Urban agriculture in Baltimore, Maryland: Documenting current irrigation practices and rainwater harvesting

Abstract

Rainwater harvesting could address freshwater shortages impacting urban agriculture caused by climate change. Limited studies exist on perceptions and concerns about using harvested rainwater for crop irrigation, and to our knowledge none have focused on urban agriculture. We surveyed urban farms and community gardens in Baltimore, Maryland, to document current irrigation practices and water sources and determine interest in rainwater harvesting. We evaluated differences in farm/garden characteristics and water uses between those harvesting rainwater, interested in rainwater harvesting, and not interested in rainwater harvesting. Most Baltimore farms/gardens surveyed were concerned about water availability (59%, n = 20), but all had access to city‐supplied water and only 14% (n = 5) reported water access preventing site expansion. Only 18% of sites used drip irrigation (n = 11), 89% (n = 33) were not using any water filtration, and most (72%, n = 23) had never attended a produce safety training. Twenty‐seven percent (n = 10) of sites were harvesting rainwater, 54% (n = 20) were interested in harvesting rainwater, and 19% (n = 7) were not interested. Ninety percent (n = 9) of the sites currently harvesting rainwater used covered rain barrels, an important strategy for preventing contamination. There were no significant associations between farm/gardens’ rainwater harvesting interests and farming practices, water access and use, or demographics of site respondents. Those currently harvesting rainwater and those interested had numerous concerns about rainwater harvesting including installation, design, and water quality, and were interested in additional information. Offering rainwater harvesting, food safety and education programs to urban farmers and gardeners could increase the safe use of this important water source.

Abbreviations

• CDC
• Centers for Disease Control and Prevention
• FSMA
• Food Safety Modernization Act
• GAP
• Good Agriculture Practices

INTRODUCTION

Climate change and population growth are increasingly impacting water availability worldwide. Weather events are becoming more extreme and unpredictable, including more frequent and severe flooding, droughts, and heat waves. These changes threaten water supplies due to scarcer, more unpredictable, and more polluted water sources (United Nations, 2015). Agriculture is the most water-intensive industry in the world, using roughly two-thirds of water supplies from rivers, lakes, and aquifers. The three most common sources of irrigation water in agriculture across the globe are municipal water, groundwater, and surface water (Mahmoud, 2018). According to a report by the United Nations Food and Agriculture Organization, 20% of global groundwater is currently overexploited and there will be a 40% water supply shortfall by 2030 unless there is a notable change in water supply management (FAO, 2017). These water stresses could impact agricultural areas, as Dong et al. (2019) documented significantly higher decreases in groundwater levels in counties in the state of Maryland with more cropland. Increased groundwater withdrawal can also lead to land subsidence (Tzampoglou et al., 2023). At the same time, 60% more food will be needed to feed the global population, and the demand for water is projected to increase by 55% by 2050 (United Nations, 2015).

Urban agriculture, defined by the U.S. Department of Agriculture (USDA) as including “the cultivation, processing, and distribution of agricultural products in urban and suburban area,” is growing quickly across the United States to address growing food insecurity, provide social and environmental benefits, and increase resilience and self-reliance, yet it is limited by irrigation water costs and the availability of high-quality water (USDA, n.d.). Most urban farmers rely on municipal water for irrigation, which is dependent on proximity to existing water mains and can be expensive, unreliable, or dependent on a city's capacity for maintenance (N. Little, personal communication, 2020). Irrigation water and labor are the main costs for many urban farms, requiring grants and donations to maintain economic viability (Arnold & Rogé, 2018; Siegner et al., 2018). A study by Arnold and Rogé (2018) found that the ability to pay for irrigation was the most important factor for urban farms’ longevity. Their study underscored the benefits of less expensive, non-potable water for urban agriculture while acknowledging infrastructure and delivery challenges. The use of potable water for agricultural irrigation means less potable water available for drinking and household use, and this highly treated water is not necessary to maintain water quality to meet food safety requirements (Uyttendaele et al., 2015).

