Challenges in Sustainability | 2016 | Volume 4 | Issue 1 | Pages 39–53

DOI: 10.12924/cis2016.04010039

ISSN: 2297–6477

Challenges in

Sustainability

Research Article

Siting Urban Agriculture as a Green Infrastructure Strategy for

Land Use Planning in Austin, TX

Charles M. Rogers* and Colleen C. Hiner

Department of Geography, Texas State University, San Marcos, TX, USA

* Corresponding author: E-Mail: [email protected]; Tel.: +1 4794458334

Submitted: 1 February 2016 | In revised form: 19 July 2016 | Accepted: 19 July 2016 |

Published: 22 August 2016

Abstract:

Green infrastructure refers to a type of land use design that mimics the natural water cycle by

using the inﬁltration capacities of vegetation, soils, and other natural processes to mitigate stormwater

runoff. As a multifunctional landscape, urban agriculture should be seen as a highly beneﬁcial tool for urban

planning not only because of its ability to function as a green stormwater management strategy, but also

due to the multiple social and environmental beneﬁts it provides. In 2012, the city of Austin adopted a major

planning approach titled the “Imagine Austin Comprehensive Plan” (IACP) outlining the city’s vision for

future growth and land use up to 2039. The plan explicitly addresses the adoption of green infrastructure as

a target for future land use with urban agriculture as a central component. Addressing this area of land use

planning will require tools that can locate suitable areas within the city ideal for the development of green

infrastructure. In this study, a process was developed to create a spatially explicit method of siting urban

agriculture as a green infrastructure tool in hydrologically sensitive areas, or areas prone to runoff, in east

Austin. The method uses geospatial software to spatially analyze open access datasets that include land

use, a digital elevation model, and prime farmland soils. Through this method a spatial relationship can be

made between areas of high surface runoff and where the priority placement of urban farms should be sited

as a useful component of green infrastructure. Planners or geospatial analysts could use such information,

along with other signiﬁcant factors and community input, to aid decision makers in the placement of urban

agriculture. This spatially explicit approach for siting potential urban farms, will support the integration of

urban agriculture as part of the land use planning of Austin.

Keywords: GIS; green infrastructure; urban agriculture; urban planning; watershed protection

1. Introduction

1.1. Green Infrastructure and Urban Agriculture

The federal Clean Water Act (CWA) of 1972 provided a

basic framework for the regulation of pollutant discharges

with the intent of protecting water quality and human health

in the United States (U.S.) [

]. To further the protective

beneﬁts of the CWA, the Environmental Protection Agency

(EPA) enacted its Combined Sewer Overﬂow Control Policy

in 1994 [

]. This policy initiative established guidelines by

which municipalities could better manage environmentally

harmful stormwater pollution events that occurred as a re-

sult of infrastructural challenges in handling both sanitary

sewage and stormwater runoff in the same sewer system

during precipitation events. In response, many cities im-

 2016 by the authors; licensee Librello, Switzerland. This open access article was published

under a Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/).

librello

plemented green infrastructure measures to help mitigate

urban stormwater runoff as an alternative to expensive wa-

ter main upgrade projects. Green infrastructure can be

deﬁned as the management of runoff through the use of

natural systems, or engineered systems that act as nat-

ural systems, to allow stormwater to inﬁltrate the ground,

recharging the water table and decreasing run-off. Accord-

ing to the EPA, Green infrastructure uses vegetation, soils,

and natural processes to manage water and create health-

ier urban environments. At the scale of a city or county,

green infrastructure refers to the patchwork of natural ar-

eas that provides habitat, ﬂood protection, cleaner air, and

cleaner water. At the scale of a neighborhood or site, green

infrastructure refers to stormwater management systems

that mimic nature by soaking up and storing water [

]. While

this deﬁnition of green infrastructure is structured on water

management, there are broader deﬁnitions of green infras-

tructure that focus on aspects such as clean air, wildlife

habitat, and preservation of forests and grasslands.

With the industrialization of our food systems, most

cities, whether intentionally or not, have developed in such

a way that farms have become incompatible with the urban

environment. There is, however, a tremendous amount of

resurgent interest in urban farming and community garden-

ing across the country. In general terms, urban agriculture

refers to “the growing, processing, and distribution of food

and nonfood plant and tree crops and the raising of live-

stock, directly for the urban market, both within and on the

fringe of an urban area” ([

], p. 2500; [

], p. 4). As evi-

denced by the recent growth of municipal urban agriculture

land inventories as a method to integrate urban agriculture

into sustainable land management policy [

], we can infer

that municipalities value the multi-beneﬁcial attributes that

urban agriculture offers. Urban agriculture has been char-

acterized as a multifunctional land use because, through

its versatility as a landscape, it offers a range of beneﬁts

to densely populated areas [

]. For example, the ecolog-

ical functions of urban agriculture provide environmental

beneﬁts in the form of biodiversity, nutrient cycling (com-

post), or wastewater re-use (stormwater and greywater);

the cultural functions improve the quality of a neighborhood

or community through its visual appeal, recreational use,

or the provision of rare foods to immigrant communities;

the socio economic functions provide access to nutritious

fresh produce for underserved communities in food deserts

facing obesity and diabetes [4].

In fact, evidence suggests that incorporating appropriate

types of urban agriculture into the urban environment will

greatly improve the overall sustainability of U.S. cities [

The EPA, in collaboration with experts from academia, state

and local governments, and nonproﬁts, released a general

listing of the beneﬁts of urban agriculture:

•

Increases surrounding property values, beautiﬁes va-

cant properties, increases a sense of community, and

provides recreational and cultural uses [8].

•

Increases inﬁltration of rainwater, reducing stormwa-

ter overﬂows and ﬂooding, decreases erosion and

topsoil removal, improves air quality, and reduces

waste by the reuse of food and garden wastes as

organic material and compost [8].

•

Increases physical activity and educates new garden-

ers on the many facets of food production from food

security to nutrition and preparation of fresh foods [

While the positive ecological and environmental impacts

of urban agriculture have been acknowledged, minimal re-

search has been conducted to determine the feasibility of

urban agriculture as a green infrastructure strategy to man-

age stormwater. Green infrastructure strategies include

a list of practices that have been studied and quantiﬁed,

making their engineering and performance outcomes pre-

dictable and reliable [

]. A lack in such quantiﬁcation, or of

carefully researched sets of best management practices,

leads to a further lack in the promotion of urban agriculture

as a green infrastructure strategy. As Dunn ([

], p. 58)

argues: “Being able to quantify the effectiveness of green

infrastructure on a small scale is one way to promote regula-

tory and enforcement acceptance, which thereby enhances

its appeal to city ofﬁcials”. As a productive, multifunctional

landscape, urban agriculture offers answers to complex

infrastructural challenges that are facing many cities.

1.2. Urban Agriculture Land Inventories

Over the last decade a variety of municipalities have begun

the process of inventorying and assessing land to determine

the potential for a broad range of urban agricultural initia-

tives. This effort is driven by a diverse range of community

stakeholders, each with an interest in growing food locally,

including capturing the social and environmental beneﬁts

urban agriculture provides. Yet, despite the realized and

potential beneﬁts for individuals and communities, urban

agriculture is largely overlooked in urban and regional plan-

ning [

]. Instead of considering opportunities to preserve

farmland or to integrate it as part of a land use management

strategy in urban environments, agricultural landscapes are

often considered by land use planners as areas for future

development [4].

As part of the urban planning process, land inventories

have been used and recognized as a basic tool for land

suitability and site selection [

]. Many municipalities and

researchers are now employing this tool to promote urban

agriculture, integrating it into public policy and planning as a

land management strategy. Horsts’ (2011) paper “A Review of

Suitable Urban Agriculture Land Inventories” provides a brief

overview of urban agriculture land inventories in various cities

[

]. Inventories have been performed in Portland, Vancou-

ver, Seattle, Cleveland-Cuyahoga County, Detroit, Chicago,

Toronto, New York City, Cincinnati, Oakland, and San Fran-

cisco [

]. Some of the inventories primarily focus on public

vacant land and some open up their inventory to include

private lands or residential lands such as lawns or rooftops.

Decisions made as part of urban agricultural land in-

ventories typically require the input of multiple criteria in-

volving social, economic, and environmental considerations

[

]. The success of multiple criteria decision making

depends on an array of knowledgeable stakeholders mak-

ing informed decisions [

]. Using this model, inventories

rely upon the explicit use of an advisory committee with rep-

resentation from municipal staff, non-proﬁt organizations,

urban gardeners, and academic researchers. Many inven-

tories emulate the Portland model, wherein an advisory

committee guided the inventory throughout the process,

particularly in establishing evaluation criteria and reviewing

preliminary results. Mendes et al. [

] points to Portland’s

inventory as more successful when compared to Vancouver

due to the way Portland engaged many community part-

ners throughout the entire process, from design to imple-

mentation, while Vancouver lacked community involvement.

