Quantification and Implications of Soil Losses from Commercial Sod Production

Abstract

Commercial sod farms occupy about 1.62 x 10(3) km(2) of the landscape of the United States. Land managers generally consider sod farms on an equal footing with other, sustainable agricultural land uses. We measured soil losses associated with sod harvesting in farms in the northeastern United States. Sod harvest resulted in soil losses ranging from 74 to 114 Mg ha(-1) yr(-1), considerably higher than the tolerable soil loss of 6.7 Mg ha(-1) yr(-1). Soil losses were proportional to time under sod production, with soil removal rates of 0.833 cm yr(-1). We estimate that sod harvesting in the United States results in the net, permanent loss of 12.0 to 18.7 Tg of agriculturally productive soil from sod farms-and associated ecosystem services-every year. The soil losses reported here have important implications in terms of land use planning, transactions involving the purchase of development rights, and tax deductions for soil depletion.

Full text

892
SSSAJ: Volume 74: Number 3 • May–June 2010
Soil Sci. Soc. Am. J. 74:892–897
Published online 10 Feb. 2010
doi:10.2136/sssaj2009.0239
Received 24 June 2009.
*Corresponding author (jamador@uri.edu).
© 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.
Quanti cation and Implications of Soil
Losses from Commercial Sod Production
Soil & Water Management & Conservation
C
ommercial sod production is an emerging agricultural industry worldwide as
a result of rising a uence in developed and developing countries (Aldous et
al., 2007). In the United States, sod is produced in all 50 states, with approximately
162,000 ha dedicated to sod production (Morris, 2006)—an area comparable to
that planted to tobacco (Nicotiana tabacum L.) (National Agricultural Statistics
Service, 2008)—and a total annual value greater than US$3.1 billion (Haydu et
al., 2006). Turfgrass provides environmental services that include amelioration of
erosion, glare, noise and air pollution, and heat buildup in urban and suburban
landscapes (Beard and Green, 1994). Although these bene ts may be experienced
at both the sod farms and in areas where the sod is eventually planted, they may
come at the expense of mining of soil as an unintended consequence of sod harvest-
ing practices. Furthermore, because soil associated with sod harvest is transported
to nonagricultural areas, there is a net loss of agriculturally productive soil from
the landscape.
In commercial sod production, conventional harvesting (by sod cutting) in-
volves removing a layer of soil just below the thatch layer of the turfgrass, which can
result in the permanent depletion of soil resources. To date, anecdotal reports from
trade organizations, government agencies, and university researchers regarding soil
losses from sod harvesting o er con icting information, with studies suggesting
that sod harvesting does not result in net soil loss (Charbonneau, 2003; Skogley
and Hesseltine, 1978; Turf Resource Center, 2009) and that it does causes signi -
cant soil loss (Carr, 1996; Sheard and Van Patter, 1978).  e striking di erences
in surface elevation commonly observed between sod  elds and historic cemeteries
found within these  elds suggest that soil losses are substantial (Fig. 1). Because
sod farms are generally considered on an equal footing with other agricultural land
uses by land managers, we suggest that the long-term e ects of sod farming on the
loss of agriculturally productive soil and associated ecosystem services need to be
quanti ed and incorporated in land use planning decisions.
David Millar
Mark Stolt
José A. Amador*
Dep. of Natural Resources Science
Coastal Institute–Kingston
Univ. of Rhode Island
Kingston, RI 02881
Commercial sod farms occupy about 1.62 × 10
3
km
2
of the landscape of the United States. Land managers
generally consider sod farms on an equal footing with other, sustainable agricultural land uses. We measured soil
losses associated with sod harvesting in farms in the northeastern United States. Sod harvest resulted in soil losses
ranging from 74 to 114 Mg ha
−1
yr
−1
, considerably higher than the tolerable soil loss of 6.7 Mg ha
−1
yr
−1
. Soil
losses were proportional to time under sod production, with soil removal rates of 0.833 cm yr
−1
. We estimate that
sod harvesting in the United States results in the net, permanent loss of 12.0 to 18.7 Tg of agriculturally productive
soil from sod farms—and associated ecosystem services—every year.  e soil losses reported here have important
implications in terms of land use planning, transactions involving the purchase of development rights, and tax
deductions for soil depletion.

