Published February 20, 2015
Hydropedology Symposium: 10 Years in the Past and 10 Years into the Future–Soil Science Issues
Hydropedological Assessments of Parcel-Level
Infiltration in an Arid Urban Ecosystem
William D. Shuster*
USEPA
Office of Research and Development
68 W. Martin Luther King Dr.
Cincinnati, OH 45268
Stephen D. Dadio
Cedarville Engineering Group, LLC
1033 S Hanover St., Suite 300
N. Coventry, PA 19465
Caitlin E. Burkman
USEPA
Office of Research and Development
68 W. Martin Luther King Dr.
Cincinnati, OH 45268
Stevan R. Earl
Julie Ann Wrigley Global Institute of
Sustainability
Arizona State Univ.
PO Box 875402
Tempe, AZ 85287
Sharon J. Hall
School of Life Sciences
Arizona State Univ.
PO Box 874501
Tempe, AZ 85287
Soil morphology and correspondent hydrologic data can contribute to qualifying and quantifying urban soil suitability and capacity to cycle stormwater
runoff. We put particular emphasis on the possibility that residential parcels
may manage their own stormwater on pervious yard areas. We assessed the
morphology of Aridisol pedons (as deep cores to approximately the 3.6m depth) via soil taxonomy, performed in situ measurements of infiltration,
and measured subsoil hydraulic conductivity in two desert parks, four residential parcels, and three dual-purpose park–stormwater retention basins in
the Phoenix, AZ, metropolitan area. Infiltration rates overall ranged between
0.4 and 1.7 cm h−1. We used borehole hydraulic conductivity as a proxy for
drainage, which ranged from a low of 0.0 to 13 cm h−1. Representing a baseline, or set of “natural” hydrologic processes, native desert sites exhibited a
relatively high capacity to infiltrate and redistribute rainfall, which, depending on storm intensity, may drain to washes or ephemeral streambeds via
subsurface runoff. Simulations of stormwater runoff on residential parcels
showed that when rooftop runoff is directed to pervious patches, rainfall is
entirely infiltrated and redistributed, which predicts good potential for parcel-level stormwater management. Our observation of an accumulation of
fine sediments in retention basin surface soils was corroborated with local
observations that drawdown times in the retention basins had increased
with time. Although further assessment is needed on a wider variety of semiarid landscapes, our results suggest a positive role for residential parcels in
detaining stormwater on site and perhaps easing the wet-weather burden on
centralized stormwater infrastructure in semiarid urban ecosystems.
Abbreviations: PR, Phoenix Residential; SR1, Scottsdale Residential 1; SR2, Scottsdale
Residential 2; TR: Tempe Residential.
T
hrough the laying down of impervious surfaces, urban development at
once creates excess stormwater runoff and limits the proportion of landscapes that offer infiltration opportunities. These conditions lead to large
volumes of unmanaged runoff that may contribute to the malfunction of wastewater infrastructure and flooding of landscapes where runoff is most concentrated
(Arnold and Gibbons, 1996; Walsh et al., 2012). Alternatively, there are potential
benefits to decentralizing stormwater management by capitalizing on and enhancing infiltration throughout the watershed (e.g., in residential parcels [Mayer et
al., 2012] and retention basins [Carlson et al., 2011; Larson and Grimm, 2012]),
thereby reducing reliance on centralized stormwater infrastructure and, potentially, increasing the use of stormwater as a renewable resource by increasing groundwater recharge (Gallo et al., 2013). The concept of using stormwater as an input to
water-stressed urban centers, particularly those in arid or semiarid climates, is par-
Soil Sci. Soc. Am. J.
doi:10.2136/sssaj2014.05.0200
Open Access.
Received 16 May 2014.
*Corresponding author ([email protected]).
© Soil Science Society of America, 5585 Guilford Rd., Madison WI 53711 USA
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by
any means, electronic or mechanical, including photocopying, recording, or any information storage
and retrieval system, without permission in writing from the publisher. Permission for printing and for
reprinting the material contained herein has been obtained by the publisher.
Soil Science Society of America Journal
ticularly relevant. For example, cities in the southwestern United
States are vulnerable to shortfalls in water availability, which is
largely dependent on the dynamic balance among extracted, imported, and stored water resources (Padowski and Jawitz, 2012).
Anthropogenic imprints on the landscape alter soil morphological characteristics (e.g., horizonation, texture, structure, and
mineralogy; Capra et al., 2013) and, in turn, interact to mediate
soil hydrology. A better understanding of remnant urban permeable surfaces and the coupled hydrology–pedology of underlying
soils can contribute to defining the practical capacity for soils to
cycle runoff (Schoeneberger and Wysocki, 2005; Dominati et al.,
2010). This knowledge can then be used to identify and leverage
soils that have the capacity to generate urban ecosystem services,
such as the management of stormwater runoff at the parcel or
municipal level (Robinson et al., 2014).
Land use profoundly alters landforms and disturbs soils, thus
regulating the capacity of soils for infiltration and redistribution
of stormwater runoff volume. The expansive urban–suburban matrix of metropolitan Phoenix, AZ, reflects a diverse mix of land use
histories. Phoenix has a pronounced agricultural lineage, beginning with the early Hohokam people, who constructed the area’s
first canal system, to modern agricultural systems that use an extensive water-distribution network focused on cotton (Gossypium
barbadense L.) and alfalfa (Medicago sativa L.), among other crops
(Redman and Kinzig, 2008). Phoenix remained relatively sparsely populated until the latter half of the 20th and early 21st centuries, when the region experienced exponential growth, which
consumed expansive areas of both agricultural and native desert
lands (Keys et al., 2007; Redman and Kinzig, 2008). Residential
areas are the most pronounced feature on the landscape, comprising >60% of the developed area in Maricopa County in central
Arizona (Maricopa Association of Governments, unpublished
data, 2012; http://www.azmag.gov/). Residential landscapes
span many forms, ranging from mesic (high water use, typically
with turfgrass and exotic or introduced large trees) to xeric (predominantly gravel substrate interspersed with native, exotic, or
introduced drought-tolerant plants), with the predominant form
varying across time and space within the developed area (Martin
et al., 2003; Grimm et al., 2013).