Mendes et al. [

] suggests this represents what the schol-

arly literature identiﬁes as a “networked movement”, where

participation in local decision making is inclusive and citi-

zen engagement is fully accepted, similar to Arnstein’s [

]

highest rungs in the ladder of citizen participation (Partner-

ship, Delegated Power, Citizen Control). Likewise, Oakland

established a Community Advisory Committee throughout

the project that provided citizen input in a number of ar-

eas: the location of potential sites, criteria for selection of

potential sites, and feedback on what type of information

would be useful in the ﬁnished inventory [

]. The best as-

set mapping has been described, especially in the case of

urban agricultural land inventories, to be a multi-stakeholder

process for action planning and policy design [4,6,18].

Many of the inventories have involved multiple stake-

holders in research and analysis phases, but have been

less inclusive when performing technical analysis includ-

ing Geographical Information System (GIS), aerial imagery

assessment, and site visits or ground-truthing [

]. The

key actors during this phase of inventory development have

been municipal staff, experts from food policy councils and

non-proﬁts, and students [

]. The partnership among

stakeholders from the city governments and student re-

searchers from local universities has created synergistic

opportunities. In Portland, Vancouver, Oakland, Seattle,

and Cleveland among others, graduate students worked

in partnerships with local municipalities to complete the

inventory, thereby gaining valuable experience while the

respective cities received cost-effective results.

Generally, the vacant land inventories followed a frame-

work of identifying vacant or open land by ownership type,

categorized as public or private, assigning suitability criteria,

then eliminating the unsuitable sites and highlighting the

best. Within this generalized framework, most inventories

created suitability criteria for urban agriculture addressing

physical and socioeconomic factors, assigned a ranking or

scoring system for criteria, presenting the study results as

publicly available reports [19].

The technical work for most inventories made use of

one or a combination of methods including aerial photo

assessment, GIS analysis, remote sensing, and site visits.

Some efforts relied extensively on GIS analysis or remote

sensing as in New York and Philadelphia [

]. The

potential exists for expanding the use of GIS and remote

sensing for urban agricultural land inventories from other

approaches developed by urban land use researchers. For

example, Myeong et al. [

] developed vegetation indexes

that estimate vegetation coverage and bare soil, criteria

that usually requires labor-intensive visual assessment, us-

ing multi-spectral and hyper-spectral data to identify urban

green areas [

]. Other inventories have made use of the

satellite imagery from the National Agricultural Imagery Pro-

gram (NAIP) overlaid onto the GIS city parcel data of vacant

land in order to select parcels containing potentially arable

land [

]. A more speciﬁc use of geospatial data in the Hal-

ifax, Canada inventory used the LiDAR data to model sun

exposure, an important aspect of most inventories suitability

criteria for potential urban agricultural sites [23].

As the practice of inventories is evolving, there are ar-

eas for improvement. In particular, many authors noted

limitations because of incomplete data, limited availability

of data, and frequency of updated data [

]. A

common limitation expressed by researchers stemmed from

low resolution and accuracy of the aerial imagery that re-

sulted in possible visual interpretation errors [

]. As

a result, researchers advised that post-analysis ground-

truthing of potential sites would be necessary to quality

check geospatial analysis [

]. Additionally, establish-

ing a measure of community support is also an area that

many inventories acknowledged needed further research

to identify variables such as cultural preferences, skills and

willingness, demand, resources, and the presence of local

leaders [

]. Inventories with community advisory

committees also noted that site visits, community outreach,

and consultation with city staff are necessary to evaluate

characteristics like soil quality, community interest, and se-

curity [

]. Soil quality, in particular, is an issue in

many urban areas that was recognized by most inventories

as an area for further research and analysis. In Oakland, for

example, the completed inventory inspired further research

about lead (Pb) contamination in the soil of potential urban

agricultural sites. This research assessed Pb levels at over

a hundred different sites identiﬁed in the inventory [26].

Beyond land identiﬁcation, land inventories have been

an effective tool to integrate urban agriculture into urban

policy and planning as a land management strategy [6]. As

a part of the planning process land inventories can identify

opportunities for urban agriculture initiatives that result in

positive changes. Some impacts have included increasing

awareness and political support for urban agriculture, ad-

vancing social and ecological sustainability, and enhancing

public involvement [

]. For example, Toronto, Seattle,

and Portland experienced notable changes resulting from

vacant land assessments. In Toronto, local zoning regula-

tions and guidelines were altered to help guide an increase

in urban agriculture [

]. In Seattle and Portland, the col-

laborative process increased community involvement and

inclusion of urban agriculture into city sustainability planning

[

]. Stakeholders have also built upon these assessments

and conducted more targeted in depth studies that relate

to issues of public health, economic development, food

security, and environmental sustainability [

]. As a tool,

land inventories do not have to function in isolation and can

be employed in conjunction with other strategies, such as

surveys or scenario planning, to advance municipal goals

such as stormwater management, reducing carbon emis-

sions, increasing food access, and supporting workforce

development [6].

Urban agriculture land inventories have been a useful

step for many cities in evaluating the potential for urban agri-

culture, though the process and resulting impacts are unique

to each city. The type of parcels considered, criteria applied,

and stakeholders involved differ depending on the objective.

The delineation of potential urban agriculture sites is only a

preliminary step in a long process of mapping the potential

of urban agriculture in a city [

]. That being said, such

demarcations can be a useful starting point as cities begin

to incorporate urban agriculture into community planning.

The tools involved with a land inventory have the potential

to facilitate participatory planning by bringing together com-

munity participants such as local residents, food activists,

academic researchers, and farmers with city planners and

government officials, in an effort to better plan and man-

age land use [

]. Nevertheless, the politics of negotiating

competing uses of land is inherently complex and difficult.

The viability of utilizing urban agriculture land inventories for

planning will depend upon identifying and negotiating the

varied interests of multiple stakeholders [23].

1.3.

Can Urban Agriculture identify as Green Infrastructure

in Austin, TX?

In 2012, the City of Austin adopted a major planning

approach titled the Imagine Austin Comprehensive Plan

(IACP) that outlines the vision for future growth and land

use in the city until 2039. The plan explicitly addresses the

adoption of green infrastructure as a targeted future land

use with urban agriculture speciﬁcally being included as a

component of the green infrastructure network.

The IACP identiﬁed eight priority programs, ranked in

order of importance, which will guide policy and implemen-

tation of the plan. The fourth priority proposes the use of

“green infrastructure to protect environmentally sensitive

areas and integrate nature into the city” [27]:

1. Invest in a compact and connected Austin.

2. Sustainably manage our water resources.

Continue to grow Austin’s economy by investing in our

workforce, education systems, entrepreneurs, and lo-

cal businesses.

Use green infrastructure to protect environmentally

sensitive areas and integrate nature into the city.

5. Grow and invest in Austin’s creative economy.

Develop and maintain household affordability through-

out Austin.

7. Create a Healthy Austin Program.

Revise Austin’s development regulations and pro-

cesses to promote a compact and connected city.

The building block actions listed as methods to imple-

ment the green infrastructure priority program include several

references to urban agriculture (below; emphasis added):

•

Integrate citywide and regional green infrastructure to

include such elements of preserves and parks, trails,

stream corridors, green streets, agricultural lands,

and the trail system into the urban environment and

the transportation network [27].

•

Incentivize appropriately-scaled and located green

infrastructure and public spaces, such as parks,

plazas, greenways, trails, urban agriculture and/or

open spaces in new development and redevelopment

projects [27].

•

Expand regional programs and planning for the pur-

chase of conservation easements and open space

for aquifer protection, stream and water quality pro-

tection, wildlife habitat conservation, and sustainable

agriculture [27].

•

Extend existing trail and greenway projects to create

an interconnected green infrastructure network that

includes such elements as preserves and parks, trails

stream corridors, green streets, greenways, and agri-

cultural lands that link all parts of Austin and connect

to nearby cities [27].

•

Permanently preserve areas of the greatest environ-

mental and agricultural value [27].

This study looks at east Austin where there is 1) an

established urban and peri-urban farm presence and 2)

high future development potential of agricultural land due

to the Austin-Round Rock Metro Area’s rapid growth (110

people per day) [

] and lack of land use constraints lead-

ing to urban sprawl [

]. Since the late 1990s, the City of

Austin has viewed the eastern side of the metro area as a

prime area of growth and development. Without an urban

growth boundary most of what the city has determined as

the “desired development zone” (DDZ) falls into much of

east Austin (Figure 1) and beyond into the Austin’s Extrater-

ritorial Jurisdiction (ETJ). The opening of the SH130 tollway,

conceived of as a north-south alternative route to interstate

35 on the east side of Austin, has further opened up large

areas of available land for new jobs, housing, and services

for Austin’s rapidly growing population. The Imagine Austin

plan conceives of growth corridors around the city, with the

SH130 corridor representing one of those growth corridors

in east Austin. According to 2012 land use data from the

City of Austin, 94,961 acres of undeveloped land, much of

it in agriculture, existed in the suburban portion of the DDZ.