SSSAJ: Volume 74: Number 3 • May–June 2010
893
We conducted a study to determine the soil losses associated
with the conventional harvest of Kentucky bluegrass (Poa praten-
sis L.) sod, the most commonly harvested sod in the northeastern
United States. Estimates of soil loss were made using two inde-
pendent methods: (i) measurement of the mass of mineral soil
associated with harvested sod, and (ii) measurement of changes
in depth to sand and gravel deposits in sod  elds in operation for
di erent amounts of time relative to adjacent forested land.
MATERIALS AND METHODS
Study Sites
Four commercial sod farms (designated A, B, C, and D) all in
Washington County, RI, were studied. Sampling was limited to ar-
eas with slopes ≤3% that were mapped as Agawam (a coarse-loamy
over sandy or sandy-skeletal, mixed, active, mesic Typic Dystrudept),
Bridgehampton (a coarse-silty, mixed, active, mesic Typic Dystrudept),
and En eld (a coarse-silty over sandy or sandy-skeletal, mixed, active,
mesic Typic Dystrudept) soils (Soil Survey Sta , 2007).  ese soils are
characterized by glacio uvial deposits (strati ed sands and gravels; out-
wash) overlain by silt loam (loess) and  ne sandy loam textured eolian
deposits (Fig. 2) (Rector, 1981). Agawam, Bridgehampton and En eld
series soils within the 0 to 3% slope range constitute the highest per-
centage (5, 29, and 31%, respectively) of land under sod production in
Rhode Island.
 e sod  elds we studied had been in production between 4 and
40 yr at the time of sampling (Table 1). Previous land uses included
potato (Solanum tuberosum L.) farming, pasture, and woodlands.
Sampling areas at each farm ranged from 19 to 183 ha. Land use history
and number of years in sod production were obtained from interviews
with sod farm operators.
Sampling
Sod strips were collected from all the sod farms on the same day
in September 2007, at least 1 wk a er the last irrigation event to en-
sure equivalent moisture conditions among sites. At each farm, three
sod strips (0.5 by 1.8 m) were collected during routine harvesting along
a 150-m transect with a 50-m interval between each strip. Sod strips
harvested at the beginning of each pass tend to be variable, and may not
be representative of strips collected along the rest of the pass.  us, all
sod strips collected in this study were done so a er several strips were
harvested at the beginning of each pass by the sod harvester.
 e sod strips were carefully placed in individual plastic bags
to minimize soil loss during transport, and processed on the day of
Fig. 1. Difference in surface elevation between a sod fi eld and a
historical cemetery found within this sod fi eld in southern Rhode
Island. Elevated cemeteries like the one pictured here are found in
many of the sod fi elds in this area. Similar elevated cemeteries were
not observed within other adjacent land uses, suggesting that the
difference in elevation is related to soil loss.
Fig. 2. Morphological features representative of the soil mapping
units used in this study. Arrows indicate the interface between the
eolian mantle and stratifi ed sand and gravel deposits in soil from a
forested area and adjacent sod fi eld.

894 SSSAJ: Volume 74: Number 3 • May–June 2010
collection.  ree adjoining, 30- by 30-cm squares were traced on the un-
derside (soil side) of each strip and a 10-cm-diam. cup cutter was used to
remove a sod plug from each square section.  e sod plugs were placed
in sealable plastic bags and stored at −10°C.