The choice of landscape type has ecological implications, as
yard type influences soil properties (Lewis et al., 2006) and biogeochemical cycling (Hall et al., 2009). In addition, research has
documented a legacy of prior land use history, with residential
landscapes that were formerly desert or agriculture exhibiting distinct soil properties (Lewis et al., 2006; Hall et al., 2009). While
mesic landscapes are and have been a prominent landscape type
in Phoenix, there is a trend toward xeric landscapes to promote
water conservation, with many municipalities providing rebates
for turfgrass removal (Kelly, 2005). As a result, many Phoenixarea residential landscapes have undergone one or more transitions from desert or agriculture to mesic or xeric landscaping and,
most recently, from mesic directly to xeric, with implications for
soil hydrology. Native soils in undisturbed landscapes may be less
compacted than residential parcels (depending on the depth to
∆
less permeable calcic and petrocalcic horizons, which are natural
features in this landscape), and they contain less imported soil
material than soils in residential parcels. Retention basins are a
common stormwater management feature on the Phoenix landscape (e.g., Larson and Grimm, 2012), and these basins often
serve a dual purpose as recreational park areas. The infiltration
capacity of retention basins may degrade due to inputs of sediments that accumulate and form more finely textured soils from
the native turf soils (which have some proportion of fines to start
with), in addition to compaction by trampling that comes with
land use as a playing field (Nascimento et al., 1999). These circumstances may lead to decreased infiltration. In this study, we
compared soil hydrologic characteristics across arid soils set in
both native desert and residential areas as a potential regulator
of infiltration and redistribution of precipitation and stormwater runoff. We contrasted the hydropedological features of native
and residential landscapes with characteristics of existing stormwater retention basins. The capacity of residential lots as passive
infrastructure for stormwater management was investigated via
modeling based on field measurements.
MATERIALS AND METHODS
Site Descriptions
This study was conducted in the Greater Phoenix metropolitan area of central Arizona (Fig. 1). Encompassing numerous
municipal centers, Phoenix is a rapidly growing urban area that
was home to more than 4.3 million residents in 2012. Situated at
the northern extent of the Sonoran Desert, the region is hot and
dry, with mean daily (1933–2013 [Western Regional Climate
Center, 2013a,b]) maximum and minimum temperatures of
30.0 and 15.2°C, respectively, and 189 mm of average rainfall
distributed bimodally as summer monsoons and winter Pacific
frontal storms. All sites lie within Major Land Resource Area 40,
the Sonoran Basin and Range (NRCS, 2006). In this study, we
assessed the soil properties of four residential parcels, two native
Sonoran desert sites, and three retention basins. The two desert
sites are located on lands owned and maintained by the Maricopa
County Regional Park System (Usery and White Tanks). These
sites are located in outlying native Sonoran Desert and have not
been subject to direct anthropogenic disturbance. The four residential lots (Phoenix Residential [PR], Scottsdale Residential 1
[SR1], Scottsdale Residential 2 [SR2], and Tempe Residential
[TR]) are located throughout Maricopa County and represent a
range of neighborhoods that were constructed within the past 30
to 40 yr. Three of the lots were formerly desert, two of which have
xeric landscapes, and the third was formerly mesic (i.e., turfgrass
lawn) but has been converted to a xeric landscape with droughttolerant shrubs. The fourth residential lot is located in a former
agricultural area (cotton) and currently has a xeric landscape with
native trees and shrubs. Human disturbance is present in the
soil surface (upper 30 cm) at all sites due to residential outdoor
landscaping (e.g., establishment of lawns and other vegetation,
the addition of surface gravel, terracing). Since construction of
the homes, these sites may have received anthropogenic manageSoil Science Society of America Journal
Fig. 1. The geography of the Phoenix, AZ, metropolitan area with sampling locations.
ment inputs, such as irrigation water, fertilizers, and pesticides, in
addition to aeolian topdressing through haboobs or dust storms
(anthropogenic and eroded dust particles). The stormwater retention basins (Mescal, Northsight, and Cactus) are located in
the city of Scottsdale, AZ. In addition to retaining stormwater,
these multiuse facilities serve as recreational fields and feature
turfgrass that is maintained and utilized throughout the year.
