e.g., curb and gutters). Several site visits

were made to identify the locations of storm drain inlets and outfalls. A 2150 Area Velocity Flow Module (Teledyne Isco, USA) was
installed in May 2014 in the storm drain at the outlet of the catchment to measure the flow rate in one-minute increments.
The SWMM model corresponding to the baseline (developed)
condition was manually calibrated for five rain events measured
in May 2014 (Fig. 2). Width, slope, imperviousness percentages,
Mannings roughness coefficients, depression storage, and infiltration parameters (Green-Ampt method) of subcatchments, and size,
length, and slope of stormwater pipelines were adjusted during the
calibration. The coefficients of determination (R2) for the five calibration events ranged from 0.371 to 0.876, while the root of mean
square errors (RMSE) ranged from 2.57 L/s to 10.56 L/s (Fig. 2). The
accuracy of the calibrated model meets the required level for this
study to evaluate the changes of the water budget due to GI
applications.
2.3. Scenarios
Three scenarios were simulated: baseline (BL), green infrastructure (GI), and natural hydrology (NH). In the NH scenario, the
catchment was modeled as being covered with native grasslands
like wheatgrass and bluegrass (Ehleringer et al., 1992), as the open
meadow is the dominant landscape at the foothill environment
next to the study catchment. Green roofs and bioretention were
implemented in the catchment for the GI scenario. The numbers
and the sizes of green roofs and bioretention were determined by
designing them to reduce the 1-year stormwater runoff volume
by 80%, which corresponds to an amount consistent with goals
for stormwater quality management plans (Horner et al., 2004;
Sullivan et al., 2010). SWMM simulations were executed with
green roofs and bioretention iteratively added to the BL scenario
until 80% runoff reduction was achieved. The green roofs were
placed on flat roofs and bioretention was placed on open ground
areas. These units were configured to match the existing designs
and recommendations for the climate of SLC (Houdeshel et al.,
2012; Houdeshel and Pomeroy, 2014).
The bioretention units were parameterized in SWMM to match
instrumented test units on the University of Utah campus (Orr,

Y. Feng et al. / Journal of Hydrology 542 (2016) 744755

747

Fig. 2. Measured and modeled outflow hydrograph for calibration events.

2013) with a 0.6 m soil layer (porosity: 0.43, field capacity: 0.21,
conductivity: 38.83 mm/h) and a 0.6 m gravel layer (void ratio:
0.6). Bottoms were lined with underdrains with a 15 mm/h drain
rate. Green roofs were modeled to mimic an existing green roof
on the Marriott Library on the University of Utah campus, which
has a 254 mm medium layer (porosity: 0.58, field capacity: 0.48,
conductivity: 81.28 mm/h) and a 25.4 mm drainage mat (void
ratio: 0.6) (Feng, 2016). Green roof underdrains were assumed to
have the same drainage rate as the bioretention units.
The hydrologic performance of GI may greatly depend on the
local weather conditions (Nawaz et al., 2015). Three years were
selected for the simulations used in this study, based on the availability of the data and the relative magnitudes compared to the
annual average precipitation (409 mm). It is difficult to find three
years having total precipitation depths perfectly distributed
around the annual average while not having missing weather
observations at the study site. Therefore, in spite of the precipitation depth being close to the annual average, 2012 (371 mm)
was assumed to represent a dry year for this study. Although the
precipitation depth of 2014 is higher than the annual average
(482 mm), it was assumed to be the average year for this study.
And 2011 (688 mm) was assumed to be the wet year for this study
(Fig. 3). Initial soil moisture was assumed to be zero for all three
scenarios to exclude its influence on the comparison. SWMM
simulations were independently executed for the calendar years
(January 1 to December 31) for each of these three years at a

Fig. 3. Cumulative precipitation depths.

one-minute time step. Annual water budgets were summed and

compared among different scenarios and different years. Results
of the average precipitation year (2014) were then used to further
explore the water budget variations among different scenarios at
the monthly, daily, and hourly time scales.

748

Y. Feng et al. / Journal of Hydrology 542 (2016) 744755

3. Results
3.1. Annual water budgets
Across all three studied years, the BL scenario always had the
largest stormwater discharge volume (Fig. 4). Specifically, the surface runoff increased from 41 mm in the NH scenario to 112 mm in
the BL scenario (171%) in the dry year, from 20 mm in the NH scenario to 104 mm in the BL scenario (416%) in the average year, and
from 30 mm to 152 mm (412%) in the wet year. The GI scenario,
however, had a closer surface discharge amount to the NH scenario, where surface discharge is the sum of surface runoff and the discharge portion from the underdrains of GI units directed into the
stormwater drainage system. Compared to the NH scenario, the
surface discharge of GI scenario increased by 31 mm (76%) in the
dry year, 37 mm (182%) in the average year, and 56 mm (191%)
in the wet year. Compared to the BL scenario, GI reduced surface
runoff by 39 mm (35%) in the dry year, 47 mm (45%) in the average
year, and 66 mm (43%) in the wet year.
ETa was the dominant component of the water budget, as it
accounted for 6063%, 7175%, and 8295% of the water budget
for the BL, GI, and NH scenarios, respectively (Fig. 4). The BL scenario had the lowest ETa amounts compared to the other two scenarios (Fig. 4), as expected. Specifically, ETa decreased from
315 mm in the NH scenario to 232 mm in the BL scenario (26%)
in the dry year, from 396 mm in the NH scenario to 287 mm in
the BL scenario (28%) in the average year, and from 650 mm to
416 mm (36%) in the wet year. The GI scenario, had closer ETa
amounts compared to NH scenario, with ETa being lower than
the NH scenario by 40 mm (13%) in the dry year, 54 mm (14%) in
the average year, and 130 mm (20%) in the wet year. Compared
to the BL scenario, GI restored annual ETa amounts by 43 mm
(18%), 55 mm (19%), and 104 mm (25%) for the dry, average, and
wet years, respectively.
Overall, development raised surface runoff annually by 171%,
416%, and 412% in dry, average and wet years, and reduced ETa
amounts by 26%, 28%, and 36%, respectively. GI reduced annual
surface runoff by 35%, 45%, and 43% and restored annual ETa
amounts by 18%, 19%, and 25% in the dry, average, and wet years,
respectively, compared to the BL scenario.
After retention in the surface storage (surface wetting and GI
storage) and discharge as surface runoff, the remaining infiltrated

Fig. 4. Water budgets of the baseline scenario (BL), the green infrastructure
scenario (GI), and the natural hydrology scenario (NH) in dry (2012), average
(2014), and wet (2011) years. (For interpretation of the references to color in this
figure legend, the reader is referred to the web version of this article.)