Estimates of Soil Loss from Analysis of Sod Strips
 e mass of mineral soil associated with sod strips was determined
by loss-on-ignition (Soil Survey Laboratory Sta , 2004). Each sod plug
was placed in a mu e furnace at 550°C for 4 h.  e mineral content of
the plant material portion of the harvested sod was determined to be
14 g kg
−1
of total sod piece mineral content by loss-on-ignition. Inputs
of mineral material through incorporation of biosolids, commonly
used as fertilizer in this area, were also determined. Biosolids from New
England Organics (Portland, ME), the main source of biosolids used
on Rhode Island sod farms, had a mean mineral content of 350 g kg
−1
as determined by loss-on-ignition. Based on an estimated biosolids ap-
plication rate of 6.72 Mg ha
−1
yr
−1
, the input of mineral material to the
soil was estimated to be 2.35 Mg ha
−1
yr
−1
.
 e net soil removal rate by sod harvesting (M
net
, Mg ha
−1
yr
−1
)
was calculated using the following equation:
net mineral mineral
MM Mαβ=− −
[1]
where M
mineral
is the total mass of mineral material removed in sod
(Mg ha
−1
yr
−1
), α is the fraction of M
mineral
associated with plants, and
β is the mass of mineral material added in biosolids (Mg ha
−1
yr
−1
).
 e thickness of soil removed as a function of the number of sod har-
vests (D
soil
, m) based on M
net
was calculated using the following equation:
net
soil
b
M
Dn
ρ
⎛⎞
=
⎜⎟
⎝⎠
[2]
where n is the number of harvests, M
net
is the soil loss rate
(Mg ha
−1
yr
−1
) (from Eq. [1]), and ρ
b
is the soil bulk density (Mg m
−3
)
measured from the soil surface (just below the turfgrass thatch layer) to a
depth of 10 cm at 10 randomly selected points using a 30-cm-long by
5-cm-diam. split core sampler.
Estimates of Soil Loss from Morphological
Soil Properties
Depth to the contact between the eolian deposits used for sod
production and the underlying sand and gravel (Fig. 2) was measured
in areas mapped as En eld soils at Sod Farms A, B, and D in  elds
that had been in commercial sod production for approximately 3, 20,
and 30 yr, respectively. Measurements were made using a bucket auger
along two, 50-m, parallel transects spaced 20 m apart that extended
across the sod  elds and into the adjacent forest (Fig. 3).  e transects
were approximately perpendicular to the border between the sod
 eld and forest. Samples were taken every 5 m along each transect (11
sampling points per transect). Additionally, samples were taken along
two, 50-m, parallel transects, one in the forest and the other in the
sod  eld.  e transects were 10 m from the forest–sod  eld boundary
and ran approximately parallel to this boundary. Samples were taken
at 5-m intervals along each transect (11 sampling points per transect).
Di erences in bulk density in the top 10 cm of the sod  elds and
adjacent forests caused by land clearing were used to account for soil
thickness di erences due to soil collapse.  e average bulk density from
0 to 10 cm on sod  elds and forest sites was 1.34 and 0.77 Mg m
−3
, re-
spectively.  e di erence in bulk density of 0.57 Mg m
−3
between these
two land uses translates into a 7.4-cm decrease in soil thickness, account-
ing for 90% of the 8.3-cm discrepancy between calculated and observed
values observed in  elds in production for 3 yr.  us, soil collapse was
accounted for by subtracting the 8.3 cm due to collapse from the di er-
ence in soil thickness between forest and sod  elds.
Statistical Analyses
A one-way analysis of variance was used to compare soil loss rates
among sod farms based on sod strip analysis, with means separation ac-
complished using Tukey’s highly signi cant di erence test. Di erences
between sod  elds and adjacent forests in terms of depth to glacio uvial
deposits were evaluated using Student’s t-test.
RESULTS AND DISCUSSION
Soil removal rates by sod harvesting based on the mass
of mineral soil in the harvested sod ranged from 74 to
114 Mg ha
−1
yr
−1
, with signi cant di erences observed between
the highest (Sod Farm D) and lowest (Sod Farm C) values (Fig.
4).  e only apparent di erence between these two farms is the
number of crops harvested: three sod crops were harvested in
Sod Farm D, whereas Sod Farm C had ?30 crops harvested.
Table 1. Time in sod production, sampling area, and previous
land use of sod farms.