As climate change continues to impact water availability worldwide, some farmers are exploring alternative water sources for irrigation, including harvested rainwater (Mahmoud, 2018; Rock et al., 2019). Rainwater harvesting collects and stores rainwater from rooftops or other surfaces for later use (Helmreich & Horn, 2008). Rainwater harvesting systems vary widely and can include covered or uncovered rain barrels, cisterns, or other containers to collect runoff from rooftops or other hard surfaces. Harvested rainwater has many potential environmental benefits including water conservation and reducing stormwater runoff (Hamilton et al., 2018). For urban farmers specifically, rainwater harvesting could address water access and availability challenges. Rainwater harvesting systems could be especially beneficial for small-scale farmers in the United States Mid-Atlantic region during summer months when heavy rainstorms can quickly fill rainwater harvesting systems, and water can be stored for use during drier periods (Bakacs et al., 2017). Despite the possible benefits of rainwater harvesting in urban agriculture and growing research on harvested rainwater quality for produce irrigation (Deng, 2021; Haberland et al., 2013), there is a lack of information on current trends in rainwater harvesting among US urban agricultural sites, including urban farms and community gardens, as well as their needs and barriers related to the use of this sustainable water source. To our knowledge, there have been no studies specifically examining urban farms/gardens’ opinions on using harvested rainwater for irrigation. To address this knowledge gap, we conducted a needs assessment survey among Baltimore, Maryland, urban farms and community gardens about their current irrigation water sources and interest, needs, and concerns related to rainwater harvesting.

Core Ideas

• There are limited studies on urban farmers’ perceptions of irrigation with harvested rainwater.
• Most Baltimore farms/gardens are concerned about water availability.
• Some (27%;10/37) Baltimore farms/gardens are harvesting rainwater, 54% (20/37) are interested.
• There is a desire for rainwater harvesting installation, design, and water quality information.

MATERIALS AND METHODS

Survey instrument

An interdisciplinary team of public health and Extension researchers developed the survey (Supporting Information 1) with input from stakeholders including farmers, local extension agents, and extension specialists. This study was approved by the University of Maryland College Park, Institutional Review Board, IRB Protocol 1601501.

Participants, who self-identified on the survey as the owner, manager, volunteer, volunteer coordinator, or “other” at the farm or garden site, were asked to answer 36 multiple-choice and open-ended questions about their site, such as ownership of the land and the size of the farm. Open-ended responses were not analyzed as part of this portion of the project. Participants identified the name of their farm or garden on the survey. The unit of analysis for all survey questions was the site. Participants were then asked about the water access and use at their site, such as concerns about water availability and the cost of water. The survey also included questions to determine if sites were currently using or were interested in installing a rainwater harvesting system. Sites using a rainwater harvesting system at the time of the survey were asked additional questions about their rainwater harvesting practices, such as size and reason for installing their system. All participants were then asked about their concerns about using a rainwater harvesting system, such as water safety concerns. Additional questions included education/outreach practices, such as attendance at a produce safety workshop, and demographics.

Study population, recruitment, and data collection

Baltimore, MD is a city of 593,490 residents as of 2019. There were approximately 33 urban farms and over 200 community gardens in the city as of 2023 (N. Little, personal communication, 2024). We defined urban farms and community gardens based on definitions used by the Farm Alliance of Baltimore: Urban farms are agricultural operations that produce more than $2000 worth of farm products for sale or donation annually and have been in production for at least one year in Baltimore City; community gardens produce less than $2000 worth of farm products for sale or donation annually and have been in production for at least one year in Baltimore City (Farm Alliance of Baltimore, n.d.). A previous survey of urban agriculture in Baltimore found that 41% of farmers were White, 37% Black, and 15% two or more races, which differs from the general population in Baltimore City based on the 2020 census where 28% of residents were White and 58% Black (U.S. Census Bureau, n.d.).

The research team distributed an online survey from July to September 2020 to participants by emailing a survey link to University of Maryland Extension listservs of Baltimore urban farmers and gardeners. The survey was also shared with community partners who distributed the survey to their membership and shared through social media. Participation was restricted to those 18 years of age and older. The introductory survey language stipulated that by completing this survey, the participant indicated that they were at least 18 years of age, understood the purpose of the survey, and freely and voluntarily chose to participate. Additionally, the consent form explained that the purpose of the study was to use feedback from urban gardeners and farmers in the Baltimore area. The online survey was developed and stored on the online survey platform Qualtrics (Qualtrics).

Statistical analysis

Prior to analysis, the data were cleaned to remove responses that did not provide information about rainwater harvesting at the farm or garden site. Statistical analyses and descriptive statistics were analyzed using R Studio (The R Foundation) and STATA (StataCorp LLC, College Station). The data were analyzed based on sites’ rainwater harvesting interest (using a rainwater harvesting system, interested in installing a rainwater harvesting system, or not interested in installing a rainwater harvesting system). Descriptive statistics were reported, such as the percentage of farms/gardens concerned about water availability on their site. Chi-square tests were used to evaluate associations between sites’ rainwater harvesting interest, other site characteristics, and respondent demographics, unless there were less than five responses in a category, upon which a Fisher's exact test was used. p-values of ≤0.05 were defined as statistically significant.