Information from the 2015 State of the Food System Report,

released by the City of Austin’s Ofﬁce of Sustainability, put

the loss of farmland each day in Travis County at 9.3 acres

and a 25% loss in farmland over the last 11 years [29].

Figure 1.

The Decker Creek and Elm Creek Watershed

study areas, located in Austin, Texas in Travis County, re-

side completely in the Suburban DDZ, an area promoted

for growth and development by the City of Austin.

Furthermore, the hydrologic environments in east Austin

are already under much pressure as a result of being located

downriver from the impervious cover of the central business dis-

trict and at the bottom of several highly urbanized watersheds,

effectively making east Austin the hydrologic drain for the city

[

]. The creeks and watersheds within east Austin have ex-

perienced significant erosion problems and flooding associated

with the increasing development and impervious cover seen

upstream over the last half century [

]. Further exacerbating

the erosion problems have been urban development patterns in

the suburban DDZ that situated buildings on small lots close to

the creeks. The Austin City Council itself has underscored the

potential implications of increasing development and its effect

on eastern watersheds:

“The eastern watersheds, with broader floodplains and more

erosive soils, pose unique challenges to creek and floodplain

protection. Development is currently being placed in close prox-

imity to erosive creek banks in headwaters areas and creating

future problems requiring significant and unsustainable public

expense to maintain and repair. This development will likely

accelerate as build-out proceeds along SH-130” [31].

As a response to city-wide watershed threats, the City Coun-

cil updated the watershed protection policies in 2013 with Phase

One of the new Watershed Protection Ordinance (WPO). The

new WPO increased stream buffers and erosion hazard zones

to ensure development is not built to close to waterways for over

400 miles of smaller headwater streams [

]. Phase Two of

the WPO began in January 2015 with a Green Infrastructure

Working Group as part of the City’s land development code

rewrite process, called CodeNEXT, to discuss how to achieve

the Imagine Austin goals of “integrating nature into the city and

creating complete communities through revisions to our zoning

and environment codes’ [

]. Currently, green infrastructure

techniques in the Watershed Department’s manual for best man-

agement practices include rain gardens, bio- filtration areas, and

vegetative filter strips; however, there is no reference to urban

agriculture as a green infrastructure strategy for new develop-

ment or as a stormwater management strategy in the city.

Nevertheless, within the last twenty years there has been

a proliferation of urban agriculture in east Austin. A variety of

urban agriculture exists in the city, ranging from community gar-

dens to small market farms (defined as farms operating on less

than one acre within the city’s full jurisdictional boundary) to

larger urban farms (over one acre) within or on the periphery

of the city. In addition, the Austin City Council has encouraged

urban agriculture by authorizing relaxed zoning regulations for

private urban farms. For example, the current land code allows

for urban farming in all zoning classes, and, at this time, there

are 23 urban farms in the city of Austin, classified either as

“market farms” if less than one acre and as “urban farms” if

more than one acre, and 52 community gardens [

]. Further-

more, in 2009, the Austin City council created the Sustainable

Urban Agriculture and Community Garden Program with the

expressed purpose of streamlining the process for establishing

community gardens and sustainable agriculture on city land,

further endorsing urban agriculture in the city.

Much of the land in the suburban DDZ is in Texas Blackland

Prairie Geographic Region and contains prime farmland soils

according to the classification by the National Resources Con-

servation Service (NRCS) (Figure 2). The soils in this region

are known for being deep and rich with organic material, mak-

ing them valuable for agricultural use [

]. With the quality of

the soils, good drainage, and flat surface, prime farmland also

serves as ideal land for development. Indeed, many new devel-

opments are going to occur in prime farmland as a result of the

city’s priority to develop in the suburban DDZ and the substantial

amount of open space in that part of the city. As a matter of de-

bate, the question is how much of the land to develop and what

positive and negative impacts are expected to occur. However,

the City of Austin, through its Imagine Austin plan, has indicated

that preserving a portion of different classifications of open land

is important for making Austin a sustainable city. The function of

urban agriculture in this plan, thus, has significance in terms of

protecting and valuing the environmental, social, and economic

well-being of Austin. Most significantly for this study, though,

is the role of urban agriculture as a means to conserve prime

farmland and serve as green infrastructure in Austin.

1.4. Purpose of Study

Urban agriculture acts as a multifunctional landscape with a

variety of benefits including the ability to offset many facets of

environmental degradation including preventing excessive runoff

[

]. The increase of new development and predominance of

urban farms in east Austin, combined with the rise of green

infrastructure as a stormwater management focus in the city,

makes east Austin an ideal study area for evaluating the poten-

tial role of urban agriculture as a green infrastructure strategy.

While this is the narrow aim of this study, the true value of urban

agriculture as green infrastructure can only be better understood

by placing the results of this research into a wider systems frame-

work that encompasses the multiple environmental, economic,

and social benefits offered by urban farms.

Figure 2.

Prime Farmland Soils in Austin, Texas almost

exclusively exist in east Austin, where the City of Austin is

encouraging growth and development to occur as part of

the Suburban DDZ.

The Imagine Austin plan explicitly addresses the

adoption of green infrastructure as a targeted future

land use with urban agriculture as a component of the

green infrastructure network. Addressing this area of

land use planning will require tools that can locate suit-

able areas within the city where urban agriculture can

best act as green infrastructure. In this study, a process

was developed to create a spatially explicit method of

siting urban agriculture as a green infrastructure tool on

hydrologically sensitive areas (HSAs), or areas prone to

runoff, in east Austin. The method uses geospatial tech-

nology to spatially analyze open access datasets that

include land use, a digital elevation model, and prime

farmland soils. Through this method a spatial relation-

ship can be made between areas of high surface runoff

and where the priority placement of urban farms should

be sited as a useful component of green infrastructure.

Planners or geospatial analysts could use such informa-

tion, along with other significant factors and community

input, to aid decision makers in the placement of urban

agriculture. Creating a spatially explicit approach for

siting potential urban farms will support the integration

of urban agriculture as part of the sustainable land use

planning of Austin.

2. Data and Methods

2.1. Study Area

This study examines the sub-watersheds of Decker

Creek and Elm Creek in east Austin (Figure 1). These

two watersheds were chosen because they are smaller

watersheds that both exist completely within the sub-

urban DDZ and reside in either the Full Purpose Ju-

risdiction or ETJ of Austin, whereas other eastern wa-

tersheds do not fit within those parameters. In ad-

dition, the SH130 growth corridor cuts through both

watersheds as well as the FM 969 growth corridor.

The impact of growth in the region and the expected

loss of prime farmland and agricultural land use make

these two watersheds ideal areas to study. Decker

Creek watershed is 17 square miles and Decker Creek

runs 12 miles from the top of the watershed to the out-

let. Decker Creek watershed is less developed than

other watersheds in Austin containing more agricultural

land, but is projected to experience a rise in population

growth from 3,156 people in 2000 to 12,341 people in

2030 or a 391% projected increase [

]. Elm Creek wa-

tershed totals only 9 square miles and Elm Creek runs

10 miles from the headwaters to the outlet [

]. Elm

Creek watershed is comprised mostly of agricultural

land, though with development looming there will be an

expected increase in population of 180%, or from 3,136

residents in 2000 to 5,643 residents in 2030 [

]. By

analyzing these watersheds, the study examines where

current farmland can best be preserved for urban agri-

culture, especially in light of the area’s continued de-

velopment, and provides insight into the functionality

of urban agriculture as a way to reduce surface runoff

pressure on developing watersheds.

2.2. Data Sourcing & Methods

The entirety of the data for the geospatial elements of

this project were sourced from open access datasets

(Table 1). Data was pulled from the “GIS Downloads”

website managed by the City of Austin. Other data was

pulled from federally managed “Data Download” map

viewers. The digital elevation model (DEM) was down-

loaded from the National Map Viewer managed by the

United States Geologic Survey. Soil data was obtained

from the Web Soil Survey Map Viewer managed by

the National Resources Conservation Service (NRCS).

Aerial imagery was downloaded from the National Agri-

culture Imagery Program (NAIP) and land cover data

came from the National Land Cover Database (NLCD).

Additional data and information about study area context

were derived from informal interviews with urban farm-

ers and City of Austin leadership, together with a site

visit to an urban farm within the study area. The human

subject’s portion of this research was approved by the

Institutional Review Board (IRB) exemption request on

October 15, 2015 with ID # L4567180.

Table 1. Data layers used to inventory urban agriculture and create a topographic index at a watershed scale.

GIS Layer Format Source

Land Use 2012 Vector-polygon City of Austin GIS Downloads (2012)

Watersheds Vector-polygon City of Austin GIS Downloads (2013)

Digital Elevation Model (DEM) 10 meter/raster National Elevation Dataset- Geospatial Data Gateway

Soil Conductivity Vector-polygon Soil Survey Geographic Database: National Resources Conservation Service.