water had three potential destinations: storage in unsaturated soil,

deep groundwater percolation, and release to atmosphere via ET.
Most of the water infiltrated into the soil was lost through ET
(Fig. 4). Very small amounts (65%) percolated into the deeper soil.
Although soil moisture is often represented in the form of volumetric or mass fractions, it is represented in this study as the water
depth stored in the soil layer. The average year shows the highest
soil moisture in each scenario compared to the other two years.
This might be due to the higher precipitation contributions (mostly
snow) near the end of 2014 when ET is relatively small (Fig. 3).
3.2. Monthly water budget variation
The monthly water budgets between different scenarios were
also compared to evaluate the monthly soil moisture balance for
the average year (2014) (Fig. 5). Of note is that SWMM accounts
for moisture stored within GI as surface storage by default. Therefore, when the precipitation exceeds the sum of the losses (percolated water, ETa, and discharge), both the storage space of GI and
soil moisture storage underneath GI may gain moisture. In this
regard, the soil storage of BL scenario was gaining moisture in January, February, March, April, August, September, November, and
December, while losing stored moisture during the rest of the year.
The soil storage of NH scenario was gaining moisture in January,
February, July, September, November, and December. The soil storage in the GI scenario was gaining in January, February, March,
August, September, November, and December.
The extent of surface storage of the GI scenario increased during
most of the months when the precipitation exceeded the sum of
the losses, while vice versa happened when the precipitation was
lower than the sum of surface discharge and ETa. The only exception was August when precipitation was higher than losses for
the first time in four months, but the surface storage was less than
the previous month.
Moisture availability determines ETa rates. Larger surface storage created extra space for GI to store moisture through dry
months in the GI scenario, compared to the other two scenarios.
In the relatively warm months of May and June, the GI scenario
had the largest ETa amounts. This happened after the soil moisture
storage of the NH scenario decreased from March onward, when
the GI scenario had extra moisture supply in the surface storage.

Fig. 5. Simulated monthly water budgets for the average water year (2014). The
three stacked columns of each month from left to right represent BL, GI, and NH
scenarios, respectively. (Note: percolated water volumes are too small to be seen at
the bottom of each stack for most months.)

Y. Feng et al. / Journal of Hydrology 542 (2016) 744755

In the relatively cold month of November, the GI scenario also had

higher ETa amounts than the NH scenario, which happened after a
dry October when the soil moisture of the NH scenario was mostly
drained.
Except November, the NH scenario had the largest ETa amounts
within months when three scenarios all received larger precipitation than the sum of the losses. The BL scenario had the lowest
ETa amounts in most months except June, July, and November,
when plants with higher ET capacity in GI or BL scenarios experienced water stress due to the loss of the moisture storage.
3.3. Daily and hourly soil moisture balance variations
The water budget simulations for the average year were further
analyzed at the daily and hourly scales, to explore the mechanism
of GI and the potential to restore the catchments water budget
after precipitation events. A period of 49 days (June 16th to August
3rd, 2014) experiencing two major rain events was selected for
evaluation (Fig. 6). The BL scenario retained the least total moisture
(sum of soil and surface storages) compared to the other two scenarios. The soil moisture storage of the GI scenario was also less
than that of the NH scenario. But by adding the surface storage
capacity of the GI to the soil moisture storage, the sum approximately matched the soil moisture storage of the NH scenario, both
in magnitude and temporal pattern.

749

Correspondingly, the BL scenario had the lowest ETa amounts

because much of the rainfall was converted to surface discharge.
The GI scenario, however, produced a temporal pattern of ETa that
more closely matched the NH scenario in magnitude and variation.
Both GI and NH scenarios have much lower surface discharge
amounts compared to the BL scenario.
Similar relationships among the three scenarios were found
when results were summed to the hourly scale (Fig. 7). Adding
the surface storage, the GI scenario had a comparatively close total
moisture storage to the NH scenario, in terms of magnitudes and
temporal variations. The BL scenario had the least total stored rainwater. Notably, the NH scenario had the largest ETa rates during
the second day (June 18th), followed by the GI scenario and then
the BL scenario. Also, the magnitude of cumulative ETa volumes
in all three scenarios are much less than the remained water in
the storages.
4. Discussion
4.1. The role of ET within the water budget
As observed in the results, ETa is the dominant component of
the water budget in this semi-arid study site, which is consistent
with previous studies. In the NH scenario, ETa accounts for 82
95% of the total water budget, which is higher than the averages

Fig. 6. Daily soil moisture storage, surface storage, ETa, and surface discharge for the average precipitation year (2014).

750

Y. Feng et al. / Journal of Hydrology 542 (2016) 744755

Fig. 7. Hourly surface storage, soil moisture storage, and ETa rates simulated by the updated SWMM from June 17th to June 18th in the average precipitation year (2014).

of the previous studies in natural watersheds that were mostly

conducted in wetter climates (Fig. 8). However, in a similar semiarid climate with an annual precipitation of 331477 mm, ETa
was estimated about twice as the annual precipitation on land
(Mariotti et al., 2002).
In the BL scenario, the ETa fraction of the total water budget is
6063%. It is hard to directly compare this result with other studies, as the range of urban ETa ratios often varies widely due to a
complex combination of local factors, including climate, imported
water, urban heat island effect, plant covers, city density, the
design of buildings and street canyons, etc. But the value falls
within the range of the previous studies (Fig. 9).

In the GI scenario, annual ETa amounts are 7175% of the total

water budget, which, similar to the NH scenario, is higher than
other estimates in the wetter climates. For example, 32% is the
fraction of ETa within the total water budget after GI is applied in
Malm, Sweden (Villarreal et al., 2004), 71% after applying the
modified household infiltration-based infrastructure and retention
ponds in Trier, Germany (Keler et al., 2012), 44% in an infiltrationbased GI applied catchment in Iowa City, Iowa (Holman-Dodds
et al., 2003), and 48% predicted for a GI scenario in Cincinnati,
OH (Chenevey, 2013).
To sum up, ETa is the dominant component of the water budget
in the climate of the study area. This is consistent with a previous

Fig. 8. A comparison of ETa fractions of predeveloped site water budgets for a range of climates.

Y. Feng et al. / Journal of Hydrology 542 (2016) 744755

751

Fig. 9. A comparison of ETa fractions of developed site water budgets for a range of study sites/climates.

study (Trinh and Chui, 2013), in which ETa was also found to be the
greatest contributing factor for all scenarios including predevelopment, postdevelopment, and a restoration scenario using green
roofs and bioretentions, even if ETa was only measured as 5% of
its own total water balance of a plot-scale green roof. However,
the ETa fractions of the water budget in this study are mostly
higher than other studies in wetter climates. ETa accounts for
6063%, 7175%, and 8295% of the water budgets in the BL, GI,
and NH scenarios, which also matches other findings that an
increase in vegetation coverage tends to raise the ETa fraction of
the urban water budget (Yao et al., 2014; Yeakley, 2014).
4.2. Water budget changes due to development
Although multiple factors have a mixed influence in increasing
or decreasing urban ETa ratios, the overall effect of urbanization is
to typically decrease ETa amounts due to vegetation reduction. For
example, due to urbanization, ETa was estimated to drop by 31%
(compared to annual average) by a study of 51 eastern U.S. watersheds (Dow and DeWalle, 2000b), 25% (compared to predevelopment condition) by a study in Leipzig, Germany (Haase, 2009),
22% (compared to predevelopment condition) by a study in the
Mill Creek watershed in Cincinnati, OH (Chenevey, 2013), and
23% (compared to predevelopment condition) by a study in the
Qinhuai river basin, China (Hao et al., 2015). In the present study,
the BL scenario experienced a decrease of ETa by 2636% from the
NH scenario, which is slightly higher than estimates by the other
published studies.
Furthermore, the BL surface runoff in this study was 171412%
greater than the surface runoff in the NH scenario. This increase is
greater than most studies in wetter climates, like the 182% increase
found in Leipzig, Germany (Haase, 2009) and the 128% increase in
Cincinnati, Ohio (Chenevey, 2013). This may indicate that the
water budget may experience more severe changes after development in semi-arid climates than in wetter climates.
4.3. Water budget restoration by GI
The potential to restore the natural water budget depends on
the type of GI selected. Different GI types can target different parts