RESULTS

Fifty-five participants completed at least some portion of the survey. Duplicate and incomplete responses were removed, leaving 38 participants that were included in the final analysis—25 of whom responded on behalf of a community garden, 10 responded on behalf of a farm, and 3 did not provide site information.

Baltimore urban farmer/gardener demographics and site characteristics

Survey respondents were predominantly White, non-Hispanic individuals, with equal numbers of male and female respondents (Table 1). Although these demographics differ from the city overall, they do align with previous reviews of Baltimore urban farmers (Little et al., 2019). Age and level of education varied among the respondents, with most individuals between the ages of 30 and 69. All respondents had at least some college or an associate degree. Comparisons based on rainwater harvesting interest are described in Section 3.3.

TABLE 1 Baltimore urban farmer/gardener demographics compared by rainwater harvesting interest using chi-square or Fisher's exact tests.

	n (%)
	Harvesting rainwater	Interested in harvesting rainwater	Not interested in harvesting rainwater	Total	p-value
Role					0.2
Owner	3 (33)	2 (12)	1 (17)	6 (19)
Manager	4 (44)	6 (35)	0 (0)	10 (31)
Volunteer	0 (0)	1 (6)	1 (17)	2 (6)
Volunteer coordinator	0 (0)	3 (18)	1 (17)	4 (13)
Other	2 (22)	5 (29)	3 (50)	10 (31)
Total	9	17	6	32
Ethnicity					0.24
Non-Hispanic or Latino	9 (90)	16 (100)	5 (83)	30 (94)
Prefer not to answer	1 (10)	0 (0)	1 (17)	2 (6)
Total	10	16	6	32
Race					0.67
Asian	1 (11)	1 (6)	0 (0)	2 (6)
Black/African American	1 (11)	5 (29)	1 (17)	7 (22)
White	6 (67)	11 (65)	4 (67)	21 (66)
Prefer not to answer	0 (0)	0 (0)	1 (17)	1 (3)
Multiple races	1 (11)	0 (0)	0 (0)	1 (3)
Total	9	17	6	32
Age					0.31
18–29	0 (0)	1 (6)	0 (0)	1 (3)
30–49	3 (33)	9 (53)	2 (33)	14 (44)
50–69	4 (44)	5 (29)	2 (33)	11 (34)
70–89	2 (22)	2 (12)	0 (0)	4 (13)
Prefer not to answer	0 (0)	0 (0)	2 (33)	2 (6)
Total	9	17	6	32
Sex					0.09
Male	7 (70)	7 (41)	1 (17)	15 (45)
Female	3 (30)	10 (59)	4 (67)	17 (52)
Prefer not to say	0 (0)	0 (0)	1 (17)	1 (3)
Total	10	17	6	33
Education					0.88
Some college or associate degree	1 (10)	2 (12)	1 (17)	4 (12)
Bachelor's degree	5 (50)	10 (59)	3 (50)	18 (55)
Graduate or professional degree	4 (40)	4 (24)	1 (17)	9 (27)
I prefer not to answer	0 (0)	1 (6)	1 (17)	2 (6)
Total	10	17	6	33

The most common farm or garden size reported was between 1000 and 10,000 ft² (93 and 939 m²) (32%, n = 12) (Table 2). Farm/garden sites were evenly distributed between private ownership by the farmer/gardener (20%, n = 11), use through the Baltimore Adopt-A-Lot program (a program through the Baltimore City Department of Housing and Community Development that allows residents, businesses, or neighborhood groups to use city-owned vacant lots) (27%, n = 10), and “other” (24%, n = 9). Other types of land ownership reported included using city-owned land and land trusts. Most sites used raised beds (34%, n = 25) or in ground (36%, n = 27) growing methods, followed by hoop houses or high tunnels (14%, n = 10). The sites mostly produced food crops for private consumption (43%, n = 28) or educational purposes (26%, n = 17).

TABLE 2 Baltimore urban farm/garden site characteristics, farming practices, and water access and use compared by rainwater harvesting interest using chi-square or Fisher's exact tests.