Sod
farm
Time in sod
production
Sampling area Previous land use
yr ha
A 33 43 potato
B 20, 30 79 woodland
C 39 183 pasture, potato
D 3, 35 19 woodland, pasture
Fig. 3. Sampling scheme used to measure depth to the interface between
the eolian materials used for sod production and the underlying sand and
gravel deposits.

SSSAJ: Volume 74: Number 3 • May–June 2010
895
 ese data may be interpreted to indicate that the amount of soil
removed per harvest decreases as the total number of harvests
increases. Sod Farm A had about the same number of harvested
crops as Sod Farm C (?30), however, and its soil removal rate
was not statistically di erent from that of either Farm C or Farm
D. Furthermore, the topography and soil mapping units were
the same for both Farms C and D. It appears that variables other
than those measured in our study were responsible for the di er-
ence observed between Farms C and D.
 e rates of soil removal reported here are more than an or-
der of magnitude higher than the tolerable soil loss (T value) of
6.7 Mg ha
−1
yr
−1
published by the NRCS for these soils (Rector,
1981).  e T value refers to the maximum rate of soil erosion that
may occur and still permit a high level of crop productivity to be
obtained economically and inde nitely (Schertz and Nearing,
2002). Assuming an average of one harvest per year, these soil
loss values are well within the 61 to 105 Mg ha
−1
harvest
−1
re-
ported by Skogley and Hesseltine (1978) for sod grown in south-
ern New England.
Direct, independent measurements of the depth to sand
and gravel deposits were made in sod  elds in production for ap-
proximately 3, 20, and 30 yr, and compared with those of adja-
cent forested areas (Fig. 5). Data from transects on sod farms in
production for di erent lengths of time all showed a statistically
signi cant, lower thickness of the eolian mantle relative to those
of adjacent forested areas, suggesting that substantial soil loss has
occurred as a result of sod production (Fig. 6). In addition, the
longer a sod  eld had been in production, the more pronounced
the di erences in thickness were relative to the adjacent forest.
 e mass of mineral soil removed during harvest (M
net
, Eq.
[1]) and ρ
b
values were used to develop an independent estimate
of the thickness of soil removed.  is resulted in estimates of
2.8, 14.3, and 27.2 cm of soil removed from sod  elds in produc-
tion for approximately 3, 20, and 30 yr,
respectively (Fig. 7). A er accounting
for soil collapse (from di erences in ρ
b
between forest and sod  eld soils), esti-
mates of soil loss based on sod strip analy-
sis (Fig. 4) were similar to those obtained
from the measurements of solum thick-
ness in sod farms and adjacent forest (Fig.
5 and 6), suggesting that both are reason-
able methods for determining soil removal
rates on land used for sod production.
 e data from both sets of estimates
were  t to a linear model (r
2
= 0.9868)
using the following equation:
it
DRtD=+
[3]
where D
t
is the depth of soil (cm) at time
t (yr), R is the rate of soil thickness loss
(cm yr
−1
), and D
i
is the depth of soil
(cm) before sod production. Based on
our data, R is −0.833 cm yr
−1
and D
i
is 0.09 cm.  is value of R is lower than
the 1.2 to 1.6 cm of soil that is reported
by some to be removed with every sod
harvesting event (Cockerham, 2007) but
considerably higher than the mean value
for soil loss due to erosion from conven-
tional agriculture of 0.394 cm yr
−1
re-
Fig. 4. Mean (n = 9) net soil removal rates from harvesting in four sod
farms in Washington County, RI. Values with the same letters were
not signifi cantly different (P ≤ 0.05). Bars represent one standard
deviation. Dashed line indicates the soil loss tolerance (T) value
(Rector, 1981).
Fig. 5. Transects of depth to sand and gravel deposits in a forest and adjacent sod fi eld for fi elds in sod
production for approximately 3, 20, and 30 yr in Washington County, RI. Data for transects that were
perpendicular (left panel) and parallel (right panel) to the boundary between the forest and sod fi eld
are shown. Note that the depth to sand and gravel deposits is closer to the soil surface in sod fi elds.