Core Sampling and Measurements
Soil coring and hydrologic assessment procedures were
outlined by Shuster et al. (2014). Soil cores (6-cm diameter, 1.3
m long) were taken to refusal with a truck-mounted Geoprobe
(Geoprobe Systems) in the front yard of each residential parcel
and within a representative location (a space between shrubs)
of the desert sites that was accessible to the truck. Core samples
were classified in the field according to soil survey standards from
NRCS (2013) and Soil Survey Staff (1993, 2010). Core samples
were described for color, texture, and transitions between layers,
and the rock or debris fragments percentage was estimated. A
second borehole was developed to measure the saturated hydraulic conductivity of shallow B horizon material as a proxy for the
shallow subsoil drainage rate. This measurement was made with a
compact constant-head permeameter (CCHP or Amoozemeter,
www.soils.org/publications/sssaj
Ksat, Inc.). Water flux data collected from the CCHP were used
to calculate Ksat:
K sat = AQ
[1]
where Ksat is the subsurface hydraulic conductivity, A is a constant based on the radius and head of water in the borehole,
and Q is the steady-state rate of water flow into the borehole
(Amoozegar, 1989). The method addresses the equilibrium outflow of water from the borehole in a quasi-spheroidal geometry
and is therefore an approximate measure of subsurface redistribution of soil moisture and drainage. We made a total of eight
measurements of unsaturated surface hydraulic conductivity,
Kunsat, at each site with tension infiltrometers run at a suction
head of 2 cm (Mini Disk infiltrometer, Decagon Devices). The
estimated Kunsat was calculated according to Zhang (1997). This
near-saturation estimate of hydraulic conductivity was made according to our assessment protocol to eliminate the flow contribution from surface cracks and larger macropores.
Data Analysis and Modeling
We took a comparative approach to interpreting data in a
hydropedological context, with an emphasis on the mean and
standard error of the hydrologic data. Soil profile graphics were
∆
developed using the R package aqp (Beaudette et al., 2013; R
Version 3.0.1). We used the USEPA Storm Water Management
Model (SWMM Version 5.0.022) to simulate losses in the pervious subareas of residential sites. Models were run with a particularly intense sequence of storms in the period 18 to 24 Jan. 2010,
which totaled approximately a 10-cm depth across five pulses.
The hydrology of each residential site was modeled as rooftop
runoff volume routed to an equivalently sized pervious yard area,
with the flow width set to that of the roof width dimension.
Therefore, parcel area, overland flow width, and landscape slope
were parameterized according to the specific dimensions of each
residential site. All residential sites had Manning’s n set to 0.01 as
an approximation for the roughness of impervious surfaces such
as rooftops and 0.032 for pervious surfaces to approximate the
resistance to flow presented by a gravelly soil surface. Each yard
area had different types and depths of gravel, the depths of which
were measured in the field and reflected in the model as a proxy
for the depth of depression storage. Infiltration losses were generated in each instance via an implementation of the Green–Ampt
method, with measured infiltration rate (as measured Kunsat,
cm h−1), a suction head to reflect the dry soil conditions that
we encountered (i.e., 20-cm tension), and a corresponding value
for soil moisture deficit (?0.4). Infiltrated water volume was
drained via an aquifer submodel based on the measured borehole Ksat. The aquifer submodel was further parameterized with
the measured soil depth, estimates of evaporative depth based on
observation, and total porosity, soil moisture, and field capacity
and wilting point, which were based on the soil texture. Excess
runoff and drainage volume were routed as inflows to a single,
freely drained outfall.
RESULTS
The desert, residential, and retention basin soils observed
in this study were all derived from fan alluvium deposits in the
Phoenix metropolitan area and ranged in classification from
coarse-loamy to loamy-skeletal, mixed, superactive, hyperthermic Typic Haplocambids. This area consists of deep sediment
that was deposited as alluvial fans from the adjacent, surrounding
mountains. These deposits are characterized primarily by their
Table 1. Unsaturated hydraulic conductivity (Kunsat, tension
infiltrometer) values taken at the landscape surface along
with soil texture and Manning’s n for different land uses in the
Phoenix, AZ, metropolitan area.
Site
Usery, desert
White Tanks, desert
Tempe Residential
Scottsdale Residential 1
Scottsdale Residential 2
Phoenix Residential
Cactus retention basin
Mescal retention basin
Northsight retention basin
† Mean ± standard error.
∆
Kunsat
Soil texture
cm h−1
0.6 ± 0.2†
sandy loam
0.6 ± 0.1
fine sandy loam
0.9 ± 0.2
clay loam
0.6 ± 0.1
fine sandy loam
1.7 ± 0.3
coarse sandy loam
1.5 ± 0.5
coarse sandy loam
0.4 ± 0.1
silt loam
0.5 ± 0.1
loam
1.1 ± 0.2 very fine sandy loam
n
8
8
8
8
14
8
8
8
8
distance from their alluvial fan source. Soils closer to the mountains are generally of a coarser texture with a higher rock content,
while soils that are located farther from the mountains tend to
have a finer texture with a lower rock content. As either Aridisols
or Entisols, these soils exhibit minimal soil development due, in
part, to the arid climate, which limits the soil moisture available
for leaching and subsequent horizon development. The Aridisols
are characterized primarily by the presence of calcareous horizons within the subsoil, whereas the Entisols are located in less
stable landscapes where calcareous horizons have not formed.
There was no difference in infiltration rate (Table 1) between the native Usery and White Tanks desert sites despite
different soil textures, which were sandy loam and fine sandy
loam, respectively, yet the drainage rate for Usery was more than
twice that of White Tanks (Table 2). The gravel content in Usery
subsoils was less (25–35%) than that for White Tanks (Table
3; 60%). Unconsolidated parent material of different types was
found at Usery as two distinct layers of gravelly loamy sands (as
50% gravel-sized particles) in sequence (Table 3). White Tanks
is at a steeper, higher elevation, and consequently less stable location within the alluvial fan, and we were unable to identify an A
horizon at this site. The depth of drilling was limited by the presence of both cobble-sized rocks throughout the profile (Table 3)
and a petrocalcic horizon at a depth of 1 to 2 m below the ground
surface.