Soil Depth to Restrictive Layer Vector-polygon Soil Survey Geographic Database: National Resources Conservation Service.

Prime Agricultural Soils Vector-Polygon Soil Survey Geographic Database: National Resources Conservation Service.

Aerial Imagery 1 meter/raster Texas Natural Resources Information System: National Agriculture Imagery Program

2.3. Geospatial Database Development

For use in this analysis, all datasets were projected to the

North American Datum 1983 (NAD83), State Plane Texas-

4203 with a Lambert Conformal projection to coincide with

data from the City of Austin. Both the soils and land use

data had to be transformed from vector to raster to cor-

respond with other raster layers. In particular, the prime

farmland data and the land use data needed transformation

to perform a simple urban agricultural inventory that would

be combined with another raster dataset.

Furthermore, a topographic index was performed to de-

lineate HSAs, areas that are prone to generate surface

runoff, within the Decker Creek and Elm Creek watersheds.

The topographic index is determined with a GIS and re-

quires a digital elevation model, soil hydraulic conductivity,

and depth of soil to restrictive layers [36–38].

2.4. HSA Delineation

The topographic index (

) is formed from two components.

The ﬁrst component is a steady state wetness index, formu-

lated from slope (

) and drainage area (

) of the watershed

[

–

]. This index determines the potential for surface

runoff within the contributing area and for each cell within

the raster. The steady state wetness index is deﬁned as:

tanβ

(1)

where

is the drainage area per unit contour length in

meters and β is the slope in radians.

The second component is soil water storage, derived

from soil hydraulic conductivity (

) and soil depth to re-

strictive layers (D) [

–

]. Soil water storage determines

the saturation probability for each cell in the raster. Soil

water storage is expressed as:

ln(K

D) (2)

where is the soil hydraulic conductivity in meters per day

and is the soil depth to restrictive layers in centimeters. In

general, the deeper the soil depth or topsoil, and the higher

the value of the soil hydraulic conductivity or the speed at

which water percolates through the ground, the lower the

likelihood of producing surface runoff [39].

The two components combine to make the soil topo-

graphic index equation, below, to determine the hydrologi-

cally sensitive areas in the watershed:

λ = ln

tanβ

− ln(K

D) (3)

To calculate the steady state wetness index, a 10-m

resolution DEM, clipped to the watershed, was processed

by an open source tool, called the Compound Topographic

Index (CTI), as part of an ArcGIS extension toolset. This

extension allows the user to simply input a DEM into the CTI

tool automatically calculating the steady state wetness in-

dex according to the ﬁrst component of the soil topographic

index equation. The CTI operates exactly like the steady

state wetness index component of the equation above and

can be shown as:

CT I = ln

tanβ

(4)

where

is drainage area “calculated as (ﬂow accumulation

+ 1)

(pixel area in m

)” [

] and

is “the slope expressed

in radians” [40].

At this point, the raster layers for each component part

of the topographic index equation were entered into the

ArcGIS Raster Calculator to calculate the HSAs of the wa-

tersheds. The resulting raster layer indicates areas within

the 10-m grid that are more or less likely to become satu-

rated when a storm event occurs. The higher topographic

index (TI) values correspond to areas most likely to become

saturated and act as a source of surface runoff, also indi-

cating the location of hydrologically sensitive areas [

During a storm event, runoff would likely accumulate and

disperse in areas with higher topographic indices than any

areas with lower topographic indices [39].

Hydrologically sensitive areas are areas most prone to

surface runoff and must be delineated in a watershed by

some threshold criteria to identify as distinct from the lower

topographic indices that are less likely to produce runoff

[

–

]. Many techniques have been used to derive such

conclusions. Agnew et al. [

] proposes using the average

saturation probability to determine HSAs. Others have used

the delineation tactic of targeting 20% of the watershed with

the highest topographic index values to prevent overland

ﬂow from reaching streams [

]. Community and stake-

holder involvement also represents a method of delineating

HSAs based on funding, feasibility, or local expertise. Qui

[

] selected an HSA threshold level for illustrative purposes

considered as reasonable and Martin-Mikle [

] expanded

upon that criteria by selecting TI values 1.5 standard devia-

tions above the mean as HSAs. Following similar protocol,

HSAs in this study were delineated by selecting topographic

index values 1.5 standard deviations above the mean for

each respective watershed. Decker Creek watershed TI

values have a mean of 11.65, resulting in a threshold TI

value level of 15. Any grids in Decker Creek with a TI value

greater than or equal to 15, therefore, are delineated as

HSAs. The threshold value for selection of HSAs in Elm

Creek is anything above 14, or 1.5 standard deviations

above the mean value of 10.20.

2.5. Delineation of Prime Farmland on Agricultural Land

Use

To determine where current land use is best suited for tran-

sitioning into some form of urban agricultural opportunity, a

simple urban agriculture land inventory was performed in

both watersheds. This land inventory was derived from two

datasets: City of Austin land use data and prime farmland

soil data from the National Resources Conservation Service

SSURGO soil database. Utilizing the ArcGIS 10.2 overlay

analysis tools on the aforementioned datasets resulted in a

new dataset of potential land suitable for urban agriculture.

The inventoried lands suitable for potential urban agriculture

are located on prime farmland soils and have an agricultural

land use classiﬁcation.

2.6.

Combining HSAs with Land Use and Prime Farmland

Data to Prioritize Locations for Urban Agriculture as

Green Infrastructure

To prioritize locations for urban agriculture as a green

infrastructure tool, the urban agriculture land inventory,

composed of land use and prime farmland data, had to

be converted to a raster layer to enable a combination of

the delineated HSAs in each watershed with TI values

1.5 standard deviations above the mean. In the Decker

Creek watershed, TI values greater than or equal to 15

were overlaid onto the potential areas for urban agricul-

ture to find HSAs located on agricultural land use and

prime farmland soils. In the Elm Creek watershed, TI

values greater than or equal to 14 were overlaid onto

the potential areas for urban agriculture to find HSAs

located on agricultural land use and prime farmlands

soils. The resulting datasets showed where in each wa-

tershed agricultural land use and prime farmland soils

should be considered as areas for potential urban agri-

culture as a green infrastructure tool according to the

spatial relationship with HSAs in the watershed. To

further prioritize the sites considered as potential sites

for urban agriculture as a green infrastructure strategy

and for conserving prime agricultural land in Austin, an

additional step derived the top ten areas in each water-

shed holding the most HSAs in acreage according to

Travis County Appraisal District (TCAD) parcels. TCAD

parcels identify bounded property lines in Travis County

and their owners.

2.7. Combining Prioritized Locations of Urban Agriculture

with TCAD Parcels to Identify Sites with the Most HSA

Acreage

TCAD parcels from the City of Austin were downloaded

and clipped to each respective watershed to delineate the

boundaries of the parcels within the watershed. The ap-

plication of the ArcGIS spatial analyst tool zonal statistics

derived how many 10 meter pixels of HSAs on potential

urban agriculture sites were located within the boundaries

of individual TCAD parcels. Parcels with no HSAs were

eliminated from analysis. Calculations utilizing the ﬁeld cal-

culator in ArcGIS were then made on the parcels containing

HSAs to derive the total amount of HSAs acreage within

each parcel. Through sorting, a ranked list of parcels con-

taining the most HSAs acreage was generated for each

watershed. This list shows the most viable parcels so that a

structured attempt can be made to identify those landown-

ers that hold the most HSAs in an effort to conserve areas

of prime agricultural land from incoming development and

locate possible urban agriculture opportunities as a method

of green infrastructure.

3. Results

3.1. Delineated HSAs on Agriculture Land Use and Prime

Farmland

The derived TI values for Elm Creek watershed range

from 3.5 to 23.3, while the TI values for Decker Creek

watershed range from 3.3 to 27.0. In Elm Creek water-

shed, areas with TI values greater than or equal to 14 (1.5

standard deviations above the mean) are delineated as

HSAs (Figure 3). In Decker Creek watershed, areas with

TI values greater than or equal to 15 (1.5 standard devia-

tion above the mean) are delineated as HSAs (Figure 4).

HSAs indicate areas prone to surface runoff as a function

of a topography’s slope and drainage area and a soil’s sat-

uration potential derived by soil hydraulic conductivity and

soil depth. To prioritize where potential urban agriculture

sites could serve as a green infrastructure strategy in both

watersheds, HSAs were combined with areas deemed as

agricultural land use by the City of Austin and those con-

taining prime farmland soils. The total area of agricultural

land use on prime farmland for Elm Creek watershed is

1218 acres and 1504 acres for Decker Creek watershed

(Table 2). The areas of the delineated HSAs within poten-

tial sites for urban agriculture as a green infrastructure

tool are 62.7 and 111.7 acres for Elm Creek and Decker

Creek, respectively. HSAs represent 5.1% and 7.4% of

the total agriculture land use on prime farmland in Elm

Creek and Decker Creek, respectively.

Figure 3.