	n (%)
	Harvesting rainwater	Interested in harvesting rainwater	Not interested in harvesting rainwater	Total	p-value
Total production area of the farm					0.76
<500 ft2 (46 m2)	3 (30)	1 (5)	2 (29)	6 (16)
500–1000 ft2 (46–93 m2)	1 (10)	3 (15)	0 (0)	4 (11)
1000–10,000 ft2 (93–939 m2)	3 (30)	6 (30)	3 (43)	12 (32)
10,000 ft2–1 acre (939–4047 m2)	2 (20)	3 (15)	1 (14)	6 (16)
1–2 acres (4047–8094 m2)	1 (10)	4 (20)	1 (14)	6 (16)
2–10 acres (8094–40,469 m2)	0 (0)	3 (15)	0 (0)	3 (8)
Total	10	20	7	37
Ownership of the land					0.55
Yes	6 (60)	4 (20)	1 (14)	11 (20)
No, the land is used through an Adopt-A-Lot program	2 (20)	6 (30)	2 (29)	10 (27)
No, someone else owns the land and has given permission to use it, other than an adopt a lot license	1 (10)	3 (15)	2 (29)	6 (16)
Other	1 (10)	6 (30)	2 (29)	9 (24)
No, I rent the land	0 (0)	1 (5)	0 (0)	1 (3)
Total	10	20	7	37
Growing methoda
Raised beds	6 (25)	13 (34)	6 (50)	25 (34)	0.64
In the ground	10 (42)	13 (34)	4 (33)	27 (36)
Greenhouse	0 (0)	3 (8)	0 (0)	3 (4)
Hoop house/high tunnel	4 (17)	5 (13)	1 (8)	10 (14)
Other	1 (4)	2 (5)	0 (0)	3 (4)
Rooftop	1 (4)	0 (0)	1 (8)	2 (3)
Hydroponics	2 (8)	2 (5)	0 (0)	4 (5)
Total	24	38	12	72
Types of products the farm producesa					0.97
Food crops and cut flowers for sale	1 (6)	5 (14)	0 (0)	6 (9)
Food crops for sale only	0 (0)	2 (6)	0 (0)	2 (3)
Food crops for private consumption	9 (50)	13 (36)	6 (55)	28 (43)
Other	2 (11)	7 (19)	3 (27)	12 (18)
Educational purposes	6 (33)	9 (25)	2 (18)	17 (26)
Total	18	36	11	65
Water availability concern for site selection
Yes	6 (60)	10 (59)	4 (57)	20 (59)	1.00
No	4 (40)	7 (41)	3 (43)	14 (41)
Total	10	17	7	34
Water access preventing site expansion					0.68
Yes	2 (22)	3 (16)	0 (0)	5 (14)
No	7 (78)	16 (84)	7 (100)	30 (86)
Total	9	19	7	35
Cost of water a concern					0.80
Yes	2 (20)	8 (40)	2 (33)	12 (34)
No	7 (80)	12 (60)	4 (67)	23 (66)
Total	9	20	6	35
Access to city/public water					0.47
Yes—through a meter pit	7 (73)	9 (45)	3 (43)	19 (51)
Yes—through another method	2 (18)	11 (55)	3 (43)	16 (43)
Other	1 (9)	0 (0)	1 (14)	2 (5)
Total	10	20	7	37
Irrigation methoda
Drip irrigation	4 (22)	5 (17)	2 (17)	11 (18)	0.73
Hose	8 (44)	14 (47)	7 (58)	29 (48)
Hand	6 (33)	7 (23)	3 (25)	16 (27)
Other	0 (0)	4 (13)	0 (0)	4 (7)
Total	18	30	12	60
Use of water filtration					0.24
Sand filter	0 (0)	2 (10)	0 (0)	2 (5)
No	8 (80)	18 (90)	7 (100)	33 (89)
Other	2 (20)	0 (0)	0 (0)	2 (5)
Total	10	20	7	37
Attended produce safety workshop or training					0.41
No	6 (75)	13 (72)	4 (67)	23 (72)
Good Agricultural Practices (GAPs)	0 (0)	4 (22)	1 (17)	5 (16)
Food Safety Modernization Act (FSMA)	1 (13)	0 (0)	0 (0)	1 (3)
Other, please specify	1 (13)	1 (6)	1 (17)	3 (9)
Total	8	18	6	32
Host educational events					0.49
Yes	2 (25)	6 (38)	3 (60)	11 (38)
No, but would be interested in hosting events	4 (50)	8 (50)	1 (20)	13 (45)
No, and would not be interested in hosting events	2 (25)	2 (13)	1 (20)	5 (17)
Total	8	16	5	29
Info that would be useful about rainwater harvesting system					0.19
Methods to collect rainwater	4 (18)	12 (14)	3 (12)	19 (14)
Materials for rain and rooftop water collection	3 (14)	13 (15)	3 (12)	19 (14)
Techniques and strategies to utilize rain and rooftop water	3 (14)	14 (16)	4 (15)	21 (15)
How to test water for food safety compliance	4 (18)	12 (14)	5 (19)	21 (15)
How to filter water to improve quality	5 (23)	12 (14)	4 (15)	21 (15)
Other	0 (0)	10 (11)	2 (8)	12 (9)
Cost savings of using collected rain or rooftop water	0 (0)	0 (0)	1 (4)	1 (1)
How to clean rain barrels/cisterns	3 (14)	15 (17)	4 (15)	22 (16)
Total	22	88	26	136