We assessed four residential parcels that reflect several contrasting but common Phoenix-area landscaping styles. For the
Tempe Residential site, surface cover was a particularly thick
(20–30 cm) layer of gravel underlain by a distinct clay loam A
horizon with an infiltration rate of 0.9 cm h−1 (Tables 1 and 4).
The drainage rate was measured as 0.5 cm h−1 in subsoils that
tended toward a finer loam (fine sandy loam; Table 2). The variegated coloration of the deeper extent of the pedon may indicate
some legacy wetness (Table 4). The Scottsdale Residential 1 site
had a thin gravel layer that was intermixed with a layer of loam
soil material, which sat on a plastic mulch layer. For the parts of
the site where we excavated the gravel, the sheeting appeared to
be in good condition without tears or holes. Infiltration into the
Table 2. Subsurface saturated hydraulic conductivity (Ksat cm
h−1, constant-head permeameter) values taken in boreholes
along with soil texture and Manning’s n for different land uses
in the Phoenix, AZ, metropolitan area.
Site
Usery, desert
White Tanks, desert
Tempe Residential
Scottsdale Residential 1
Scottsdale Residential 2
Phoenix Residential
Ksat
Soil texture
7.2 ± 0.2†
sandy loam
3.0 ± 0.5
fine sandy loam
0.5
loam–fine sandy loam
1.2 ± 0.3
fine sandy loam
13.1 ± 10.1
coarse sandy loam
7.4 ± 0.4
coarse sandy loam
n
5
2
1
5
2
2
Cactus retention basin
0.0 ± 0.0
sandy loam–
fine sandy loam
2
Mescal retention basin
0.2 ± 0.2
fine sandy loam
4
0.7 ± 0.4
sandy loam–
fine sandy loam
4
Northsight retention basin
† Mean ± standard error.
Soil Science Society of America Journal
Table 3. The morphological features of pedons at the desert sites.
Site
Depth
Horizon
Munsell color
Texture†
cm
0–56
A
10YR 4/3
sl
56–76
Bk1
7.5YR 5/4
sl
76–99
Bk2
7.5YR 6/4
sl
99–224
Bk3
7.5YR 7/3
sl
224–305
C1
10YR 6/4
ls
305–366
C2
7.5YR 7/3
ls
White Tanks
0–61
Bk
7.5YR 5/6
fsl
61–91
C
7.5YR 5/4
sl
† sl, sandy loam; ls, loamy sand; fsl, fine sandy loam.
‡ gr, gravelly; vgr, very gravelly; cb, cobbly; vcb, very cobbly.
Usery
gravelly loam just above the plastic sheet measured 0.6 ± 0.1 cm
h−1; once the wetting front reached the plastic mulch, lateral
movement of water was assumed (Tables 1 and 4). The plastic
sheet overlies a 2A, 2Bk, and 2C sequence (Fig. 2), and the 2B
horizon drained at a rate of 1.2 ± 0.3 cm h−1 (Table 2). This set of
surface horizons was then underlain by an older buried Aridisol
indicated as 3A, 3Bk, and 3C, which was probably formed during the Pleistocene (Fig. 2). This buried Aridisol has an older,
deeper B horizon, which differentiates it from the more recent
surficial Aridisol. We observed a relatively high infiltration rate
at Scottsdale Residential 2 (coarse sandy loam, 1.7 cm h−1) and
Parent material
fan alluvium
fan alluvium
Rock fragments
Rock fragment
modifier‡
Rock fragment
type
%
25
35
25
25
50
50
20
60
gr
vgr
gr
gr
vgr
vgr
cb
vcb
gravels
cobbles
a comparatively very high drainage rate (13.1 cm h−1) through
the rocky sandy loam (Tables 1, 2, and 4). Soil development was
more advanced than at the other sites, with a thin B horizon (Fig.
2) and darker sandy loam C horizon. The proportion of gravel
in the C1 and C2 horizons was also the overall highest that was
observed in the study (ranging between 70 and 90%; Table 4).
Following the removal of turf from the Phoenix Residential
landscape, an 8-cm layer of gravel sat over a coarse sandy loam A
horizon with an infiltration rate of 1.5 cm h−1 (Tables 1 and 2).
Our measurement of drainage integrated across the sandy loam
2Bk and 2C horizons, and the drainage rate was 7.4 cm h−1,
Table 4. The morphological features of pedons at residential land use sites.
Site
Tempe
Residential
Depth
cm
0–38
38–53
53–71
71–163
163–269
Horizon†
Munsell color
Texture‡
Parent material
^C/A1
2A2
2Bk1
2Bk2
2C1
–¶
7.5YR 4/4
7.5YR 6/4
7.5YR 7/4
7.5YR 8/3
variegated
cl
l
l
l
fsl
anthropogenic
fan alluvium
269–366
2C2
7.5YR 6/3
variegated
sl
Scottsdale
Residential 1
0–20
^C/A1
–
l
anthropogenic
20–51
2A2
7.5YR 5/6
sl
fan alluvium
51–122
2Bk
7.5YR 7/3
l
122–254
2C
7.5YR 7/4
l
254–279
3A
7.5YR 5/4
fsl
279–335
3Bk
7.5YR 8/2
fsl
335–366
3C
7.5YR 7/4
fsl
Scottsdale
0–15
Bw
5YR 4/6
csl
fan alluvium
Residential 2
15–122
C1
7.5YR 5/6
csl
122–152
C2
–
csl
Phoenix
0–10
^C/A1
–
csl
anthropogenic
Residential
10–20
2A2
10YR 5/4
csl
fan alluvium
20–53
2Bk1
10YR 6/4
csl
53–71
2Bk2
10YR 7/4
csl
71–366
2C
10YR 7/3
csl
† ^, anthropogenic activity contributed to development of the horizon.