896 SSSAJ: Volume 74: Number 3 • May–June 2010
ported by Montgomery (2007). Given an average depth to sand
and gravel deposits of 83 cm for this soil mapping unit (Fig. 5),
the model predicts complete loss of the eolian mantle 96 yr af-
ter the onset of sod production (Fig. 7). Because sod  elds are
traditionally plowed to a depth of 30 cm, however, management
problems associated with the incorporation of sand and gravel
materials with eolian deposits can be expected a er ?60 yr of
sod production.
 e time course for management problems to develop
may be unique to soils with the speci c morphological features
studied here (e.g., eolian mantle over strati ed sand and gravel
deposits formed in a glaciated landscape; Fig. 2). Regardless of
geographic location or soil morphology, however, soil removal
is an intrinsic part of conventional sod harvesting, such that sub-
stantial soil losses will always be associated with these operations.
Because the amount of soil removed is constrained by the thick-
ness of the cut sod—and most sod producers try to minimize this
thickness to lower transportation costs (Perez et al., 1995; Turf
Resource Center, 2009)—the rates of soil removal reported here
probably represent a lower limit of soil removed by conventional
sod harvesting operations.
Our results have implications for the long-term sustainabil-
ity of sod production as it is currently practiced in the United
States, and to the function of commercial sod farms in rural and
suburban landscapes. Land for sod production and open space
are both in high demand in suburban areas, the former to pro-
vide landscaping materials for new construction, the latter be-
cause of a desire to conserve the rural character of the landscape
(Geoghegan, 2002; Machado et al., 2006). As such, sod farms
are considered desirable as open space, on a par with other agri-
cultural operations and with woodlands. Although commercial
sod farms may meet the de nition of open space, the soil min-
ing aspects of sod production do not appear to be taken into
account by land use managers and environmental regulators.
Conventional sod farming clearly results in considerable net per-
manent depletion of agriculturally productive soil, a nonrenew-
able resource, relative to other agricultural and nonagricultural
land uses.  is should be an important element in making deci-
sions about the function of sod farming in the land use planning
process. Clearly those areas to which the sod—and associated
soil—is transported and planted accrue a number of bene ts,
including amelioration of erosion, glare, noise and air pollution,
and heat buildup (Beard and Green, 1994). Such bene ts, how-
ever, come at the expense of a permanent loss of agriculturally
productive soil from sod farms.
Agricultural land in suburban areas is o en the subject of
transactions known as purchase of development rights (Daniels,
1991). In these transactions, a government or private organiza-
tion interested in open space preservation purchases the rights
to develop a parcel of land, with the seller maintaining owner-
ship of the land but bound by contract not to develop it, and
the purchaser of the development rights being responsible for
ensuring the owner adheres to the contract. Such transactions
are common in areas with high development pressure due to de-
mographic and market forces (Daniels, 1991). When the intent
is to preserve productive agricultural land as part of the suburban
landscape, sod farming falls short of the requirement for long-
term sustainability of current practices that is part of the criteria
for the purchase of development rights by federal (e.g., Federal
Agriculture Improvement and Reform Act of 1996, 1996) and
state agencies charged with the stewardship of soil resources. On
the other hand, if the intent is to simply preserve open space, the
soil mining associated with sod farming may not interfere with
the ful llment of this function.  ere generally is an implicit as-
sumption by the entities acquiring development rights and land
managers, however, that current agricultural practices will not
have signi cant, long-term e ects on the provision of ecosystem
services by the purchased land—an assumption that, in light of
our results, does not appear to be warranted for sod farms. Soil
Fig. 6. Mean (n = 9–11) values of depth to sand and gravel deposits
in forest and the adjacent sod fi eld for fi elds in sod production for
approximately 3, 20, and 30 yr in Washington County, RI, based
on transect data (Fig. 5). Bars represent one standard deviation.
*Signifi cant difference between forest and sod fi eld (P ≤ 0.05).