of the water budget. For example, in an Iowa City, Iowa study

(Holman-Dodds et al., 2003), infiltration-based GI was estimated
to raise a catchments groundwater recharge by 100%, making it
close to that of predevelopment conditions. However, the GI in
the Iowa City study was only able to increase ETa by 8% compared
to the developed condition, which was an increase of 3% of the
total water budget (from 41% to 44%). In a study of a catchment
in Cincinnati, OH (Chenevey, 2013), a GI plan incorporating rain
barrels, porous pavement, and green roofs was estimated to raise
ETa amounts by 19% compared to the developed condition, but
the ETa fraction (48%) of the total water budget did not show a significant change compared to the developed scenario without GI
(40%).
In the present study, the GI increases annual ETa amounts by
1825% compared to the BL scenario, which is 1115% of the water
budget. As the GI selected were bioretention and green roofs,
which both have significant PET, the results suggest GI designs
can effectively raise the catchments ETa close to the predeveloped
levels. This is also supported by a study in Malm, Sweden, which
used green roofs, open channels, and detention ponds that have
even higher ETa potential. The GI was found to raise the catchments annual ETa amounts by 74% compared to the developed
condition (Villarreal et al., 2004).
Overall in the present study, the GI scenario showed a water
budget closer to the NH scenario than the BL scenario for all types
of precipitation years (dry, average, wet). The GI scenario reduces
annual runoff by 3545%, which is higher than other studies, like
19% in Malm, Sweden (Villarreal et al., 2004), 33% in Cincinnati,
OH (Chenevey, 2013), or 35% in Iowa City, IA (Holman-Dodds
et al., 2003). This indicates that GI can more significantly affect
the water budget in a semi-arid climate than wetter climates.
4.4. Storage space of GI
Within storage space in soil and GI, moisture collected in moist
months (like December to February) can be stored and then used
by ET in the subsequent drier months (Fig. 5). ETa rates determine
how fast the storage capacity will be recovered for the next precipitation event (Voyde et al., 2010b; Krebs et al., 2016). It might be
necessary to point out that the soil moisture storage and surface

752

Y. Feng et al. / Journal of Hydrology 542 (2016) 744755

storage terms are used as state variables here to represent the

remaining moisture levels cut off by the end of each time step,
which could not reflect the dynamics of moisture variations within
the time step. Given the monthly time step (Fig. 5), for instance,
zero soil moisture storage of the NH scenario in May and June
could not reflect the inner-month soil moisture variations, but
should stand for the remaining soil moisture levels at the end of
the month. In fact, the moisture levels of the NH scenario experience rises and falls within May and June, respectively, but their
moisture levels at the end of each month are zero. Still speaking
of the NH scenario, the source of the water to support ETa of May
comes from the rainfall in May and the remaining moisture collected in April; while the ETa of June is entirely supplied by the
rainfall in June (Fig. 5). In reality, the moisture contents contained
within the plants and the air and the landscape irrigation could be
extra water sources for ET, which are not considered in this study.
During the months having enough incoming precipitation in the
study area (January, February, September) or when antecedent soil
moisture storage is sufficient (March, April, August, October), the
natural plants with highest PET, simulated by the NH scenario,
generate higher ETa amounts than the BL scenario with low vegetation cover and the GI scenario with the lower PET (Fig. 5). On
the other hand, during the months when soil moisture storage
was insufficient to support high ET rates (May, June, and November), the storage layers of GI, especially the gravel layer in bioretention and the drainage mat in green roofs, provide extra stored
moisture to maintain ETa amounts even higher than the NH plants
which experienced water stress (Fig. 5).
GI can add storage space in urban catchments to approximately
match the storage capacity of the predevelopment condition
(Figs. 6 and 7). The extra storage of GI can create ETa patterns close
to the predevelopment condition (NH scenario) at the event scale.
Due to the lack of storage space in the developed condition (BL scenario), the cumulative ETa amounts were 15% and 27% lower than
GI and NH scenarios after 49 days (Fig. 6), and 14% and 41% lower
than GI and NH scenarios after 2 days (Fig. 7). This indicates the
importance of the storage space in determining the performance
of GI.
To sum up, in restoring the water budget, one of the most
important features of GI is providing concentrated storage to
match the storage of the predevelopment condition, as GI can significantly raise the annual ETa amounts close to the level of the natural landscapes although the formers ET capacity tends to be
lower than the latter (Figs. 57).
4.5. Land use change
The pervious areas in the BL, GI, and NH scenarios occupy 41%,
54%, and 95% of the corresponding total catchment areas, respectively. With the replacement of the impervious surfaces by pervious areas including bioretention areas and green roofs in GI and
NH scenarios, the water budgets change accordingly in the average
precipitation year (2014) (Table 1). From the BL to the GI scenario,
an 11% increase of ETa and a 10% decrease of runoff of the total
water budget, however, only require 13% impervious surfaces
replaced within the total catchment. Therefore, replacing 1% of
impervious areas by GI within the total catchment may bring a
0.87% increase of ETa and a 0.74% decrease of runoff from the
developed condition.
If assuming the natural landscape as one type of GI, a 23%
increase of ETa and a 17% decrease of runoff within the total water
budget require 54% impervious surfaces replaced by natural landscape within the total catchment. This means that replacing 1%
impervious areas by natural landscape within the total catchment
may bring a 0.42% increase of ETa and a 0.32% decrease of runoff.
This indicates that to gain the extra restoration effect by choosing

Table 1
ETa and runoff restorations created by the pervious areas changes in the average
precipitation year (2014).
Transitions

Percent of replaced
impervious area of
the total catchment
(%)

Percent of ETa
increase of the
total water
budget (%)

Percent of runoff
decrease of the
total water budget
(%)

BL -> NH
BL -> GI

54
13

23
11

17
10

the traditional landscape may require retrofitting a much larger

built city space. The retrofit of the water budget in terms of choosing to promote ET or groundwater recharge also depends on the
designs of each GI type (Houdeshel et al., 2012), which may affect
the restoration efficiency when seeking to recreate a local hydrology. Future studies are needed to address this uncertainty and further explore the impacts of GI designs with different hydraulic and
ecological considerations on restoring the water budget for different local environmental conditions.
4.6. Water budget restoration coefficient
A water budget restoration coefficient (WBRC) is proposed to
evaluate the overall restoration effect of different GI plans in
restoring the water budget. The WBRC is calculated as:

WBRC 1 

X


f i jf 0i  f i j ;

where f i is the percent of the water budget component in predevel0

oped condition (NH scenario in this study), and f i is the percent of
the water budget component in the studied condition (e.g., BL, GI).
The range of this coefficient should be from zero to less than or
equal to one. Larger values of the WBRC suggest the studied water
budget is closer to the predevelopment water budget, and that the
infrastructure or land covers have a high restoration effect.
Surface runoff, ET, percolation, soil moisture storage, and surface storage were included in the calculation of WBRC in this study.
The WBRC of the BL, GI, and NH scenarios for the dry, average and
wet precipitation years were calculated using Eq. (1), assuming the
NH scenario was the predeveloped condition (Table 2). The GI scenario has coefficients more close to one than the BL scenario in any
water year, which is consistent with the previously reported
results indicating GIs potential to restore the natural hydrology
of the catchment. More generally, the coefficient can be used to
determine the degree of the water budget restoration by GI. As
an example, the WBRC was calculated for three other studies based
on their published data (Table 2). In the GI scenario, WBRC of the
dry and average years in the present study were higher than WBRC
of all other studies, which indicates that the GI applications in
these two cases reached better restoration effects. From BL scenario to GI scenario, WBRC of the dry, average, and wet years of
the present study have increased by 14%, 13%, and 22%, respectively, compared to 4% and 9% in the latter two other studies,
respectively. This further supports the previous observation that
GI has a higher potential to restore the water budget in the semiTable 2
The water budget restoration coefficients of three scenarios in three water years and
estimated values for other studies.

Dry Year
Average Year
Wet Year
Villarreal et al. (2004)
Chenevey (2013)
Holman-Dodds et al. (2003)

BL

GI

NH

0.79
0.80
0.67

0.84
0.81

0.90
0.90
0.82
0.85
0.87
0.88

1.00
1.00
1.00
1.00
1.00
1.00

Y. Feng et al. / Journal of Hydrology 542 (2016) 744755

arid climate than the humid climate. After GI was applied, the
water budgets have achieved 90%, 90%, and 82% of the predevelopment state in the dry, average, and wet years, respectively.
WBRC can also be used to evaluate how far the urban water
budget can be changed away from the predevelopment condition
due to urbanization. In the BL scenario, the WBRC of this study is
all lower than other studies in wetter climates. This may further
support the inference as above that the water budget may vary
more severely after development, and GI may be more effective
to restore the water budget in the semi-arid climate compared
with humid climate. But when the precipitation is higher, the
restoration effect may become closer to the wetter climate, like
the case of the GI scenario in the wet year (Table 2).
5. Conclusion
The effect of GI in restoring the urban water budget to predeveloped conditions was explored in the semi-arid climate of Salt Lake
City, Utah. Three scenarios (BL, GI, and NH) were compared in three
types of precipitation years (dry, average, and wet). Water budget
variations among scenarios and water years were presented to
analyze the effect and mechanism of GI to restore the catchments
natural water budget.
The GI scenario was shown to produce a water budget closer to
the NH scenario than the BL scenario in all three types of precipitation years. ETa accounts for 6063%, 7175%, and 8295% of
the water budget for the BL, GI, and NH scenarios, respectively,
which are relatively higher than the ratios found by other studies
in wetter climates. Compared to the predeveloped condition, the
surface discharge of the developed condition is raised by 171%,
416%, and 412% in the dry, average, and wet years, respectively;
ETa was reduced by 26%, 28%, and 36%, correspondingly. Compared
to the developed condition, GI annually reduces surface runoff by
35%, 45%, and 43% and restores ETa amounts by 18%, 19%, and
25% for the dry, average, and wet years, respectively. The proposed
WBRC further supports the hypothesis that GI can restore the
urban water budget close to the natural hydrology. Based on the
proposed WBRC, the water budgets have been restored due to GI
applications by 14%, 13%, and 22% in the dry, average, and wet
years. After GI was applied, the water budgets have achieved
90%, 90%, and 82% of the predevelopment state in the dry, average,
and wet years, respectively.
Comparison with other studies indicates the water budget may
vary more severely after development, and GI may be more effective to restore the water budget in a semi-arid climate than more
humid climates. But when the precipitation amount becomes closer between the climate types, the restoration effect becomes less
different. The results and the WBRC developed by this study would
be useful in practice to support quantifying the stormwater credit
for the new stormwater management guidelines like Envision and
LEED.
Acknowledgements
This research was supported by the NSF EPSCoR grant IIA
1208732 awarded to Utah State University, as part of the State of
Utah EPSCoR Research Infrastructure Improvement Award. Additional support was provided by the Global Change and Sustainability Center at the University of Utah in collaboration with the iUTAH
EPSCoR Program. Any opinions, findings, and conclusions or recommendations expressed are those of the authors and do not necessarily reflect the views of the National Science Foundation. The
authors want to thank the Urban Water Research Group at the
University of Utah for their dedicated support to this project, and
Will McDonald for his contributions to land cover classifications