The spatial distribution of hydrologically sensitive areas (HSAs) in

Elm Creek Watershed in Austin, TX. HSAs in Elm Creek watershed have

topographic index (TI) values 1.5 standard deviations above the mean or

greater than 14. These particular HSAs are on potential urban agriculture

sites as determined by their location on land deemed agricultural by the

City of Austin and considered prime farmland by the National Resources

Conservation Service (NRCS).

Figure 4.

The spatial distribution of hydro-

logically sensitive areas (HSAs) in Decker

Creek Watershed in Austin, TX. HSAs

in Decker Creek watershed have topo-

graphic index (TI) values 1.5 standard de-

viations above the mean or greater than

15. These particular HSAs are on poten-

tial urban agriculture sites as determined

by their location on land deemed agri-

cultural by the City of Austin and consid-

ered prime farmland by the National Re-

sources Conservation Service (NRCS).

Table 2.

Potential urban agriculture land totals in Elm Creek and in Decker Creek watersheds: agricultural land use on

prime farmland, HSAs on prime farmland with agricultural land use, and HSAs on prime farmland with agricultural land

use without Critical Water Quality Zone (CWQZ) restrictions.

Elm Creek Watershed Decker Creek Watershed

Total Acreage of Agricultural Land Use with Prime Farmland 1504 acres 1218 acres

Total Acreage of HSAs on Agricultural Land Use with Prime Farmland 62.7 acres 111.7 acres

Total Acreage of HSAs in CWQZ 20.4 acres 17.5 acres

Total HSAs left after CWQZ restrictions omitted 42.3 acres 94.2 acres

3.2. Delineated HSAs Protected Under Current Land Use

Control

The Critical Water Quality Zone (CWQZ) restrictions

set forth by the City of Austin’s 2013 watershed protec-

tion ordinance restricts development around waterways

by establishing a buffer associated with the size of the

waterway [

]. The suburban DDZ watersheds, includ-

ing Decker Creek and Elm Creek watersheds, maintain

unique buffer classifications associated with the size of

the waterways. In minor waterways, CWQZ boundaries

are located 100 feet from the centerline of a waterway

[

]. Intermediate waterways retain CWQZ boundaries

200 feet from the centerline of a waterway [

]. Finally,

major waterways have a restriction of no development

within 300 feet of the CWQZ [41].

The areas of the delineated HSAs found on prime

farmland with agricultural land use that are already pro-

tected under CWQZ restrictions total 20.4 acres for the

Elm Creek watershed (Table 2) and 17.5 acres for Decker

Creek watershed (Table 2). With the Elm Creek water-

shed that leaves 42.3 acres of HSAs on prime farmland

with agriculture land use unprotected from potential devel-

opment (Table 2). While, in the Decker Creek watershed

94.2 acres of HSAs remain unprotected from future de-

velopment (Table 2).

3.3. Prioritized HSAs on TCAD Parcels

In an effort to include tangible results for the City of Austin

(as suggested by the Food Policy Manager at the Ofﬁce of

Sustainability) so that they may reach out to landowners

about conserving portions of their land to incoming devel-

opmental threat, TCAD parcels were sorted into a “top ten

list” of areas that contain the most HSAs. These areas

not only hold the potential to serve as urban agricultural

opportunities, protecting prime farmland, but also to act as

a green infrastructure land use control that further protects

the watershed beyond the CWZQ zone.

In both watersheds, the spatial distribution of the HSAs

in the parcels varies because the parcels are based on an

individual’s ownership. As a result, some of the parcels

show an even spread of HSAs throughout the parcel, while

other locations show clusters of HSAs only in portions of

the parcel. As a means of validation, aerial imagery from

the National Agricultural Imagery Program (NAIP) and land

use data from the National Land Cover Database (NLCD)

were used to identify land use within the parcels to validate

potential urban agriculture sites.

The top ten parcels in Elm Creek watershed contain

51.5 acres, or 82%, of all HSAs. The total acreage for

HSAs ranges from 10.7 acres in Parcel 1 down to 1.9

acres in Parcel 10 (Table 3). In Parcel 1, a mid-size

parcel containing 57 acres, the HSAs spread throughout

the parcel in an even density, while in most others the

HSAs cluster in certain areas of the parcel. Parcel 1

encompasses 41 acres of prime farmland and 26% of

that land contains HSAs. With the relative density of

HSAs in Parcel 1, this parcel may best be conserved as

an entire plot suitable for a large urban farm. Verification

with NAIP imagery and land cover data from the NLCD,

demonstrates that this plot contains a large pasture and

hay field with HSAs (Figure 5). In contrast, Parcels 2

and 3 contain 8.9 and 7.3 acres of HSAs, respectively,

but include only a minimal percentage of HSAs relative

to their total percentage of HSAs on agricultural land

use with prime farmland. For example, Parcel 2 has 132

acres of agricultural land use on prime farmland but only

6.8% contain HSAs. The HSAs are highly clustered in

the southeast portion of the parcel and, through NAIP

and NLCD verification, the densest network of HSAs

lies on cultivated fields (Figure 5). In this case, it would

make sense to target the portion of the land with HSAs

for conservation, adding protection to the watershed by

allowing it to remain free of development, enabling it to

become a potential site for urban agriculture.

In contrast to Elm Creek, the top ten parcels in Decker

Creek watershed encompass a total of 57.5 acres of HSAs,

or 51%, of all delineated HSAs in the watershed. Delineated

HSAs in the Decker Creek watershed parcels range from a

total of 11.1 to 2.9 acres (Table 4). The spatial distribution of

HSAs in the Decker Creek watershed parcels vary, though

in the northwest corner of the watershed ﬁve parcels fea-

ture somewhat evenly spread out HSAs. Through validation

with NAIP imagery and NLCD data, Parcels 6 and 7, in

the northwest corner, contain HSAs primarily on cultivated

ﬁelds, and have 4.6 and 4.0 acres of HSAs, respectively

(Figure 6). Parcel 6 contains only prime farmland and 93%

of the land in parcel 7 is prime farmland (Table 4). The per-

cent of HSAs on prime farmland is 14.5% for parcel 6 and

18.9% for parcel 7. These two parcels, taken together as

an urban agriculture site, could protect 8.6 acres of HSAs

and 52.8 acres of prime farmland. As a contrast, parcel 8

contains 2.9 acres of HSA, but when veriﬁed through NAIP

imagery and NLCD data, most of the HSAs are located

on uncultivated ﬁelds within deciduous forest and scrub-

land (Figure 6). While maintaining this area as open land,

protected from development, still has beneﬁts for the water-

shed, its use as a potential site for urban agriculture as a

green infrastructure strategy may not be the best use of the

land. Rather, leaving the land undeveloped arguably poses

the greatest beneﬁt to the watershed.

Figure 5.

TCAD parcels in Elm Creek watershed. TCAD

parcel 1 shows an even spread of HSAs across the land

area, while TCAD parcel 2 shows a clustering of HSAs in

the southeast portion of the parcel.

Table 3.

Potential urban agriculture sites in Elm Creek

watershed relative to their affiliation with TCAD parcels

and the HSAs located on agricultural land use with prime

farmland (Prime Ag).

Elm Creek Watershed

TCAD

Parcel

HSAs

acreage

Total

Land

Acreage

Prime

Acres

Percent of

HSAs on

Total Land

Acreage

Percent of

HSAs on

Prime Ag

Acres

1 10.7 57.4 41.3 18.60 26.00

2 8.9 147.3 132.3 6.10 6.80

3 7.3 275.4 256 2.70 2.90

4 5.4 165.1 154.3 3.30 3.50

5 4.4 82.2 60.5 5.40 7.30

6 4 29.5 19.2 13.60 20.80

7 3.5 79 75 4.40 4.70%

8 2.8 27.2 9.2 10.10 29.90

9 2.6 42.4 33.2 6.10 7.80

10 1.9 20.6 20.6 9.40 9.40

Figure 6.

TCAD parcels in Decker Creek watershed. TCAD

parcel 6 and 7 show an even spread of HSAs across a cul-

tivated land area, while TCAD parcel 8 shows an uneven

spread of HSAs mostly in a landscape of deciduous forest

and scrubland.

Table 4.

Potential urban agriculture sites in Decker Creek

watershed relative to their affiliation with TCAD parcels

and the HSAs located on agricultural land use with prime

farmland (Prime Ag).