‡ cl, clay loam; l, loam; fsl, fine sandy loam; sl, sandy loam; csl, coarse sandy loam.
§ gr, gravelly; xgr, extra gravelly.
¶ Not applicable.
www.soils.org/publications/sssaj
Rock
fragments
%
90
0
10
10
30
Rock fragment
modifier§
30
gr
0
0
10
10
0
0
0
20
70
90
100
15
20
15
15
–
–
–
–
–
–
–
gr
xgr
xgr
–
gr
gr
gr
gr
–
–
–
–
gr
Rock fragment type
landscape gravels
gravels
–
–
gravels
–
–
–
gravels
weathered rock
landscape gravels
gravels
∆
Fig. 2. Soil profiles for each reference and suburban site graphed according to depth and Munsell color for each diagnostic horizon. The thick, black
line for Scottsdale Residential 1 indicates the depth of plastic sheeting; ^ indicates that anthropogenic activity contributed to horizon development.
with refusal at 3.7 m (Fig. 2; Tables 1, 2, and 4). Modeling results
run for each of the four residential parcels indicated that runoff
generated from rooftop impervious areas was categorically fully
detained (infiltrated and redistributed) in pervious yard areas.
Each of the stormwater retention basins doubled as recreational facilities and had complete turf cover [Bermuda grass,
Cynodon dactylon (L.) Pers.] across a level field. The Cactus site
had a silt loam surface soil (ranging in thickness from 8 to 20 cm;
Table 5) with an infiltration rate of 0.4 ± 0.1 cm h−1 (Table 1),
and a drainage rate of zero was measured in the lower extent of
the fine sandy loam B horizon (Table 2). Boreholes at the Cactus
site were prone to collapse and yielded only a single usable measurement of drainage rate. The Mescal site had a loam surface soil
that was 10 to 15 cm thick with an infiltration rate of 0.5 ± 0.1
cm h−1 (Tables 1 and 5). The drainage rate was measured across
fine sandy loam B horizons overlying a sandy loam C horizon,
with 15 to 20% gravel content (Tables 2 and 5). The 15-cmthick, very fine sandy loam surface soil at the Northsight location
infiltrated water at a rate of 1.1 ± 0.1 cm h−1 (Tables 1 and 5).
We averaged drainage measurements across fine sandy loam to
sandy loam B (?30-cm depth, n = 2) and C horizons (?90-cm
depth, n = 2), which yielded an overall site drainage rate of 1.1 ±
0.2 cm h−1 (Fig. 2; Tables 2 and 5).
DISCUSSION
Soil development in each location was driven by some degree of anthropogenic influence. The desert reserve areas and
SR2 had surface horizons that developed from fan alluvium, but
the balance of the residential sites had landscapes with different
thicknesses and types of gravel materials, and surface soils in the
∆
retention parks apparently formed from an accumulation of fine
sediments over native fan alluvium. Soils that formed higher in
the landscape (White Tanks, PR, and SR2) were more coarsely
textured with a higher rock content. Soils that formed lower in
the landscape (Usery, SR1, and TR) had a higher proportion
of fines, and there was a clear anthropogenic–aeolian influence
on the development of fine surface soils in the retention basins.
We found evidence of disturbance in the native desert soils. The
50-cm thickness of the darker (10YR 4/3) mineral surface layer
of the Usery profile was unexpected because surface horizons
in Aridisols are usually thin (5–10 cm) due to evapotranspiration and a consequent decrease in profile development. This
evaporative demand largely drives near-surface soil development
processes in aridic (torric) moisture regimes (Liu et al., 1995).
The thicker surface layer was possibly an artifact of grading activities involved in developing an adjacent roadbed, such that
additional surface material was layered over the native soils (see
Lyford and Qashu, 1969). The percolation of water into Usery
was more thorough than that at White Tanks, which was limited
by a relatively shallow impermeable horizon. The layering of soils
at White Tanks led us to speculate that rainfall would infiltrate
and move through the B horizon, and at approximately the 1-m
depth, water would encounter an impermeable petrocalcic layer,
move laterally, and then emerge as seeps or base flow in ephemeral streambeds and washes.
The residential parcels that we assessed had landscapes that
largely reflected the objectives of land developers, which were
modified by subsequent homeowners, and generally had mixtures of xeriscaping techniques with interspersed vegetation.
The plastic sheeting at SR1 was installed before construction
Soil Science Society of America Journal
Table 5. The morphological features of pedons at dual-purpose park–stormwater retention basins.
Site
Cactus
Depth
cm
0–20
20–33
33–122
122–295
295–345
Horizon†
Munsell color
Texture‡
Parent material
Ap
Bw
C1
2C2
3C3
10YR 4/3
7.5YR 6/6
7.5YR 7/4
7.5YR 7/3
7.5YR 7/3
variegated
sil
sl
sl
fsl
ls
anthropogenic/aeolian
fan alluvium
345–366
2C4
7.5YR 4/4
l
0–15
^Ap
10YR 4/3
l
anthropogenic/aeolian
15–53
Bw1
7.5YR 7/3
fsl
fan alluvium
53–97
Bw2
7.5YR 6/6
sl
97–122
Bw3
7.5YR 5/6
fsl
122–218
2C1
7.5YR 5/6
sl
218–244
2C2
7.5YR 6/6
fsl
Northsight
0–15
Ap
10YR 4/4
l
anthropogenic/aeolian
15–152
Bw
10YR 5/6
sl
fan alluvium
152–224
2C1
7.5YR 6/6
sl
224–310
2C2
7.5YR 5/6
fsl
310–345
3C3
7.5YR 7/4
sl
345–366
3C4
7.5YR 6/4
fsl
† ^, anthropogenic activity contributed to development of the horizon.