753

based on Lidar data and satellite images. The authors are also
grateful for the received insightful suggestions from anonymous
reviewers and editors.
References
Alexandri, E., Jones, P., 2008. Temperature decreases in an urban canyon due to
green walls and green roofs in diverse climates. Build. Environ. 43 (4), 480493.
Alfredo, K., Montalto, F., Goldstein, A., 2010. Observed and modeled performances of
prototype green roof test plots subjected to simulated low- and high-intensity
precipitations in a laboratory experiment. J. Hydrol. Eng. 15 (6), 444457.
Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop Evapotranspiration
Guidelines for Computing Crop Water Requirements FAO Irrigation and
Drainage Paper 56. FAO - Food and Agriculture Organization of the United
Nations, Rome.
Ambrose, R.F., Winfrey, B.K., 2015. Comparison of stormwater biofiltration systems
in Southeast Australia and Southern California. Wiley Interdiscip. Rev.: Water 2
(2), 131146.
Aragao, L.E.O.C., 2012. Environmental science: the rainforests water pump. Nature
489 (7415), 217218.
Arnow, T., Mattick, R.E., 1968. Thickness of Valley Fill in the Jordan Valley East of
Great Salt Lake, Utah. 600B.
Astaraie-Imani, M., Kapelan, Z., Fu, G., Butler, D., 2012. Assessing the combined
effects of urbanisation and climate change on the river water quality in an
integrated urban wastewater system in the UK. J. Environ. Manage. 112 (0), 19.
Baik, J.-J., Kwak, K.-H., Park, S.-B., Ryu, Y.-H., 2012. Effects of building roof greening
on air quality in street canyons. Atmos. Environ. 61 (0), 4855.
Bailey, H., 1979. Semi-arid climates: their definition and distribution. Agric. Semiarid Environ. Springer, 7397.
Bair, F.E., 1992. Weather of US Cities. Gale Research Incorporated.
Balling, R.C., Brazel, S.W., 1987. The impact of rapid urbanization on pan
evaporation in Phoenix, Arizona. J. Climatol. 7 (6), 593597.
Barrio, E.P.D., 1998. Analysis of the green roofs cooling potential in buildings.
Energy Build. 27 (2), 179193.
Barron, O.V., Barr, A.D., Donn, M.J., 2013. Effect of urbanisation on the water balance
of a catchment with shallow groundwater. J. Hydrol. 485 (0), 162176.
Bijoor, N.S., Pataki, D.E., Haver, D., Famiglietti, J.S., 2014. A comparative study of the
water budgets of lawns under three management scenarios. Urban Ecosyst. 17
(4), 10951117.
Boggs, J.L., Sun, G., 2011. Urbanization alters watershed hydrology in the Piedmont
of North Carolina. Ecohydrology 4 (2), 256264.
Booth, D.B. et al., 2004. Reviving urban streams: land use, hydrology, biology, and
human behavior. J. Am. Water Resour. Assoc. 40 (5), 13511364.
Brown, R., Hunt, W., Davis, A., Traver, R., Olszewski, J., 2009. Bioretention/
bioinfiltration performance in the mid-Atlantic. In: Proceedings of the World
Environmental and Water Resources Congress, pp. 904913.
Brun, S.E., Band, L.E., 2000. Simulating runoff behavior in an urbanizing watershed.
Comput. Environ. Urban Syst. 24 (1), 522.
Burian, S.J., Pomeroy, C.A., 2010. Urban impacts on the water cycle and potential
green infrastructure implications. In: Aitkenhead-Peterson, J., Volder, A. (Eds.),
Urban Ecosystem Ecology. American Society of Agronomy, pp. 277296.
Burian, S.J., Shepherd, J.M., 2005. Effect of urbanization on the diurnal rainfall
pattern in Houston. Hydrol. Process. 19 (5), 10891103.
Burns, M.J., Fletcher, T.D., Walsh, C.J., Ladson, A.R., Hatt, B.E., 2012. Hydrologic
shortcomings of conventional urban stormwater management and
opportunities for reform. Landscape Urban Plan. 105 (3), 230240.
Chapin III, F.S., Kofinas, G.P., Folke, C., Chapin, M.C., 2009. Principles of Ecosystem
Stewardship: Resilience-Based Natural Resource Management in a Changing
World. Springer Science & Business Media.
Chenevey, B., 2013. Development and Its Impact on the Water Balance of an Urban
Watershed M.S. Thesis. University of Cincinnati.
Chisala, B., Lerner, D., 2008. Distribution of sewer exfiltration to urban groundwater.
Proc. ICE-Water Manage. 161 (6), 333341.
Claessens, L., Hopkinson, C., Rastetter, E., Vallino, J., 2006. Effect of historical
changes in land use and climate on the water budget of an urbanizing
watershed. Water Resour. Res. 42 (3).
Culbertson, T.L., Hutchinson, S.L., 2004. Assessing Bioretention Cell Function in a
Midwest Continental Climate, 2004 ASAE Annual Meeting. American Society of
Agricultural and Biological Engineers, St. Joseph, Michigan, Ontario, Canada.
Currie, D.J., 1991. Energy and large-scale patterns of animal- and plant-species
richness. Am. Nat. 137 (1), 2749.
DeBusk, K., Hunt, W., Line, D., 2011. Bioretention outflow: does it mimic nonurban
watershed shallow interflow? J. Hydrol. Eng. 16 (3), 274279.
DiGiovanni, K., Montalto, F., Gaffin, S., Rosenzweig, C., 2013. Applicability of
classical predictive equations for the estimation of evapotranspiration from
urban green spaces: green roof results. J. Hydrol. Eng. 18 (1), 99107.
Dimoudi, A., Nikolopoulou, M., 2003. Vegetation in the urban environment:
microclimatic analysis and benefits. Energy Build. 35 (1), 6976.
Dow, C.L., DeWalle, D.R., 2000a. Trends in evaporation and Bowen Ratio on
urbanizing watersheds in eastern United States. Water Resour. Res. 36 (7),
18351843.
Dow, C.L., DeWalle, D.R., 2000b. Trends in evaporation and Bowen ratio on
urbanizing watersheds in eastern United States. Water Resour. Res. 36 (7),
18351843.