Water availability, uses, treatments, and food safety training

Most respondents were concerned about water availability when selecting a garden or farm site (59%, n = 20) and 14% (n = 5) expressed that water access had prevented them from expanding their farm site (Table 2; Figure 1). Nearly two-thirds of respondents (n = 23) reported that the cost of water was not a challenge on their farm. There was no significant correlation between water costs concerns and Adopt-a-Lot agreements (data not shown), despite anecdotal evidence that farmers who own their land express concern about water bill costs (N. Little, personal communication, 2020). All respondents reported having access to city or public water on site. Watering by hose was the most common method of irrigation for all sites (48%, n = 29), followed by watering by hand (27%, n = 16) and drip irrigation (18%, n = 11). Most Baltimore farms and gardens (89%, n = 33) were not using any water filtration. Most respondents (72%, n = 23) had never attended a produce safety course or training (Table 2).

[IMAGE OMITTED. SEE PDF]

Rainwater harvesting practices

Twenty-seven percent (n = 10) of farms/gardens were harvesting rainwater, 54% (n = 20) were not harvesting rainwater but were interested in installing a system, and 19% (n = 7) were not harvesting rainwater and not interested in installing a system (Figure 2). Covered barrels and collection from rooftop runoff were the most widely used methods for harvesting rainwater reported, with 90% (n = 9) of sites that used harvested rainwater utilizing covered barrels (Table 3). Only one respondent provided additional information about harvested rainwater collection that was not from a roof surface, explaining that water was collected in underground cisterns. It is possible that rainwater is being captured as runoff from parking lot surfaces (the Community Learning Garden at the University of Maryland uses this technique for irrigation water); however, we cannot be sure what “other” rainwater harvesting systems might be in use at our respondents’ sites (UMD Community Learning Garden, n.d.). Sites harvesting rainwater were using rainwater harvesting systems that could collect a range of water volumes (<50–250 gallons; 189–946 L), with one site's system able to capture 6000 gallons (22,712 L). Most of the rainwater harvesting systems were 10 years old or less (70%, n = 7), while one respondent reported no longer using the installed rainwater harvesting system at their site. The rainwater harvesting systems were mostly paid for using funds from city grants/programs (36%, n = 4) or private donations (27%, n = 3).

[IMAGE OMITTED. SEE PDF]

TABLE 3 Characteristics of Baltimore urban farm/garden rainwater harvesting systems (n = 10).

	n (%)
Type of rainwater harvesting system
Use covered rain barrels	9 (50)
Use uncovered rain barrels	1 (6)
Use cisterns	1 (6)
Collect runoff from rooftops	6 (33)
Other	1 (6)
Gallons (liters) rainwater harvesting system collects
<50 gallons (<189 L)	4 (44)
51–100 gallons (193–379 L)	3 (33)
101–250 gallons (382–946 L)	2 (22)
Length of time using rainwater harvesting system
1–5 years	3 (30)
5–10 years	4 (40)
Longer than 10 years	2 (20)
No longer use the installed system	1 (10)
Reason for installing rainwater harvesting system
To save money	5 (17)
To benefit the environment	9 (31)
To be more self sufficient	9 (31)
To ensure water availability in case of a city service interruption	3 (10)
Connecting to city water was too difficult/not possible	2 (7)
Other	1 (3)
How rainwater harvesting system was paid for
Funds from your organization	2 (18)
City grant/program	3 (27)
Private donations	3 (27)
Other	3 (27)

The most common reasons for installing rainwater harvesting systems were to benefit the environment (31%, n = 9) and become more self-sufficient (31%, n = 9) (Table 3). However, 67% (n = 6) of respondents reported encountering a significant maintenance or performance issue with the system since installation. These system challenges included requiring pumps to deliver the water, clogged lines, leaks, and loss of institutional knowledge in that “knowing how to operate the system has been lost to changes in personnel.”