Decker Creek Watershed

TCAD

Parcel

HSAs

acreage

Total

Land

Acreage

Prime

Acres

Percent of

HSAs on

Total Land

Acreage

Percent of

HSAs on

Prime Ag

Acres

1 11.1 89.1 66.8 12.40 16.60

2 8.5 66.4 48.1 12.70 17.60

3 7.8 72 46.9 10.80 16.60

4 7.1 100.3 68.9 7.10 10.30

5 5.7 38.1 31.7 15.10 18.10

6 4.6 31.7 31.7 14.50 14.50

7 4 22.6 21.1 17.70 18.90

8 2.9 66.3 32.3 4.40 9.10

9 2.9 101.3 89 2.90 3.20

10 2.9 31.5 13 9.20 22.20

3.4. Delineated HSAs on Urban Farms

Green Gate Farms is a ﬁve-acre farm and the only classi-

ﬁed urban farm in the Elm Creek watershed. The owners

currently rent the land to farm from an owner who holds

106 acres of the land in the Elm Creek watershed. Analysis

of the HSA data in the Elm Creek watershed results in no

delineated HSAs within the boundaries of the farm. When

HSAs are classiﬁed as existing on prime farmland with a TI

value greater than or equal to 14, there are still no HSAs in

the farm boundary, but there are 7.7 acres of HSAs within

the wider 106 acre boundary under scrutiny. The densest

network of HSAs resides within just 150 feet of the farm

boundaries in an open ﬁeld adjacent to the farm. When the

CWQZ is accounted for, 2.5 acres of HSAs are protected

from development leaving 2.2 acres of HSAs unprotected

still within a dense network neighboring the farm.

Tecolote Organic Farm is the only urban farm within the

Decker Creek watershed. It comprises 65 acres in total and

contains 0.5 acres of HSAs on prime farmland. There are

only two 10m pixels within the CWQZ accounting for protec-

tion of four one-hundredths of an acre. In effect, Tecolote

Organic Farm is protecting 0.46 acres of HSAs on prime

farmland or 92% of the total HSAs. When HSAs are delin-

eated solely by a TI value greater than or equal to 15 and

not according to land use classiﬁcation or prime farmland,

HSAs total 2.1 acres on the farm.

4. Discussion

The Decker Creek and Elm Creek watersheds reside on

the periphery of Austin and remain relatively undeveloped

at this point. However, city projections suggest the water-

sheds will be rapidly urbanizing in the near future as Austin

continues to increase in population and expand its urban

impervious cover. The implication that much of the land in

the suburban DDZ will be developed at some point poses a

signiﬁcant threat to those watersheds and open land within

them including areas that hold value as prime farmland.

Given population growth demands, it will be important to

grow smartly, identifying areas best suited for development

and organized around interconnected open spaces that pro-

tect the environment and add a social and economic beneﬁt

as integrated elements of a community, neighborhood, and

city. The IACP provides the framework for this type of smart

growth in Austin but will need to be implemented through

new sets of land use controls and tools that allow planners

and decision makers to make informed decisions about

where to develop and where not to develop. As the city

redevelops its land development code and initiates Phase

Two of the WPO concerning green infrastructure, tools such

as the one outlined in this research can provide valuable

information to incorporate urban agriculture as part of the

green infrastructure strategy in Austin.

4.1. Implementation of Urban Farms on HSAs: Best

Management Practices (BMPs)

The approach outlined in this research prioritized HSAs

by their spatial connection to potential urban agriculture

sites that could act as a green infrastructure tool. HSAs

in a watershed generate overland runoff and require land

uses that limit the amount of water resource degradation.

High-intensity urban land uses placed on HSAs such as a

commercial development or low-density residential develop-

ment have a signiﬁcant impact on watershed degradation.

For example, an increase in impervious cover increases the

amount of stormwater runoff already generated in the area

resulting in potential ﬂoods, incised creek beds, or more

non-point source pollutant loads from residential lawns,

leaking septic systems, or carbon particulates from streets

and parking lots [

]. Furthermore, common stormwater

infrastructure such as catch basins, detention basins, pipes

and culverts quickly disperses polluted stormwater runoff

to downstream waterways. Urban farms can decrease the

amount of surface runoff that would otherwise ﬂow to an

already overwhelmed storm drain by remaining as unde-

veloped land and increasing the amount of water that the

land can soak up. HSAs on potential urban agricultural land

protect the watershed by keeping it as low-intensity land

use, with the proper BMPs.

Traditional farming techniques typically have negative

impacts on water quality due to farming practices such as

fertilization along with pesticides and herbicide use. As a

result, HSAs in agricultural land uses have a higher poten-

tial to export pollutants from agriculture ﬁelds to streams.

If HSAs were to be located on potential sites for urban

agriculture, it would be necessary to implement BMPs that

provide risk reduction techniques to reduce contamination

of stormwater runoff from urban farms. Below are some of

the best practices for urban agriculture [38,42,43]:

•

Use organic farming principals that require no syn-

thetic pesticides or fertilizers;

•

Construct berms along the edges where the im-

pervious surface and farm meet to prevent erosion

and runoff;

•

Incorporate bio swales and retention ponds to collect

runoff and promote inﬁltration;

•

Use crop rotation and plant cover crops to hold soil

in place;

•

Avoid input of animal manure, instead use organically

produced compost as a fertilizer;

•

Install a rainwater re-use system that captures rain-

water then filters it into an underground tank for

irrigation use.

Real Food Farm is an example of a farm that integrates

stormwater management practices into urban farming. The

farm is located in Baltimore, Maryland, in the Chesapeake

Bay watershed, a watershed affected by the urbanization of

the area leading to large amounts of polluted stormwater

runoff. The farm installed a rainwater re-use system that

captures rainwater off the hoop houses and stores it an

underground cistern. Additionally, the farm incorporated a

retention pond and constructed bio swales and berms to

help mitigate stormwater runoff from the farm [43].

Another BMP model that the City of Austin could poten-

tially implement for urban farms is a sediment and erosion

control plan. Seattle’s urban farm code requires a man-

agement plan if an urban farm exceeds 4,000 square feet

[

]. One provision included within that plan states a given

proposed sediment and erosion control program for farm-

ers to follow. To approve a farm, the city considers the

potential impacts and mitigation of how a farm’s proposed

sediment and erosion control measures will affect the im-

pacts of runoff on the surrounding watershed. In addition,

Seattle resides in the King County Conservation District

which provides free soil nutrient testing on up to ﬁve sam-

ples, including compost [

]. This program allows farmers

to use soil amendments wisely in an effort to reduce water

pollution from over-fertilization.

In a sheer size comparison, traditional farms compared

to urban farms also typically have a broader footprint on the

surrounding environment. The larger the farm the higher

the potential to export non-point source pollution into wa-

terways. In Austin, urban farms (again, classiﬁed as farms

over one acre) average only 4.5 acres in total land area,

and, moreover, the actual plots that produce food occupy

fewer acres. Often embedded within the total land area

are woods, pasture, and fallow ground. If these areas re-

main undeveloped they may occupy space where HSAs

are located. Hence, potential urban farms using BMPs

and occupying space around future development in the two

researched watersheds can provide valuable community

beneﬁts in the form of providing access to healthy food,

enriching green space, and protecting prime farmland, as

well as the added environmental beneﬁt of leaving HSAs as

undeveloped land.

4.2. Opportunities at the Local Scale

The situation at Green Gate Farms in Elm Creek watershed

offers an instructive view of the pressures current and po-

tential urban farms confront as development encroaches

into the rural-urban fringe of Austin. In addition, Green Gate

Farms shows the value urban farms bring to a community

and why urban agriculture in the City of Austin should be

considered as a valuable land management tool.

Green Gate Farm is an organic farm occupying ﬁve

acres of leased land on a larger parcel of land that also con-

tains a large RV park and undeveloped land. The land sits

in an area experiencing major growth with the construction

of new subdivisions. An informal interview with one of the

farmers provided the following information. The farmers of

the land have been farming on it for the last 10 years paying

rent to one owner. Recently, the land was sold to a new

owner who hopes to develop an upscale RV park and man-

ufactured homes on the farmstead. Until recently, the two

original farmers lived on the land in an old farmhouse, but a

new condition applied by the current landowner stipulated

that the house can longer be used as a residence, forcing

the farmers to move off the land. The farmhouse now oper-

ates as an ofﬁce for the farm and a non-proﬁt called New

Farm Institute with a mission “to educate, assist and inspire

citizens and a new generation of sustainable farmers, with

a focus on the urban fringe” [

]. Currently, the farmers are

still under lease until summer 2016 and continue to operate

the farm, but whether or not they will be able to continue to

farm the land is uncertain.

Furthermore, the farmer indicated that the mission of

the farm has always been community oriented. As former

health professionals, the farmers’ priorities have been to

provide fresh produce to underserved families living in a

part of east Austin that is recognized by the USDA as a food

desert, an area with limited access to fresh food or exist-

ing one mile or more from a grocery store [

]. As part of

the mission to create a community resource for neighbors

of all incomes, the farm accepts Supplemental Nutrition

Assistance Program (SNAP) beneﬁts (i.e. food stamps)

and Women, Infants, and Children (WIC) vouchers so that

citizens in the community can buy healthy local produce.