754

Y. Feng et al. / Journal of Hydrology 542 (2016) 744755

Du, J. et al., 2012. Assessing the effects of urbanization on annual runoff and flood
events using an integrated hydrological modeling system for Qinhuai River
basin, China. J. Hydrol. 464465 (0), 127139.
Ehleringer, J.R., Negus, N.C., Arnow, L.A., Arnow, T., McNulty, I.B., 1992. Red Butte
Canyon Research Natural Area: history, flora, geology, climate, and ecology. RED
52 (2).
Elliott, A.H., Trowsdale, S.A., 2007. A review of models for low impact urban
stormwater drainage. Environ. Modell. Softw. 22 (3), 394405.
Ellis, J.B., 2013. Sustainable surface water management and green infrastructure in
UK urban catchment planning. J. Environ. Plan. Manage. 56 (1), 2441.
Ellis, J.B., Viavattene, C., 2014. Sustainable urban drainage system modeling for
managing urban surface water flood risk. CLEAN Soil, Air, Water 42 (2), 153
159.
Eltahir, E.A.B., 1998. A soil moisturerainfall feedback mechanism: 1. Theory and
observations. Water Resour. Res. 34 (4), 765776.
Eubank, M.E., Brough, R.C., 1979. Mark Eubanks Utah Weather. Weatherbank.
Falkenmark, M., Rockstrm, J., 2004. Balancing Water for Humans and Nature: The
New Approach in Ecohydrology. Earthscan.
Fassman, E., Blackbourn, S., 2010. Urban runoff mitigation by a permeable
pavement system over impermeable soils. J. Hydrol. Eng. 15 (6), 475485.
Feng, Y., 2016. Potential of Stormwater Green Infrastructure to Recreate the Natural
Water Budget of a Semiarid Urban Catchment Doctoral Dissertation. The
University of Utah.
Feng, Y., Burian, S., 2016. Improving evapotranspiration mechanisms in the U.S.
Environmental Protection Agencys Storm Water Management Model. J. Hydrol.
Eng. 21 (10). http://ascelibrary.org/doi/abs/10.1061/%28ASCE%29HE.19435584.0001419.
Fioretti, R., Palla, A., Lanza, L.G., Principi, P., 2010. Green roof energy and water
related performance in the Mediterranean climate. Build. Environ. 45 (8), 1890
1904.
Fletcher, T.D., Andrieu, H., Hamel, P., 2013. Understanding, management and
modelling of urban hydrology and its consequences for receiving waters: a state
of the art. Adv. Water Resour. 51 (0), 261279.
Foley, J.A. et al., 2005. Global consequences of land use. Science 309 (5734), 570
574.
Foster, S.S.D., Morris, B.L., Lawrence, A.R., 1994. 3. Effects of urbanization on
groundwater recharge. In: Wilkinson, W.B. (Ed.), Groundwater Problems in
Urban Areas, pp. 4363.
Gartland, L., 2010. Heat Islands: Understanding and Mitigating Heat in Urban Areas.
Routledge.
Getter, K.L., Rowe, D.B., 2006. The role of extensive green roofs in sustainable
development. HortScience 41 (5), 12761285.
Gillies, R.R., Brim Box, J., Symanzik, J., Rodemaker, E.J., 2003. Effects of urbanization
on the aquatic fauna of the Line Creek watershed, Atlantaa satellite
perspective. Remote Sens. Environ. 86 (3), 411422.
Gbel, P., Coldewey, W.G., 2013. Concept for near-natural storm water control in
urban areas. Environ. Earth Sci. 70 (5), 19451950.
Gbel, P. et al., 2004. Near-natural stormwater management and its effects on the
water budget and groundwater surface in urban areas taking account of the
hydrogeological conditions. J. Hydrol. 299 (34), 267283.
Gober, P. et al., 2009. Using watered landscapes to manipulate urban heat island
effects: how much water will it take to cool Phoenix? J. Am. Plan. Assoc. 76 (1),
109121.
Gober, P. et al., 2012. Tradeoffs between water conservation and temperature
amelioration in Phoenix and Portland: implications for urban sustainability.
Urban Geogr. 33 (7), 10301054.
Grimmond, C.S.B., Oke, T.R., 1986. Urban water balance: 2. Results from a suburb of
Vancouver, British Columbia. Water Resour. Res. 22 (10), 14041412.
Guan, M., Sillanp, N., Koivusalo, H., 2015a. Modelling and assessment of
hydrological changes in a developing urban catchment. Hydrol. Process. 29
(13), 28802894.
Guan, M., Sillanp, N., Koivusalo, H., 2015b. Assessment of LID practices for
restoring pre-development runoff regime in an urbanized catchment in
southern Finland. Water Sci. Technol. 71 (10), 14851491.
Guan, M., Sillanp, N., Koivusalo, H., 2016. Storm runoff response to rainfall
pattern, magnitude and urbanization in a developing urban catchment. Hydrol.
Process. 30 (4), 543557.
Gwenzi, W., Nyamadzawo, G., 2014. Hydrological impacts of urbanization and
urban roof water harvesting in water-limited catchments: a review. Environ.
Process. 1 (4), 573593.
Haase, D., 2009. Effects of urbanisation on the water balance a long-term
trajectory. Environ. Impact Assess. Rev. 29 (4), 211219.
Hand, L.M., Shepherd, J.M., 2009. An investigation of warm-season spatial rainfall
variability in Oklahoma City: possible linkages to urbanization and prevailing
wind. J. Appl. Meteorol. Climatol. 48 (2), 251269.
Hao, L. et al., 2015. Urbanization dramatically altered the water balances of a paddy
field-dominated basin in southern China. Hydrol. Earth Syst. Sci. 19 (7), 3319
3331.
Hasse, J.E., Lathrop, R.G., 2003. Land resource impact indicators of urban sprawl.
Appl. Geogr. 23 (23), 159175.
He, B., Wang, Y., Takase, K., Mouri, G., Razafindrabe, B.N., 2009. Estimating land use
impacts on regional scale urban water balance and groundwater recharge.
Water Resour. Manage 23 (9), 18631873.
He, M., Hogue, T., 2012. Integrating hydrologic modeling and land use projections
for evaluation of hydrologic response and regional water supply impacts in
semi-arid environments. Environ. Earth Sci. 65 (6), 16711685.

Hibbs, B.J., Sharp, J.M., 2012. Hydrogeological impacts of urbanization. Environ. Eng.
Geosci. 18 (1), 324.
Holman-Dodds, J.K., Bradley, A.A., Potter, K.W., 2003. Evaluation of hydrologic
benefits of infiltration based urban storm water management. J. Am. Water
Resour. Assoc. 39 (1), 205215.
Horner, R.R., Lim, H., Burges, S.J., 2004. Hydrologic Monitoring of the Seattle UltraUrban Stormwater Management Projects: Summary of the 20002003 Water
Years. Water Resources Series Technical Report, 181.
Houdeshel, C.D., Pomeroy, C.A., Hultine, K.R., 2012. Bioretention design for xeric
climates based on ecological principles. J. Am. Water Resour. Assoc. 48 (6),
11781190.
Houdeshel, D., Pomeroy, C., 2014. Storm-water bioinfiltration as no-irrigation
landscaping alternative in semiarid climates. J. Irrig. Drainage Eng. 140 (2),
06013004.
Interlandi, S.J., Crockett, C.S., 2003. Recent water quality trends in the Schuylkill
River, Pennsylvania, USA: a preliminary assessment of the relative influences of
climate, river discharge and suburban development. Water Res. 37 (8), 1737
1748.
Jacobson, C.R., 2011. Identification and quantification of the hydrological impacts of
imperviousness in urban catchments: a review. J. Environ. Manage. 92 (6),
14381448.
Jarden, K.M., Jefferson, A.J., Grieser, J.M., 2016. Assessing the effects of catchmentscale urban green infrastructure retrofits on hydrograph characteristics. Hydrol.
Process. 10 (15), 15361550. http://onlinelibrary.wiley.com/doi/10.1002/hyp.
10736/abstract.
Jeppesen, J., Christensen, S., Ladekarl, U.L., 2011. Modelling the historical water
cycle of the Copenhagen area 18502003. J. Hydrol. 404 (34), 117129.
Jung, M. et al., 2010. Recent decline in the global land evapotranspiration trend due
to limited moisture supply. Nature 467, 951954.
Kaufmann, R.K. et al., 2007. Climate response to rapid urban growth: evidence of a
human-induced precipitation deficit. J. Clim. 20 (10), 22992306.
Kelly, R.P. et al., 2011. Mitigating local causes of ocean acidification with existing
laws. Science 332 (6033), 10361037.
Keler, S., Meyer, B., Seeling, S., Tressel, E., Krein, A., 2012. Influence of near-tonature stormwater management on the local water balance using the example
of an urban development area. Water Environ. Res. 84 (5), 441451.
Kidmose, J., Troldborg, L., Refsgaard, J.C., Bischoff, N., 2015. Coupling of a distributed
hydrological model with an urban storm water model for impact analysis of
forced infiltration. J. Hydrol. 525, 506520.
Kingston, D.G., Todd, M.C., Taylor, R.G., Thompson, J.R., Arnell, N.W., 2009.
Uncertainty in the estimation of potential evapotranspiration under climate
change. Geophys. Res. Lett. 36 (20), L20403.
Koster, R.D. et al., 2004. Regions of strong coupling between soil moisture and
precipitation. Science 305 (5687), 11381140.
Krayenhoff, E.S., Voogt, J.A., 2010. Impacts of urban albedo increase on local air
temperature at daily-annual time scales: model results and synthesis of
previous work. J. Appl. Meteorol. Climatol. 49 (8), 16341648.
Krebs, G., Kuoppamki, K., Kokkonen, T., Koivusalo, H., 2016. Simulation of green
roof test bed runoff. Hydrol. Process. 30 (2), 250262.
Kumar, R., Kaushik, S.C., 2005. Performance evaluation of green roof and shading for
thermal protection of buildings. Build. Environ. 40 (11), 15051511.
Lazzarin, R.M., Castellotti, F., Busato, F., 2005. Experimental measurements and
numerical modelling of a green roof. Energy Build. 37 (12), 12601267.
Lee, J., Heaney, J., 2003. Estimation of urban imperviousness and its impacts on
storm water systems. J. Water Resour. Plan. Manage. 129 (5), 419426.
Lerner, D.N., 1990. Groundwater recharge in urban areas. Atmos. Environ. Part B:
Urban Atmos. 24 (1), 2933.
Lerner, D.N., 2002. Identifying and quantifying urban recharge: a review. HYJO 10
(1), 143152.
Lerner, D.N., Barrett, M.H., 1996. Urban groundwater issues in the United Kingdom.
HYJO 4 (1), 8089.
Levallius, J., 2005. Green Roofs on Municipal Buildings in Lund: Modeling Potential
Environmental Benefits. Lund University, Sweden.
Li, H., Sharkey, L., Hunt, W., Davis, A., 2009. Mitigation of impervious surface
hydrology using bioretention in North Carolina and Maryland. J. Hydrol. Eng. 14
(4), 407415.
Liu, Y.B., Gebremeskel, S., De Smedt, F., Hoffmann, L., Pfister, L., 2006. Predicting
storm runoff from different land-use classes using a geographical information
system-based distributed model. Hydrol. Process. 20 (3), 533548.
Loperfido, J.V., Noe, G.B., Jarnagin, S.T., Hogan, D.M., 2014. Effects of distributed and
centralized stormwater best management practices and land cover on urban
stream hydrology at the catchment scale. J. Hydrol. 519 (Part C(0)), 25842595.
Mariotti, A., Struglia, M.V., Zeng, N., Lau, K.M., 2002. The hydrological cycle in the
Mediterranean region and implications for the water budget of the
Mediterranean Sea. J. Clim. 15 (13), 16741690.
Merrell, D.J., 2013. A Two-Dimensional Storm Drain Model for Cyber-Infrastructures
for Water Resources M.S. Thesis. Brigham Young University.
Mitchell, V.G., Cleugh, H.A., Grimmond, C.S.B., Xu, J., 2008. Linking urban water
balance and energy balance models to analyse urban design options. Hydrol.
Process. 22 (16), 28912900.
Moglia, M., Perez, P., Burn, S., 2010. Modelling an urban water system on the edge of
chaos. Environ. Modell. Softw. 25 (12), 15281538.
Monteith, J.L., 1965. Evaporation and environment. In: Fogg, G.E. (Ed.), Symposium
of the Society for Experimental Biology, The State and Movement of Water in
Living Organisms. Academic Press, Inc., NY, Swansea, pp. 205234.
Narain, S., 2012. Sanitation for all. Nature 486 (7402), 185.