Respondents, both those who already had rainwater harvesting systems and those interested in installing them, had concerns regarding the implementation and use of these systems, although 60% (n = 6) of those already harvesting rainwater had no concerns (Figure 3). The most common concerns for sites already harvesting rainwater were that the collected volume of water was not sufficient to meet irrigation needs, lack of hard surface to collect water, and lack of knowledge on how to maintain the system. Sites interested in harvesting rainwater, but not doing so yet, had a wider range of concerns. The primary concerns for this group included uncertainty about how to design (10%) and install (9%) a rainwater harvesting system, uncertainty about city rules and regulations (9%), concerns that systems would attract mosquitoes and other pests (9%), and costs (8%). Respondents indicated that a variety of information about rainwater harvesting practices would be useful for them, including techniques for using rainwater, materials for rain and rooftop water collection, how to test water for food safety compliance, and how to filter water to improve quality.

[IMAGE OMITTED. SEE PDF]

Comparisons between farms/gardens using and not using harvested rainwater

There were no significant differences in terms of demographics between respondents representing farm and garden sites that harvested rainwater, interested in harvesting rainwater, and not interested in harvesting rainwater (Table 1). More respondents from sites harvesting rainwater had a graduate or professional degree (40%, n = 4) compared to those that were interested in installing a rainwater harvesting system (24%, n = 4), and those that were not interested in installing a rainwater harvesting system (17%, n = 1), although these differences were not significant (p = 0.88).

There were also no significant differences in terms of site characteristics or farming practices based on rainwater harvesting use, although some comparisons are of interest for future research (Table 2). More respondents that represented sites harvesting rainwater owned the land on which they farmed (60%, n = 6) (p = 0.27) and had smaller sites (<500 ft²/46 m²) compared to those not harvesting rainwater (p = 0.98) (Table 2). Access to city water did not differ significantly between rainwater harvesting practices, as all sites reported having access to this water (p = 0.47). However, more sites harvesting rainwater (73%, n = 7) had access to city water from a metered pit compared to 45% (n = 9) of those interested in installing a rainwater harvesting system and 43% (n = 3) not interested in installing a rainwater harvesting system (Figure 4). Neither water availability nor cost concerns differed significantly based on rainwater harvesting practice (Figure 1). More sites currently harvesting rainwater reported that access to water prevented them from expanding their farm/garden site than those who were not currently harvesting rainwater, but this result was also not significant (p = 0.68).

[IMAGE OMITTED. SEE PDF]

More sites that harvested rainwater used some type of water filtration compared to those not harvesting rainwater, but this was not significant (p = 0.24) (Figure 4). Twenty percent (n = 2) of sites harvesting rainwater used a water filtration method compared to 10% (n = 2) of those interested in installing a system, and 0% (n = 0) of those not interested in installing a system. A higher percentage of sites harvesting rainwater employed drip irrigation as their irrigation method compared to those not harvesting rainwater, but this was not significant (p = 0.73). There were no significant differences in food safety training between respondents who represented sites harvesting rainwater and those who did not—most respondents in categories had never attended a training (p = 0.41) (Table 2). Respondents from sites harvesting rainwater were most interested in information related to water filtration to improve quality compared to other respondents (Table 2).

DISCUSSION

Water availability and rainwater harvesting

Most farm and garden respondents were concerned about water availability for site selection, but all had access to public/city water. Baltimore is unique in that it offers water access at a flat rate of $120 per growing season to sites that are either city or “privately owned community open space[s]” within the city through their Water Access Program (Baltimore City Department of Housing & Community Development, n.d.). Most respondents also noted that water access did not prevent them from expanding their sites and that water costs were not a concern. According to Bakacs et al. (2017), rainwater harvesting can be beneficial to US Mid-Atlantic farmers that do not have access to municipal water since heavy rainstorms can quickly fill rainwater harvesting systems and store the water for drier days. Because all of the sites in our study had access to city water, they may have been less concerned about water availability. However, we know that access to municipal water for urban farmers varies by city in the United States, and even more when we look globally (McDonald et al., 2014; Mitlin, n.d.; Nolasco, 2013; Water, n.d.). For example, access to municipal water was restricted for urban agriculture in New York City in 2001 as a result of drought, spurring many urban farmers and gardeners to turn to rainwater harvesting with the help of GrowNYC. New York City now has over 150 urban farms and gardens that harvest rainwater (GrowNYC, 2012). In the current study, more sites that harvested rainwater were prevented from expanding their site because of water access than those not harvesting rainwater. The lack of water access could have driven some farms and gardens to investigate alternative irrigation water sources, resulting in their use of harvested rainwater. In our study, farms and gardens’ concerns about water availability for site selection did not significantly impact their perspectives on rainwater harvesting. These results differed from Suri et al. (2019), who found that concern about water availability was significantly correlated with farmers’ willingness to use nontraditional water sources (e.g., recycled water, untreated surface water, and brackish surface water). Future studies among urban farmers and gardeners in diverse geographic regions will help further elucidate associations between water availability concerns and willingness to use harvested rainwater.