In addition, the farm has a robust community supported

agriculture (CSA) program that provides boxes of organic

produce, meat, eggs, and ﬂowers to participating individ-

uals that choose to enroll in a weekly, monthly, or yearly

subscription service. During an on-site observation of the

farm, it became clear that community involvement is some-

thing that is not just talked about, but actually exists. Many

people dropped by on a Saturday afternoon to purchase

certiﬁed organic vegetables, dairy, meat, and eggs from the

farm stand, which is open to the public Tuesday, Friday and

Saturday. At least two families purchased produce with food

beneﬁt cards. In addition, they have an open-door policy

that allows the public to experience a working urban farm

providing educational opportunities as wells as a relaxing

environment. On this particular Saturday, the farm hosted

a children’s birthday party surrounded by volunteers and

workshare members preparing the ﬁelds for Fall planting.

To protect beneﬁcial urban farms like Green Gate Farm

and ensure that future urban agricultural opportunities exist

on prime farmland within the suburban DDZ, the Austin Sus-

tainable Food Policy Board (SFPB) serve as advisors to the

Austin City Council and the Travis County Commissioners

Court to improve the food system in Austin. As part of the

land development code update in Austin (CodeNEXT), the

all-volunteer SFPB working group is working with commu-

nity and board members“to improve upon the existing code

in a way that meets the needs of communities, farmers,

and regulators in the interest of a healthy, safe, secure, and

sustainable food system for all of Austin” [

]. The group’s

work includes providing recommendations on desirable land

use policies in the suburban DDZ [47] (below):

• Prioritize preservation of prime farmland;

• Establish limits on sub-dividing farmland;

•

Utilize community gardens in new housing developments;

•

Allow for conversion of underutilized industrial

sites/strip-malls into urban farms.

The Green Infrastructure working group in Austin is also

currently working on integrating green infrastructure as part

of land development code update to meet the provisions

set forth in the Imagine Austin Comprehensive Plan. The

SFPB working group and the Green Infrastructure working

group both provide paths by which their recommendations

can be integrated into the code revision process and ensure

that urban agriculture is considered a viable asset in the up-

dated land development code. The methods outlined in this

research to delineate HSAs on potential urban agricultural

sites as a green infrastructure strategy could prove useful

to both working groups as they look for ways to conserve

portions of Austin’s green space.

For example, consider the situation at Green Gate Farm

again: while the five acre farmstead does not contain any

prime farmland soils nor HSAs, the undeveloped fields next

to the farm do contain 2.2 acres of HSAs on prime farmland

that are not protected under the CWQZ. If an updated land

development code existed establishing land use controls that

“prioritized preservation of prime farmland” to include HSAs

as a part of the criteria that functions as green infrastructure

tool in the WPO then a portion of this land could be protected

from development and potentially be used as urban agricul-

ture site. Under this type of land management scenario

there could be an opportunity to move Green Gate Farm

onto the adjacent prime farmland that occupies the HSAs

in order to preserve environmental and social functionality

and services offered by the farmstead when the new owner

builds the RV Park. The protection of Green Gate Farm

would serve not only the underserved communities’ needs

with continued healthy fresh produce but also provide critical

watershed protection from the incoming development.

4.3. Limitations of the Research

The ability to remotely prioritize HSAs and potential ur-

ban agriculture sites is one advantage of employing a GIS

based approach as opposed to the costly and time con-

suming practice of visiting all sites. While the soil data

collected provides the necessary components to complete

the soil topographic index, it would be ideal to have on-site

collected data of the soil inﬁltration rates and soil depths as-

sociated with urban agricultural land in east Austin, enabling

a comparative analysis that quantiﬁes exactly how much

urban farms help to reduce stormwater runoff. Furthermore,

ground truthing HSAs on the prioritized sites to verify that

there is a potential for surface runoff would further validate

the GIS based approach of delineating HSAs. Additionally,

2014 aerial imagery from NAIP was used to verify selected

sites in this study where cultivated cropland could be recog-

nized as potential urban agricultural land. It is also possible

to ground truth data using Google Earth, similarly to Taylor

and Lovell [

], who used this method to identify backyard

agriculture in Chicago. Nonetheless, before making any

policy decisions, on-the-ground site-checking of selected

HSAs on potential urban farms should be a necessary com-

ponent of any urban agriculture land inventory [13].

Another possible limitation of this study are the derived

threshold criteria determining the HSAs in both watersheds.

In this study, the threshold criteria for selecting HSAs are

TI values greater than or equal to 1.5 standard deviations

above the mean. This approach follows well-researched

methods from Mickle et al. [

] and Qui [

] who use sim-

ilar derivations. However, more information is needed on

the local characteristics of the watersheds in Austin that

could possibly determine a more suitable threshold value.

This could result in either leaving the value the same, or

increasing or decreasing the delineations of HSAs in the

study area depending on if that value is equal to, greater

than, or less than the value used in this study.

Another limitation of the study is the lack of analysis of

other important factors related to the feasibility of urban agri-

culture. When determining land use for future development,

urban agriculture faces strong competition from housing

and commercial developers; simply creating urban farms

instead of housing or businesses could decrease affordable

housing options or minimize job creating commercial devel-

opments. Like the set of BMPs urban farms use to mitigate

stormwater runoff, BMPs should also be created that prop-

erly locate urban agriculture without hindering access to

affordable housing.

5. Conclusion

While urban agriculture is a valuable tool in a city’s toolbox

of methods which can be used to create a more sustainable

city, it is by no means a solution to all the problems a city

endures. This study will add only one piece of knowledge to

an ongoing discussion and debate about urban agriculture’s

role in developing more sustainable cities. Nevertheless, this

research may help to further establish urban agriculture as

part of the discussion about strategies that could lead to a

more sustainable city. The social and economic benefits of

urban agriculture are well established and well known, but

the scientific inquiry into the environmental benefits of urban

agriculture is still lagging behind current needs and popular

enthusiasm. This research aims to add a spatially explicit

method to urban agriculture’s potential as a green infrastruc-

ture strategy that demonstrates the ability of urban farms to

mitigate surface runoff and provide environmental benefits to

a city. By validating urban agriculture as green infrastructure

it will help to integrate urban agriculture into public policy and

urban planning as a land management strategy.

References and Notes

[1]

Summary of the Clean Water Act: EPA Laws and Regulations. U.S.

Environmental Protection Agency. Available from: http://www.epa.

gov/laws-regulations/summary-clean-water-act.

[2]

Combined Sewer Overﬂows (CSOs): National Pollutant Dis-

charge Elimination System (NPDES). U.S. Environmental Protection

Agency. Available from: http://www.epa.gov/npdes/combined-sewer-

overﬂows-csos.

[3]

What is Green Infrastructure? U.S. Environmental Protection Agency.

Available from: http://www.epa.gov/green-infrastructure/what-green-

infrastructure.

[4]

Lovell ST. Multifunctional Urban Agriculture for Sustainable Land Use

Planning in the United States. Sustainability. 2010;2(8):2499–2522.

doi:10.3390/su2082499.

[5]

Mougeot LJA. Growing better cities: urban agriculture for

sustainable development. Choice Reviews Online. 2007;43.

doi:10.5860/choice.44-2695.

[6]

Mendes W, Balmer K, Kaethler T, Rhoads A. Using Land Inven-

tories to Plan for Urban Agriculture: Experiences From Portland

and Vancouver. Journal of the American Planning Association.

2008;74(4):435–449. doi:10.1080/01944360802354923.

[7]

Vitiello D. Growing Edible Cities. In: Birch E, Wachter S, editors.

Growing Greener Cities. Philadelphia, PA, USA: University of Penn-

sylvania Press. 10.9783/9780812204094.259.

[8]

Brownﬁelds and Urban Agriculture: Interim Guidelines for Safe Gar-

dening Practices. U.S. Environmental Protection Agency. Available

from: http://www.epa.gov/sites/production/ﬁles/2015-09/documents/

bf urban ag.pdf.

[9]

Dunn AD. Siting Green Infrastructure: Legal and Policy Solu-

tions to Alleviate Urban Poverty and Promote Healthy Communities.

Boston, MA, USA: Pace Law Faculty Publications; 2010. Available

from: http://digitalcommons.pace.edu/cgi/viewcontent.cgi?article=

1557&context=lawfaculty.

[10]

Pothukuchi K, Kaufman JL. The Food System. Jour-

nal of the American Planning Association. 2000;66(2):113–124.

doi:10.1080/01944360008976093.

[11]

Malczewski J. GIS-based multicriteria decision analysis: A survey

of the literature. International Journal of Geographical Information

Science. 2006;20(7):703–726. doi:10.1080/13658810600661508.

[12]

Horst M. Review of Suitable Urban Agriculture Land In-

ventories. American Planning Association. Journal of

the American Planning Association. 2011;Available from:

https://planning-org-uploaded-media.s3.amazonaws.com/

legacy resources/resources/ontheradar/food/pdf/horstpaper.pdf.

[13]

Erickson DL, Lovell ST, M

endez VE. Identifying, quanti-

fying and classifying agricultural opportunities for land use

planning. Landscape and Urban Planning. 2013;118:29–39.

doi:10.1016/j.landurbplan.2013.05.004.