Y. Feng et al. / Journal of Hydrology 542 (2016) 744755

Nawaz, R., McDonald, A., Postoyko, S., 2015. Hydrological performance of a fullscale extensive green roof located in a temperate climate. Ecol. Eng. 82, 6680.
Nie, W. et al., 2011. Assessing impacts of Landuse and Landcover changes on
hydrology for the upper San Pedro watershed. J. Hydrol. 407 (14), 105114.
NOAA, 2013. Salt Lake City Climate Information.
Oke, T.R., 1979. Advectively-assisted evapotranspiration from irrigated urban
vegetation. Bound.-Layer Meteorol. 17 (2), 167173.
Olszewski, J., Davis, A., 2013. Comparing the hydrologic performance of a
bioretention cell with predevelopment values. J. Irrig. Drainage Eng. 139 (2),
124130.
Orr, A.M., 2013. Transpiration Performance in Bioretention Systems Designed for
Semiarid Climates M.S. Thesis. The University of Utah.
Ouldboukhitine, S.-E., Belarbi, R., Jaffal, I., Trabelsi, A., 2011. Assessment of green
roof thermal behavior: a coupled heat and mass transfer model. Build. Environ.
46 (12), 26242631.
Pauleit, S., Ennos, R., Golding, Y., 2005. Modeling the environmental impacts of
urban land use and land cover changea study in Merseyside, UK. Landscape
Urban Plan. 71 (24), 295310.
Petrucci, G., Rioust, E., Deroubaix, J.-F., Tassin, B., 2013. Do stormwater source
control policies deliver the right hydrologic outcomes? J. Hydrol. 485, 188200.
Powell, S.L., Cohen, W.B., Yang, Z., Pierce, J.D., Alberti, M., 2008. Quantification of
impervious surface in the Snohomish Water Resources Inventory Area of
Western Washington from 19722006. Remote Sens. Environ. 112 (4), 1895
1908.
Ramier, D., Berthier, E., Andrieu, H., 2011. The hydrological behaviour of urban
streets: long-term observations and modelling of runoff losses and rainfall
runoff transformation. Hydrol. Process. 25 (14), 21612178.
Rose, S., Peters, N.E., 2001. Effects of urbanization on streamflow in the Atlanta area
(Georgia, USA): a comparative hydrological approach. Hydrol. Process. 15 (8),
14411457.
Rosenfeld, D., 2000. Suppression of rain and snow by urban and industrial air
pollution. Science 287 (5459), 17931796.
Russell, J., Cohn, R., 2012. Climate of Salt Lake City. LENNEX Corp.
Rutland, D., Dukes, M., 2012. Performance of rain delay features on signal-based
evapotranspiration irrigation controllers. J. Irrig. Drainage Eng. 138 (11), 978
983.
Saadatian, O. et al., 2013. A review of energy aspects of green roofs. Renew. Sustain.
Energy Rev. 23 (0), 155168.
Sailor, D.J., 1995. Simulated urban climate response to modifications in surface
albedo and vegetative cover. J. Appl. Meteorol. 34 (7), 16941704.
Scalenghe, R., Marsan, F.A., 2009. The anthropogenic sealing of soils in urban areas.
Landscape Urban Plan. 90 (12), 110.
Schr, C., Lthi, D., Beyerle, U., Heise, E., 1999. The soil-precipitation feedback: a
process study with a regional climate model. J. Clim. 12 (3), 722741.
Seneviratne, S.I. et al., 2010. Investigating soil moistureclimate interactions in a
changing climate: a review. Earth Sci. Rev. 99 (34), 125161.
Shashua-Bar, L., Pearlmutter, D., Erell, E., 2009. The cooling efficiency of urban
landscape strategies in a hot dry climate. Landscape Urban Plan. 92 (34), 179
186.
Shepherd, J.M., 2006. Evidence of urban-induced precipitation variability in arid
climate regimes. J. Arid Environ. 67 (4), 607628.
Shepherd, J.M., Burian, S.J., 2003. Detection of urban-induced rainfall anomalies in a
major coastal city. Earth Int. 7 (4), 117.
Sherwood, S., Fu, Q., 2014. A drier future? Science 343 (6172), 737739.
Shields, C., Tague, C., 2012. Assessing the role of parameter and input uncertainty in
ecohydrologic modeling: implications for a semi-arid and urbanizing coastal
California catchment. Ecosystems, 117.
Shuster, W., Gehring, R., Gerken, J., 2007. Prospects for enhanced groundwater
recharge via infiltration of urban storm water runoff: a case study. J. Soil Water
Conserv. 62 (3), 129137.
Shuster, W.D., Bonta, J., Thurston, H., Warnemuende, E., Smith, D.R., 2005. Impacts
of impervious surface on watershed hydrology: a review. Urban Water J. 2 (4),
263275.
Simpson, M.G., 2007. Low Impact Development Modeling to Manage Urban Storm
Water Runoff and Restore Predevelopment Site Hydrology. Colorado State
University.
Spracklen, D.V., Arnold, S.R., Taylor, C.M., 2012. Observations of increased tropical
rainfall preceded by air passage over forests. Nature 489 (7415), 282285.
Sullivan, M., Busiek, B., Bourne, H., Bell, S., 2010. Green infrastructure and NPDES
permits: one step at a time. Proc. Water Environ. Federation 2010 (8), 7801
7813.
Sun, G. et al., 2011. Upscaling key ecosystem functions across the conterminous
United States by a water-centric ecosystem model. J. Geophys. Res.: Biogeosci.
116 (G3), 116.