Characteristics of farmers/gardeners harvesting rainwater

There were no significant associations between any farming/gardening methods, water concerns and use, education, or demographics and rainwater harvesting practices, yet some interesting differences were present. The lack of significant associations could indicate that interest in rainwater harvesting is influenced by alternative factors, such as a general lack of knowledge about harvesting rainwater practices, social norms, or political beliefs, or could be an artifact of this study's small sample size. Dhaka et al. (2010) found that age, farming experience, innovativeness, environmental consciousness, and exposure to mass media significantly and positively impacted farmers in Bundi, India's perceptions of climate change. To adapt to climate change, 34% (n = 169) of respondents indicated that they started harvesting rainwater. In our study, more males than females were using a rainwater harvesting system. Previous literature has not specifically examined the relationship between gender differences and the use of harvested rainwater; however, this difference could be the result of males’ more general openness to exploring water alternatives compared to females. A study by Jianjun et al. (2015) found that male farmers in Yongqiao District, China, were more likely to use new technology for water conservation and spend more money on irrigation infrastructure compared to females. Yet, this differed from a study by Kanyi et al. (2016), which found that women farmers in Kenya had more positive attitudes about harvested rainwater compared to male farmers. Thus, more research should be conducted with a larger sample size to understand the role of gender in rainwater harvesting decisions.

Most farms and gardens that were harvesting rainwater in Baltimore tended to have smaller sites (less than 500 ft²/46 m²) compared to those not harvesting rainwater. This may suggest that it might be more feasible to irrigate a smaller amount of land with harvested rainwater. The existing literature on the impact of farm size on rainwater harvesting use is all from African countries, and the findings are contradictory. Bessah et al. (2022) and Wakeyo and Gardebroek (2017) found that large-scale farmers in Ghana and Ethiopia were more likely to harvest rainwater, whereas Mangisoni et al. (2019) saw that farmers in southern Malawi with smaller sites were more likely to harvest rainwater (2019). This might suggest that the impact of farm size on the use of rainwater harvesting might differ between countries because of climate, economic, or other variables. Additionally, more farmers that owned their land used rainwater harvesting systems, suggesting that making an investment in these systems is more acceptable to those who control their site. Mangisoni et al. (2019) similarly found that Malawian farmers with secure land tenure were more likely to harvest rainwater.

Rainwater harvesting and food safety considerations

The water source and irrigation method used in urban agriculture can impact the safety of the produce grown because of the possible presence of microbiological and heavy metal contaminants in various water sources (Uyttendaele et al., 2015). The Centers for Disease Control and Prevention (CDC) estimates that of 48 million people in the United States sickened each year from foodborne illnesses, 46% are due to produce (CDC, 2018; USDA, 2019). Since produce is commonly consumed raw, it poses a greater risk for foodborne pathogens compared to foods that are cooked or processed. While it is difficult to tell the sources of contamination, contaminated water can increase pathogens on produce (FDA, 1998) and has been linked to the source of several reported foodborne outbreaks (Uyttendaele et al., 2015). Rainwater harvested from a roof can collect contaminants present on the roof's surface and roof material, such as fecal bacteria and pathogens, heavy metals, pesticides, and PAHs (Sanchez et al., 2015). Uncovered rain barrels are more susceptible to contaminants and require screens to keep out twigs, leaves, animals, and mosquitoes (Ardoin, 2022). Ninety percent (n = 9) of the sites in our study were using covered rain barrels as part of their rainwater harvesting systems. Irrigation methods that are less likely to allow water to contact the produce pose the lowest food safety risk (Rock et al., 2019). Thus, subsurface and drip irrigation methods pose a lower risk, while overhead irrigation methods pose the highest risk (Uyttendaele et al., 2015). In our study most farms and gardens were irrigating by hose, followed by hand irrigation, and only 18% were using drip irrigation. Continuing food safety trainings and targeting urban farmers and gardeners could further reduce food safety risks from irrigation water sources. Water treatment methods can also reduce the risk of contamination. Common filtration methods for harvested rainwater include settling tanks, disinfection, membrane filtration, reverse osmosis, heat treatment, solar disinfection, and slow sand filtration followed by chlorination (Sanchez et al., 2015). However, in our study the majority of all sites (89%, n = 33) were not using any treatment or filtration for their water. Two out of the 10 sites harvesting rainwater (20%) reported using some type of water filtration. Although most sites were using municipal water for irrigation, depending on storage conditions, which could allow for algal and/or bacterial growth, filtration might still be helpful for improving water quality. In addition, in older cities, aging infrastructure can leach lead into municipal water, although this is not a documented problem in Baltimore. Lead can be removed by some filters (EPA, n.d.). Baltimore has however detected bacteria in their drinking treatment system in recent years (DPW, 2023; Portnoy, 2022).