Fig. 7. Relationship between depth to sand and gravel deposits and
years in sod production based on measurement of mineral soil loss
with harvested sod and on direct measurements of depth to sand and
gravel materials. Data from both sets of measurements were fi t to a
linear model (r
2
= 0.9868) using the equation: D
t
= Rt + D
i
(Eq. [3]),
where D
t
= depth of soil (cm) at time t (yr), R = rate of soil thickness
loss (cm yr
−1
) and D
i
= depth of soil (cm) before sod production. For
these data, R is −0.833 cm yr
−1
and D
i
is 0.09 cm.

SSSAJ: Volume 74: Number 3 • May–June 2010
897
loss from sod harvesting is inextricably linked with a diminishing
capacity of the land to carry out many of its functions within
the landscape, including crop production, water in ltration and
storage, retention and recycling of nutrients, C sequestration,
and consumption of greenhouse gases. Sod farms do provide eco-
systems services, such as reduced soil erosion, C sequestration,
and wildlife habitat (Beard and Green, 1994; Milesi et al., 2005).
Unlike agricultural practices that preserve the soil and its func-
tions over the long term, however, the bene ts of commercial sod
farming are  nite, constrained by the rate at which the soil on
which the sod is grown is depleted by harvesting.
 e permanent loss of soil also presents problems in the
context of land leased for sod farming, in which the owner is
not compensated  nancially for the loss of soil resources as a re-
sult of sod harvesting. In addition, there are implications with
regard to whether tax deductions may be taken by sod famers,
or land owners that lease land to sod farmers, for soil deple-
tion. According to Revenue Ruling 79-267 of the U.S. Internal
Revenue Service (Internal Revenue Service, 1979): “ e amount
of naturally occurring soil exhausted with each sod harvest can-
not be established, and a deduction for cost depletion is not al-
lowable.” Contrary to these assertions, our data show that the
amount of naturally occurring soil removed with each sod har-
vest can indeed be established, suggesting that reconsideration of
this ruling may be warranted.
Assuming a total area of sod harvested each year of 162,000 ha
(Morris, 2006; National Agricultural Statistics Service, 2008) and
based on the soil removal rates reported in this study (Fig. 4), we
estimate that between 12.0 and 18.7 Tg of soil is permanently
removed from agricultural land by sod harvesting in the United
States every year. Development and enforcement of criteria that
explicitly require preservation of soil resources as a condition for
the purchase of development rights by private and government
entities would contribute substantially to preventing the loss of
soil resources from commercial sod farming. Soil loss associated
with sod farming may also be ameliorated by the adoption of sus-
tainable sod farming practices, such as growing sod on recyclable
organic materials over plastic sheeting (Decker, 2001), and a shi
to more traditional means of establishing turf, such as seeding or
hydroseeding, by landscaping professionals and consumers.
CONCLUSIONS
Our results show that soil removal associated with the con-
ventional harvesting of Kentucky bluegrass blend sod results in
a net loss of agriculturally productive soil that is more than an
order of magnitude larger than the tolerable soil loss value.  e
data also show that soil loss is directly proportional to years un-
der sod production.  is information needs to be taken into con-
sideration when evaluating the costs and bene ts of sod farming.
Speci cally, our results suggest that decisions concerning the
land use planning process, transactions involving the purchase
of development rights, and tax credits for soil depletion should
incorporate the soil loss associated with conventional sod har-
vesting practices.
ACKNOWLEDGMENTS
 is research was funded in part by a grant from the USDA-NRCS to
M. Stolt and J. Amador. We wish to thank the anonymous land owners
for access to their property.  is paper is contribution no. 5232 from the
Rhode Island Agricultural Experiment Station.
REFERENCES
Aldous, D.E., K.W. McAuli e, and N.M. Power. 2007.  e turfgrass industry:
Australia, New Zealand, and the Asia-Paci c region. Chron. Hortic.
47:17–19.
Beard, J.B., and R.L. Green. 1994.  e role of turfgrasses in environmental
protection and their bene ts to humans. J. Environ. Qual. 23:452–460.