755

Sun, T., Bou-Zeid, E., Wang, Z.-H., Zerba, E., Ni, G.-H., 2013. Hydrometeorological
determinants of green roof performance via a vertically-resolved model for heat
and water transport. Build. Environ. 60, 211224.
Takebayashi, H., Moriyama, M., 2007. Surface heat budget on green roof and high
reflection roof for mitigation of urban heat island. Build. Environ. 42 (8), 2971
2979.
Taylor, C.M., de Jeu, R.A.M., Guichard, F., Harris, P.P., Dorigo, W.A., 2012. Afternoon
rain more likely over drier soils. Nature 489 (7416), 423426.
Thompson, J.R., Green, A.J., Kingston, D.G., 2014. Potential evapotranspirationrelated uncertainty in climate change impacts on river flow: an assessment for
the Mekong River basin. J. Hydrol. 510 (0), 259279.
Trinh, D.H., Chui, T.F.M., 2013. Assessing the hydrologic restoration of an urbanized
area via an integrated distributed hydrological model. Hydrol. Earth Syst. Sci. 17
(12), 47894801.
U.S. Geological Survey, 2015. Groundwater Watch at Site 404531111510101.
USEPA, 2000. Stormwater Phase II Final Rule Fact Sheet.
USEPA, 2008. Reducing Urban Heat Islands: Compendium of Strategies.
Villarreal, E.L., Semadeni-Davies, A., Bengtsson, L., 2004. Inner city stormwater
control using a combination of best management practices. Ecol. Eng. 22 (45),
279298.
Vrsmarty, C.J., Green, P., Salisbury, J., Lammers, R.B., 2000. Global water
resources: vulnerability from climate change and population growth. Science
289 (5477), 284288.
Voyde, E., Fassman, E., Simcock, R., 2010a. Hydrology of an extensive living roof
under sub-tropical climate conditions in Auckland, New Zealand. J. Hydrol. 394
(34), 384395.
Voyde, E., Fassman, E., Simcock, R., Wells, J., 2010b. Quantifying evapotranspiration
rates for New Zealand green roofs. Hydrol. Eng. 15 (6), 395403.
Walsh, C.J. et al., 2016. Principles for urban stormwater management to protect
stream ecosystems. Freshwater Sci. 35 (1), 398411. http://www.journals.
uchicago.edu/doi/abs/10.1086/685284.
Wella-Hewage, C.S., Alankarage Hewa, G., Pezzaniti, D., 2016. Can water sensitive
urban design systems help to preserve natural channel-forming flow regimes in
an urbanised catchment? Water Sci. Technol. 73 (1), 7887.
Weng, Q., 2001. Modeling urban growth effects on surface runoff with the
integration of remote sensing and GIS. Environ. Manage. 28 (6), 737748.
White, M.D., Greer, K.A., 2006. The effects of watershed urbanization on the stream
hydrology and riparian vegetation of Los Peasquitos Creek, California.
Landscape Urban Plan. 74 (2), 125138.
Whitford, V., Ennos, A.R., Handley, J.F., 2001. City form and natural process
indicators for the ecological performance of urban areas and their application to
Merseyside, UK. Landscape Urban Plan. 57 (2), 91103.
Wijesekara, G.N. et al., 2012. Assessing the impact of future land-use changes on
hydrological processes in the Elbow River watershed in southern Alberta,
Canada. J. Hydrol. 412413 (0), 220232.
Wong, G.K.L., Jim, C.Y., 2015. Identifying keystone meteorological factors of greenroof stormwater retention to inform design and planning. Landscape Urban
Plan. 143, 173182.
Wu, H., 2015. Protecting Stream Ecosystem Health in the Face of Rapid Urbanization
and Climate Change Ph.D. Dissertation. The University of Oregon.
Xiao, F., Simcik, M.F., Gulliver, J.S., 2012. Perfluoroalkyl acids in urban stormwater
runoff: influence of land use. Water Res. 46 (20), 66016608.
Yang, J., Wang, Z.-H., Georgescu, M., Chen, F., Tewari, M., 2016. Assessing the impact
of enhanced hydrological processes on urban hydrometeorology with
application to two cities in contrasting climates. J. Hydrometeorol. 17 (4),
10311047.
Yao, L., Wei, W., Chen, L., 2016. How does imperviousness impact the urban rainfallrunoff process under various storm cases? Ecol. Ind. 60, 893905.
Yao, Y. et al., 2014. Estimation of the terrestrial water budget over northern China
by merging multiple datasets. J. Hydrol. 519 (Part A), 5068.
Yeakley, J.A., 2014. Urban Hydrology in the Pacific Northwest, Wild Salmonids in
the Urbanizing Pacific Northwest. Springer, pp. 5974.
Zahmatkesh, Z., Burian, S., Karamouz, M., Tavakol-Davani, H., Goharian, E., 2014.
Low-impact development practices to mitigate climate change effects on urban
stormwater runoff: case study of New York City. J. Irrig. Drainage Eng. 141 (1),
04014043.
Zgheib, S., Moilleron, R., Chebbo, G., 2012. Priority pollutants in urban stormwater:
Part 1 case of separate storm sewers. Water Res. 46 (20), 66836692.
Zhang, L., Kennedy, C., 2006. Determination of sustainable yield in urban
groundwater systems: Beijing, China. J. Hydrol. Eng. 11 (1), 2128.
Zhang, X., Zhang, X., Hu, S., Liu, T., Li, G., 2013. Runoff and sediment modeling in a
peri-urban artificial landscape: case study of Olympic Forest Park in Beijing. J.
Hydrol. 485 (0), 126138.