Rainwater harvesting education needs for urban agriculture

Farms and gardens represented in our study had a wide variety of concerns related to rainwater harvesting. Sites interested in installing a rainwater harvesting system, but not currently harvesting rainwater, reported the most concerns. Their main concerns included the lack of knowledge about designing, installing, and maintaining a rainwater harvesting system, lack of awareness of city rules/regulations, microbial contamination of water, and system costs. However, these same sites were also the most interested in receiving additional information about using rainwater harvesting systems, including how to test water for food safety compliance and techniques for utilizing harvested rainwater. This suggests that while farmers and gardeners may have concerns, these concerns could be alleviated with education and training on rainwater harvesting and food safety practices. If farmers and gardeners felt more comfortable with using rainwater harvesting systems, the use of rainwater harvesting may increase. Previous studies agree that increased knowledge about sustainable agriculture practices and sustainable water sources leads to increased willingness to use these practices or water sources (Adnan et al., 2018; Hoffman et al., 2014; Suri et al., 2019; Zeweld et al., 2017). For example, Suri et al. (2019) found that farmers that were knowledgeable about nontraditional water (including agricultural runoff, treated wastewater, recycled water, produced water, untreated surface water, and brackish surface and groundwater) were significantly more likely to be willing to use these water sources.

Limitations

Our study provides valuable information about the current water uses, concerns, and rainwater harvesting practices among Baltimore urban farmers and gardeners. However, a larger number of respondents would provide a fuller picture of water issues among Baltimore growers. Urban farms and gardens, as well as their water needs, can vary by geographic region, therefore additional studies are needed in multiple regions and cities to determine the generalizability of our results. Although our study took place only among Baltimore farms and gardens, the results begin to show important education and training gaps and needs among urban farms and gardens concerning rainwater harvesting. Despite these limitations, this preliminary study begins to fill in a knowledge gap about the use of harvested rainwater in urban agriculture.

CONCLUSION

Our study provides an overall picture of current urban agricultural practices and perspectives on rainwater harvesting among a subset of Baltimore City farms and community gardens. Farms’ and gardens’ interest in rainwater harvesting was not significantly correlated with their current farming practices, water concerns and use, or demographics. These findings suggest that there could be alternative factors that influence farms and gardens’ interest in rainwater harvesting, such as social norms or political beliefs. It might also suggest that farms and gardens, regardless of their farming practices and water access, could be encouraged to use harvested rainwater.

Our results suggest that more food safety education and training, especially on water quality, is needed for urban farms and gardens. Most of the sites in our study were not using known strategies to improve food safety, such as water filtration or drip irrigation. Our respondents had multiple concerns about rainwater harvesting but expressed openness to learn more. However, the majority of the farms and gardens that we surveyed had never attended a food safety training. There is a valuable opportunity to expand the use of harvested rainwater in urban agriculture which could provide numerous environmental, social, and health benefits including improving crop yields and food access for urban communities, increasing resiliency and sustainability, and reducing environmental impacts, but the promotion of rainwater harvesting in urban agriculture needs to be paired with food safety training to increase the safe use of this water source.

AUTHOR CONTRIBUTIONS

Abriana Segal: Formal analysis; writing—original draft. Niya Khanjar: Formal analysis; writing—original draft. Julie Yang: Visualization; writing—review and editing. Kelsey Brooks: Investigation; methodology; writing—review and editing. Marcus Williams: Investigation; resources; writing—review and editing. Neith Little: Conceptualization; investigation; validation; writing—review and editing. Andrew Lazur: Conceptualization; funding acquisition; investigation; methodology; writing—review and editing. Rachel Rosenberg Goldstein: Conceptualization; formal analysis; funding acquisition; investigation; project administration; resources; supervision; writing—review and editing.

ACKNOWLEDGMENTS

The authors would like to thank the farmer and gardener participants and Marina Costa for her help in updating the figures. Funding for this project was provided by the University of Maryland College of Agriculture and Natural Resources, Interdisciplinary Projects for Community Resilience in Urban and Peri-urban Environments Seed Grant.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

References

Adnan, N., Nordin, S. M., Rahman, I., & Noor, A. N. (2018). The effects of knowledge transfer on farmers decision making toward sustainable agriculture practices. World Journal of Science, Technology and Sustainable Development, 15(1), 98–115. [DOI: https://dx.doi.org/10.1108/wjstsd-11-2016-0062]