‡ sil, silt loam; sl, sandy loam; fsl, fine sandy loam; ls, loamy sand; l, loam.
§ gr, gravelly; vgr, very gravelly.
¶ Not applicable.
Mescal
of the residence (?1978) with the intent of providing weed
control and, possibly, moisture management (e.g., Kelly, 2005)
and is contiguous except for holes where shrubs, trees, or utility
standpipes are present. Overall, the impermeable plastic sheeting limits the infiltration depth and arrests the role of drainage,
which was measured in a borehole at >1 cm h−1. We would also
expect little or no soil development at this location. Any change
in the gravelly topsoil composition would probably come from
fine-textured aeolian deposits that are washed into the gravel
layer and continue to collect on the surface of the plastic mulch.
While SR1 may have a reduced overall detention capacity due
to the impermeable plastic layer, the backyard area offered additional capacity to infiltrate and redistribute stormwater runoff
volume. The SR2 site was near the edge of development, and the
landscape at this location appeared to be more similar to the surrounding desert. The gravel and cobble that underlies the thin A
horizon at SR2 is unconsolidated but tightly packed, contributing to high variability in the drainage rate (Table 2). A gravel
surface layer was used as part of the landscaping at TR, SR1, and
PR and provided considerable detention capacity. These gravels
exhibited free lateral drainage (as measured by double-ring infiltration techniques; data not shown). Infiltration rates for soils
below the gravel are much less than that of the gravel (or impermeable for SR1), and water movement may be modulated by different processes. The inflow may move downward through the
gravel layer, at which point the water reaching the soil surface
would either infiltrate immediately or produce runoff as infiltration excess. In the latter case, runoff flow may then follow the
landscape contour. Residential landscapes were sloped toward
www.soils.org/publications/sssaj
Rock fragments
%
0
20
30
10
50
0
0
20
30
15
25
10
0
25
50
0
25
10
Rock fragment
modifier§
–¶
gr
gr
–
vgr
–
–
gr
gr
gr
gr
–
–
gr
vgr
–
gr
–
Rock fragment
type
gravels
–
gravels
–
gravels
–
gravels
sidewalks and streets, so any lateral flow that might seep out (i.e.,
daylights) would probably be routed to the street-level stormwater collection system. Based on our simulations, however, pervious patches can infiltrate and redistribute a sequence of closely
spaced wet-weather flows.
An unresolved question is whether these permeable patches
in the aggregate can provide an adequate detention capacity to
infiltrate and redistribute typical wet-weather stormwater volumes at the scale of the local municipal sewer service area. In
this conception of a decentralized stormwater management approach, each parcel would attempt to route all wet-weather flows
into pervious patches. These soils appear to be able to absorb an
intentional routing of roof downspout flow and swimming pool
backwash volume. Our modeling approach, however, involved
routing flow from rooftops to pervious areas along a runoff flow
width equal to the width of the roof. This scenario assumes that
roof runoff is delivered to the pervious area via a flow spreader,
which prevents runoff flows from concentrating and keeps the
drainage area as large as possible.
Stormwater management in the southwestern United States
is problematic to the extent that a large proportion of the total
annual rainfall record arrives in monsoons. These storms can be
brief, yet hyetographs often indicate intense precipitation rates
that can create flooding conditions in urban drainage areas.
Although evaporation is a predominant and considerable source
of loss in the local hydrologic cycle, it acts far too slowly to manage the wet-weather stormwater volume. This is one reason that
retention basins and other centralized stormwater infrastructure features are used. Soils under various land use management
∆
can act in different capacities as passive or otherwise infiltrative
stormwater control measures. However, the retention basins
studied exhibited unfavorable infiltration and drainage conditions compared with the same assessments made in native desert
and residential land uses. Our taxonomic data indicates mixed
anthropogenic–aeolian surface soils in the retention basins. The
finer surface soils in the parks are probably the product of what
may have originally been sandy loam turf soils. The sandy loam
surface soils probably evolved along the textural continuum to
a fine sandy loam due to ongoing enrichment with fines and
grit sediments settling out from periodic stormwater inundation and, potentially, also the accumulation of wind-deposited
silt-sized particles (Reheis, 2006; Reynolds et al., 2006; Neff et
al., 2008). The finer topsoils in each of the basins serve to decrease the infiltration rate and thereby increase the amount of
time for post-storm drawdown of the runoff volume. A mixture
of fine sandy loam and sandy loam subsoils apparently do not
have high permeability, as reflected in the low measured drainage
rates. Although we did not measure the bulk density or conduct
penetrometry in these basins, it stands to reason that year-round
trampling of the playing fields would contribute to compaction
and a subsequent decrease in infiltration. The shallow turf root
mass, which is most dense in the first few centimeters of the soil
surface, would do little to mitigate against the effects of compaction, which can persist deeper into the soil profile.
CONCLUSIONS
This field study incorporated taxonomic and hydrologic
data to illustrate the potential for distributed, parcel-level infiltration as a tool for municipal stormwater management.