Carr, M.Y. 1996. Metal levels in Rhode Island municipal sewage sludges and their
resulting appropriateness for agricultural use by sod farms. Honors thesis.
Brown Univ., Providence, RI.
Charbonneau, P. 2003. Soil loss during sod production. Available at www.omafra.
gov.on.ca/english/crops/facts/info_sodprod.htm#loss (veri ed 19 Jan.
2010). Ontario Ministry of Agriculture, Food and Rural A airs, Ottawa.
Cockerham, S.T. 2007. Turfgrass culture for sod production. p. 139–150. In M.
Pessarakli (ed.) Handbook of turfgrass management and physiology. CRC
Press, Boca Raton, FL.
Daniels, T.L. 1991.  e purchase of development rights. J. Am. Plann. Assoc.
57:421–431.
Decker, H.F. 2001. Producing sods over plastic in soilless media. Hortic. Rev.
27:317–352.
Federal Agriculture Improvement and Reform Act of 1996, Pub. L. no. 104–127,
§388, 110 Stat. 888.
Geoghegan, J. 2002.  e value of open spaces in residential land use. Land Use
Policy 19:91–98.
Haydu, J.J., A.W. Hodges, and C.R. Hall. 2006. Economic impacts of the turfgrass
and lawncare industry in the United States. Food and Resource Economics
Dep., Inst. of Food and Agric. Sci., Univ. of Florida, Gainesville.
Internal Revenue Service. 1979. 142.181 Soil removed with sod harvest. Vol. Rev.
Rul. 79–267, 1979–2 C.B. 243.
Machado, E.A., D.M. Stoms, and J. Kreitler. 2006. Prioritizing farmland
preservation cost-e ectively for multiple objectives. J. Soil Water Conserv.
61:250–258.
Milesi, C., S. Running, C. Elvidge, J. Dietz, B. Tuttle, and R. Nemani. 2005.
Mapping and modeling the biogeochemical cycling of turf grasses in the
United States. Environ. Manage. 36:426–438.
Montgomery, D.R. 2007. Soil erosion and agricultural sustainability. Proc. Natl.
Acad. Sci. 104:13268–13272.
Morris, K. 2006.  e National Turfgrass Research Initiative. Green Sect. Record,
Sept.–Oct., p. 26–30.
National Agricultural Statistics Service. 2008. Acreage. Available at
usda.mannlib.cornell.edu/usda/current/Acre/Acre-06-30-2008.pdf
(veri ed 19 Jan. 2010). NASS, Washington, DC.
Perez, A., J. Harwood, D. Johnson, A. Somwaru, and G. Zepp. 1995. Turfgrass
sod: An economic assessment of the feasibility of providing multiple-peril
crop insurance. USDA Econ. Res. Serv., Washington, DC.
Rector, D.D. 1981. Soil survey of Rhode Island. U.S. Gov. Print. O ce,
Washington, DC.
Schertz, D.L., and M.A. Nearing. 2002. Erosion tolerance/soil loss tolerances.
p. 640–643. In R. Lal (ed.) Encyclopedia of soil science. Marcel Dekker,
New York.
Sheard, R.W., and M. Van Patter. 1978. Soil modi cation during nursery sod
production. Dep. of Land Resource Science, Univ. of Guelph, Guelph,
ON, Canada.
Skogley, C.R., and B.B. Hesseltine. 1978. Soil loss and organic matter return in
sod production. Univ. of Rhode Island, Kingston.
Soil Survey Laboratory Sta . 2004. Soil survey laboratory methods manual. Soil
Surv. Invest. Rep. 42. Version 4.0. U.S. Gov. Print. O ce, Washington, DC.
Soil Survey Sta . 2007. O cial soil series descriptions. Available at soils.usda.
gov/technical/classi cation/osd/index.html (veri ed 19 Jan. 2010).
NRCS, Washington, DC.
Turf Resource Center. 2009. What is turfgrass sod? Available at www.
turfgrasssod.org/images/documents/040120091843151.pdf (veri ed 19
Jan. 2010). Turf Resour. Ctr., East Dundee, IL.