[14]

Balmer K, Gill J, Kaplinger H, Miller J, Peterson M, Rhoads A, et al.

The Diggable City: Making Urban Agriculture a Planning Priority. Mas-

ter of Urban and Regional Planning Workshop Projects. 2016;Paper

52. Available from: http://pdxscholar.library.pdx.edu/cgi/viewcontent.

cgi?article=1051&context=usp murp.

[15]

Park JR, Stabler MJ, Mortimer SR, Jones PJ, Ansell DJ, Parker

GPD. The use of a multiple criteria decision analysis to eval-

uate the effectiveness of landscape and habitat enhancement

mechanisms: An example from the South Downs. Journal of

Environmental Planning and Management. 2004;47(5):773–793.

doi:10.1080/0964056042000274470.

[16]

Arnstein SR. A Ladder Of Citizen Participation. Jour-

nal of the American Institute of Planners. 1969;35(4):216–224.

doi:10.1080/01944366908977225.

[17]

McClintock N, Cooper J. Cultivating the Commons—An Assess-

ment of the Potential for Urban Agriculture on Oakland’s Public Land.

Berkeley, CA, USA: Department of Geography, University of Cali-

fornia; 2010. Available from: http://www.web.pdx.edu/

∼

ncm3/ﬁles/

Cultivating the Commons2010.pdf.

[18]

Zhang FF, Cai J, G L. How urban agriculture is reshaping peri-urban

Beijing? Open House International. 2009;34(2):15–24.

[19]

Chin D, Infahsaeng T, Jakus I, Oorthuys V. Urban Farming in

Boston: A Survey of Opportunities. Boston, MA, USA: Tufts University.

Available from: http://as.tufts.edu/uep/sites/all/themes/asbase/assets/

documents/fieldProjectReports/2013/Team 8 Final Report 2013.pdf.

[20]

Ackerman K. The Potential for Urban Agriculture in New York

City: Growing Capacity, Food Security, and Green Infrastructure.

New York, NY, USA: Urban Design Lab, Columbia University; 2012.

Available from: http://urbandesignlab.columbia.edu/ﬁles/2015/04/

4 urban agriculture nyc.pdf.

[21]

Kremer P, DeLiberty TL. Local food practices and growing po-

tential: Mapping the case of Philadelphia. Applied Geography.

2011;31(4):1252–1261. doi:10.1016/j.apgeog.2011.01.007.

[22]

Myeong S, Nowak D, Hopkins P, Brock R. Urban cover mapping us-

ing digital, high-spatial resolution aerial imagery. Urban Ecosystems.

2001;5(4). 10.1023/A:1025687711588. 10.1023/A:1025687711588.

[23]

McClintock N, Cooper J, Khandeshi S. Assessing the potential contri-

bution of vacant land to urban vegetable production and consumption

in Oakland, California. Landscape and Urban Planning. 2013;111:46–

58. doi:10.1016/j.landurbplan.2012.12.009.

[24]

Horst M. Growing Green: An inventory of Public Lands Suit-

able for Community Gardening in Seattle, Washington. Seattle,

WA, USA: University of Washington; 2008. Available from:

http://www.seattle.gov/Documents/Departments/Neighborhoods/

PPatch/Growing-Green.pdf.

[25]

Colasanti K. Assessing the local food supply capacity of Detroit,

Michigan. Journal of Agriculture, Food Systems, and Community

Development. 2010;pp. 41–58. doi:10.5304/jafscd.2010.012.002.

[26]

McClintock N. Assessing soil lead contamination at multiple scales

in Oakland, California: Implications for urban agriculture and en-

vironmental justice. Applied Geography. 2012;35(1-2):460–473.

doi:10.1016/j.apgeog.2012.10.001.

[27]

Imagine Austin Comprehensive Plan. City of Austin, TX, USA. Avail-

able from: https://austintexas.gov/sites/default/ﬁles/ﬁles/Planning/

ImagineAustin/webiacpreduced.pdf.

[28]

Top Ten Demographic Trends in Austin, Texas. City of Austin, TX,

USA. Available from: https://www.austintexas.gov/page/top-ten-

demographic-trends-austin-texas.

[29]

State of the Food System Report. City of Austin, TX, USA. Avail-

able from: http://austintexas.gov/sites/default/ﬁles/ﬁles/Sustainability/

AustinFoodReport050115 - FINAL.pdf.

[30]

Karvonen A. Politics of Urban Runoff. Cambridge, MA, USA: MIT

Press; 2011.

[31]

Watershed Ordinance History: Watershed Protection Ordinance

Stakeholder Process 2011–2012. City of Austin, TX, USA.

Available from: https://web.archive.org/web/20130915093925/http:

//austintexas.gov/page/watershed-protection-ordinance.

[32]

Watershed Ordinance History. City of Austin, TX, USA. Avail-

able from: https://web.archive.org/web/20130915093925/http://

austintexas.gov/page/watershed-protection-ordinance.

[33]

Community Inventory: Natural Environment. City of Austin, TX, USA.

Available from: http://ftp://ftp.ci.austin.tx.us/GIS-Data/planning/compplan/

combined community inventory NaturalEnvironment v7.pdf.

[34]

What is Urban Agriculture? U.S. Environmental Protection

Agency. Available from: http://www3.epa.gov/region1/eco/uep/

urbanagriculture.html.

[35]

Environmental Integrity Index. City of Austin, TX, USA. Avail-

able from: http://www.austintexas.gov/department/environmental-

integrity-index.

[36]

Agnew LJ, Lyon S, G

erard-Marchant P, Collins VB, Lembo AJ, Steen-

huis TS, et al. Identifying hydrologically sensitive areas: Bridging the

gap between science and application. Journal of Environmental Man-

agement. 2006;78(1):63–76. doi:10.1016/j.jenvman.2005.04.021.

[37]

Martin-Mikle CJ, de Beurs KM, Julian JP, Mayer PM. Identify-

ing priority sites for low impact development (LID) in a mixed-

use watershed. Landscape and Urban Planning. 2015;140:29–41.

doi:10.1016/j.landurbplan.2015.04.002.

[38]

Qiu Z, Hall C, Drewes D, Messinger G, Prato T, Hale K, et al. Hydrolog-

ically Sensitive Areas, Land Use Controls, and Protection of Healthy

Watersheds. J Water Resour Plann Manage. 2014;140(7):04014011.

doi:10.1061/(asce)wr.1943-5452.0000376.

[39]

Qiu Z, Hall C, Drewes D, Messinger G, Prato T, Hale K, et al. Hydrolog-

ically Sensitive Areas, Land Use Controls, and Protection of Healthy

Watersheds. J Water Resour Plann Manage. 2014;140(7):04014011.

doi:10.1061/(asce)wr.1943-5452.0000376.

[40]

Evans J, Oakleaf J, Cushman S, Theobald D. Quantitative Meth-

ods in Spatial Ecology: An ArcGIS Toolbox for Surface Gradi-

ent and Geomorphometric Modeling. Version 2.0-0. Available

from: http://evansmurphy.wix.com/evansspatial#!arcgis-gradient-

metrics-toolbox/crro.

[41]

Municode Library: Critical Water Quality Zones Established. City of

Austin, TX, USA. Available from: https://www.municode.com/library/

tx/austin/codes/code of ordinances?nodeId=TIT25LADE CH25-

8EN SUBCHAPTER AWAQU ART2WACLZOES.

[42]

Walter M, Walter M, Brooks E, Steenhuis T, Boll J, Weiler K. Hydro-

logically sensitive areas: Variable source area hydrology implications

for water quality risk assessment. Journal of Soil and Water Conser-

vation. 2000;3:227–284.

[43]

Boian M, Hughes J, Deardorff L. Urban Farms: A Green Infrastruc-

ture Tool for the Chesapeake Bay. American Rivers; 2015. Avail-

able from: http://www.americanrivers.org/assets/pdfs/reports-and-

publications/UrbanAgricultureReport Final.pdf.

[44]

Tip 244: Urban Agriculture. City of Seattle, WA, USA. Available from:

http://www.seattle.gov/DPD/Publications/CAM/cam244.pdf.

[45]

Green Gate Farms. New Farm Institute. Available from: http:

//greengatefarms.net/new-farm-institute.

[46]

Food Access Research Atlas, U.S. Department of Agriculture.

Available from: http://www.ers.usda.gov/data-products/food-access-

research-atlas/go-to-the-atlas.aspx.

[47]

Recommendation To Code Next From The Sustainable

Food Policy Board. City of Austin, TX, USA. Available

from: https://www.austintexas.gov/sites/default/ﬁles/ﬁles/

Planning and Zoning/CodeNEXT/Sustainable Food Policy Board

RecommendationToCodeNext.pdf.

[48]

Taylor JR, Lovell ST. Mapping public and private spaces of ur-

ban agriculture in Chicago through the analysis of high-resolution

aerial images in Google Earth. Landscape and Urban Planning.

2012;108(1):57–70. doi:10.1016/j.landurbplan.2012.08.001.