Hydropedological data indicated that the infiltration performance of retention basins is low, probably due to the development of finer surface soils and compaction. Although this pilot
study assessed relatively few sites in each of native desert, residential, and retention basins, there are some clear delineations in
the capacity of the soils associated with these land use types as it
relates to infiltration capacity. Native desert soils were subject to
some level of disturbance (enhancement of the A horizon, possibly by road grading) and had a relatively high overall capacity to
infiltrate and redistribute precipitation. One of the native desert
sites (White Tanks) had a shallower soil overall and would be
expected to shunt any infiltrated water to washes or ephemeral
streambeds via subsurface runoff. Because these desert park areas
are not developed, wet-weather flooding in the low areas and activation of ephemeral streams could be part of the natural hydrologic cycle in these reserve areas.
Residential parcels exhibited hydrologic characteristics
and soil morphology that provide the capacity for infiltration of
stormwater volume and redistribution of soil moisture throughout the pedon. Taken as an aggregate, our findings suggest that
residential yards may have sufficient infiltration capacity to detain the runoff volume that they produce. Public campaigns to
promote on-site detention (e.g., by minimizing impervious areas, planting trees, enriching soils with organic matter, etc.) may
∆
contribute to increasing the capacity of these areas to store precipitation and, possibly, enhance groundwater recharge. While
complete detention is not feasible, the contribution of residential
yards and other areas to passive infiltration may reduce the burden on the centralized stormwater infrastructure. For example,
our assessment highlighted the decreased efficiency of retention
basins for drawdown. City officials have remarked that it has taken longer each year for equivalent volumes of water to infiltrate,
and it is clear that an increasingly fine-textured topsoil can drive
this trend, which would be further aggravated by compaction.
These field-based diagnoses may help focus operation and maintenance priorities for the retention basins. Our limited number
of hydropedological assessments indicate a potential for stormwater management via infiltration into urban Aridisols, although
further study is needed to develop finer scale mapping of soil hydrology in this arid conurbation.
ACKNOWLEDGMENTS
We thank Maricopa County Parks and Recreation for access to desert
locations. This contribution and work of authors S.R. Earl and S.J.
Hall for this manuscript is supported, in part, by the National Science
Foundation under Grant no. BCS-1026865 (CAP LTER). Hannah
Heavenrich, Russ Losco, and Donnie Vineyard provided valuable
field assistance. This project was supported in part by an appointment
(C.E. Burkman) to the Research Participation Program at the USEPA
National Risk Management Research Laboratory administered by the
Oak Ridge Institute for Science and Education through an interagency
agreement between the USDOE and USEPA.
REFERENCES
Amoozegar, A. 1989. Comparison of the Glover solution with the simultaneous
equations: Approach for measuring hydraulic conductivity. Soil Sci. Soc.
Am. J. 53:1362–1367. doi:10.2136/sssaj1989.03615995005300050010x
Arnold, C.L., Jr., and C.J. Gibbons. 1996. Impervious surface coverage: The
emergence of a key environmental indicator. J. Am. Plann. Assoc. 62:243–
258. doi:10.1080/01944369608975688
Beaudette, D.E., P. Roudier, and A.T. O’Geen. 2013. Algorithms for quantitative
pedology: A toolkit for soil scientists. Comput. Geosci. 52:258–268.
doi:10.1016/j.cageo.2012.10.020
Capra, G.F., S. Vacca, E. Cabula, E. Grilli, and A. Buondonno. 2013. Through the
decades: Taxonomic proposals for human-altered and human-transported
soil classification. Soil Horiz. 54(2). doi:10.2136/sh12-12-0033
Carlson, M.A., K.A. Lohse, J.S. McIntosh, and J.E.T. McLain. 2011. Impacts of
urbanization on groundwater quality and recharge in a semi-arid alluvial
basin. J. Hydrol. 409:196–211. doi:10.1016/j.jhydrol.2011.08.020
Dominati, E., M. Patterson, and A. Mackay. 2010. A framework for classifying
and quantifying the natural capital and ecosystem services of soils. Ecol.
Econ. 69:1858–1868. doi:10.1016/j.ecolecon.2010.05.002
Gallo, E.L., P.D. Brooks, K.A. Lohse, and J.E.T. McLain. 2013. Land cover controls
on summer discharge and runoff solution chemistry of semi-arid urban
catchments. J. Hydrol. 485:37–53. doi:10.1016/j.jhydrol.2012.11.054
Grimm, N.B., C.L. Redman, C.G. Boone, D.L. Childers, S.L. Harlan, and
B.L. Turner II. 2013. Viewing the urban socioecological system through
a sustainability lens: Lessons and prospects from the Central Arizona–
Phoenix LTER Program. In: S.J. Singh et al., editors, Long term socioecological research. Human–Environ. Interact. Ser. 2. Springer, Dordrecht,
the Netherlands. p. 217–246.
Hall, S.J., B. Ahmed, P. Ortiz, R.C. Davies, R.A. Sponseller, and N.B. Grimm.
2009. Urbanization alters soil microbial functioning in the Sonoran
Desert. Ecosystems 12:654–671. doi:10.1007/s10021-009-9249-1
Kelly, J. 2005. Converting turf to a xeriscape landscape: How to eliminate a
bermudagrass lawn using glyphosate. Publ. AZ1371. Univ. of Arizona
Coop. Ext., Tucson.
Keys, E., E.A. Wentz, and C.L. Redman. 2007. The spatial structure of land use
Soil Science Society of America Journal
from 1970–2000 in the Phoenix, Arizona, metropolitan area. Prof. Geogr.
59:131–147. doi:10.1111/j.1467-9272.2007.00596.x
Larson, E.K., and N.B. Grimm. 2012. Small-scale and extensive
hydrogeomorphic modification and water redistribution in a desert city
and implications for regional nitrogen removal. Urban Ecosyst. 15:71–85.
doi:10.1007/s11252-011-0208-1
Lewis, D.B., J.P. Kaye, C. Gries, A.P. Kinzig, and C.L. Redman. 2006. Agrarian
legacy in soil nutrient pools of urbanizing arid lands. Global Change Biol.
12:703–709. doi:10.1111/j.1365-2486.2006.01126.x
Liu, B., F. Phillips, S. Hoines, A.R. Campbell, and P. Sharma. 1995. Water
movement in desert soil traced by hydrogen and oxygen isotopes,
chloride, and chloride-36, southern Arizona. J. Hydrol. 168:91–110.
doi:10.1016/0022-1694(94)02646-S
Lyford, F.P., and H.K. Qashu. 1969. Infiltration rates as affected by desert vegetation.
Water Resour. Res. 5:1373–1376. doi:10.1029/WR005i006p01373
Martin, C.A., K.A. Peterson, and L.B. Stabler. 2003. Residential landscaping in
Phoenix, Arizona, U.S.: Practices and preferences relative to covenants,
codes, and restrictions. J. Arboric. 29(1):9–17.
Mayer, A.L., W.D. Shuster, J.J. Beaulieu, M.E. Hopton, L.K. Rhea, A.H. Roy,
and H.W. Thurston. 2012. Building green infrastructure via citizen
participation: A six-year study in the Shepherd Creek (Ohio, USA).
Environ. Pract. 14:57–67. doi:10.1017/S1466046611000494
Nascimento, N.O., J.B. Ellis, M.B. Baptista, and J.-C. Deutsch. 1999. Using
detention basins: Operational experience and lessons. Urban Water
1:113–124. doi:10.1016/S1462-0758(00)00009-1
Neff, J.C., A.P. Ballantyne, G.L. Farmer, N.M. Mahowald, J.L. Conroy, C.C.
Landry, et al. 2008. Increasing eolian dust deposition in the western
United States linked to human activity. Nat. Geosci. 1:189–195.
doi:10.1038/ngeo133
NRCS. 2006. Land Resource Regions and Major Land Resource Areas of the
United States, the Caribbean, and the Pacific Basin. Agric. Handbk. 296.
U.S. Gov. Print. Office, Washington, DC.
NRCS. 2013. 627.08(d)(2): Pedon descriptions. In: National soil survey
handbook, Title 430-VI. NRCS, Washington, DC.
Padowski, J.C., and J.W. Jawitz. 2012. Water availability and vulnerability of
225 large cities in the United States. Water Resour. Res. 48:W12529.
doi:10.1029/2012WR012335
Redman, C.L., and A.P. Kinzig. 2008. Water can flow uphill: A narrative of
www.soils.org/publications/sssaj
central Arizona. In: C.L. Redman and D.R. Foster, editors, Agrarian
landscapes in transition: Comparisons of long-term ecological and cultural
change. Oxford Univ. Press, New York. p. 238–271.
Reheis, M.C. 2006. A 16-year record of eolian dust in southern Nevada and
California, USA: Controls on dust generation and accumulation. J. Arid
Environ. 67:487–520. doi:10.1016/j.jaridenv.2006.03.006
Reynolds, R.L., M. Reheis, J. Yount, and P. Lamothe. 2006. Composition of
aeolian dust in natural traps on isolated surfaces of the central Mojave
Desert: Insights to mixing, sources, and nutrient inputs. J. Arid Environ.
66:42–61. doi:10.1016/j.jaridenv.2005.06.031
Robinson, D.A., I. Fraser, E.J. Dominati, B. Davídsðóttir, J.O.G. Jónsson, L.
Jones, et al. 2014. On the value of soil resources in the context of natural
capital and ecosystem service delivery. Soil Sci. Soc. Am. J. 78:685–700.
doi:10.2136/sssaj2014.01.0017
Schoeneberger, P.J., and D.A. Wysocki. 2005. Hydrology of soils and deep
regolith: A nexus between soil geography, ecosystems and land management.
Geoderma 126:117–128. doi:10.1016/j.geoderma.2004.11.010
Shuster, W.D., S. Dadio, P. Drohan, R. Losco, and J. Shaffer. 2014. Residential
demolition and its impact on vacant lot hydrology: Implications for the
management of stormwater and sewer system overflows. Landsc. Urban
Plan. 125:48–56. doi:10.1016/j.landurbplan.2014.02.003
Soil Survey Staff. 1993. Soil survey manual. Agric. Handbk. 18. U.S. Gov. Print.
Office, Washington, DC.
Soil Survey Staff. 2010. Keys to soil taxonomy. 11th ed. NRCS, Washington, DC.
Walsh, C.J., T.D. Fletcher, and M.J. Burns. 2012. Urban stormwater runoff:
A new class of environmental flow problem. PLoS ONE 7:E45814.
doi:10.1371/journal.pone.0045814
Western Regional Climate Center. 2013a. PHOENIX WSFO AP, ARIZONA
(026481): Period of record monthly climate summary. WRCC, Reno,
NV. http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl?az6481 (accessed 24
Apr. 2014).
Western Regional Climate Center. 2013b. APACHE JUNCTION 5 NE,
ARIZONA (020288): Period of record monthly climate summary.
WRCC, Reno, NV. http://www.wrcc.dri.edu/cgi-bin/cliMAIN.
pl?az0288 (accessed 24 Apr. 2014).
Zhang, R. 1997. Determination of soil sorptivity and hydraulic conductivity
from the disk infiltrometer. Soil Sci. Soc. Am. J. 61:1024–1030.
doi:10.2136/sssaj1997.03615995006100040005x
∆