In recent decades, as global population has continued to increase, so has the demand for food (Ackerman et al. 2014; Opitz et al. 2016). This demand is only projected to rise as not only the population increases, but also the percentage of inhabitants in urban areas increases as well (Lin et al. 2015; Opitz et al. 2016). This situation has led to many communities experiencing food insecurity, primarily in urban areas throughout the globe (Ackerman et al. 2014; Lin et al. 2015). It has been widely documented that low-income and disadvantaged communities have less access to nutritionally dense foods, and these areas of reduced access are often called food deserts (Lin et al. 2015; Opitz et al. 2016). Food insecurity and food deserts are now among the most pressing issues in US cities (Meenar and Hoover 2012). To address this rise in food demand, especially for nutritionally adequate food, various forms of urban agriculture have risen in popularity. Additional motivations, such as the desire for locally grown food, the fact that culturally important foods may not be available in grocery stores, the need to reduce inconveniences related to supply chain issues such as those seen during the coronavirus disease 2019 pandemic, and the environmental and health benefits of urban agriculture, are also contributing to its growth (Gunia 2020; Kuta 2021; Lin et al. 2015; McDougall et al. 2019; McGril 2021; Mok et al. 2014; Van Tuijl et al. 2018).
Urban agriculture has many definitions, but can be described simply as the process of growing food crops, or ornamental and medicinal plants—and even raising livestock—within cities and towns (Goldstein et al. 2016; Lin et al. 2015; Opitz et al. 2016). Some of the more common types of urban agriculture include community gardens, backyard gardens, and rooftop gardens, which are also referred to as green roofs (Mok et al. 2014). Although hydroponic and other indoor systems are available, outdoor soil or soil-based media production methods remain among the most common forms of urban agriculture (Mok et al. 2014). Furthermore, controlled environment agriculture and vertical farming require a huge initial investment and energy, making these methods less sustainable and suitable for small-scale farmers in an urban setting than outdoor vegetable production such as raised beds or green roofs (Barbosa et al. 2015; McDougall et al. 2019). It was estimated that urban agriculture could fulfill ∼15% to 20% of the global food supply (Knizhnik 2012; Lin et al. 2015), reaching as high as 90% of local vegetable, egg, and milk needs, and 70% of poultry needs in some cities (Nugent 2002). Since the early 1990s, urban agriculture in the United States has grown by more than 30%, primarily in underserved communities (Lin et al. 2015). Urban agriculture could provide 7% to 8% of the current vegetable consumption in Oakland, CA, USA (McClintock et al. 2013) and 15% of the food supply in Sydney, Australia (McDougall et al. 2020), when unoccupied urban land areas are used for food production.
Urban agriculture has also gained popularity in the past two decades because of greater public awareness and concern for carbon footprints. Environmental benefits of urban agriculture include supporting native biodiversity by providing food and habitat resources, mitigating air pollution and urban heat island effects, providing stormwater management, and lowering energy use required for food transport (Ackerman et al. 2014; Lin et al. 2015; Mok et al. 2014). Benefits of urban agriculture can also include community building, mitigation of childhood obesity and malnutrition, improved mental and physical health, and educational benefits to students (Bahamonde 2019; Colman 2017; Ghose and Pettygrove 2014; Lin et al. 2015; Meenar and Hoover 2012; Monroe 2015; Nogeire-McRae et al. 2018; Ohly et al. 2016; van Averbeke 2007). In particular, creating urban farms in low-income communities can revitalize these communities by promoting social cohesion and improving economic well-being (Angotti 2015).
Benefits of urban agriculture to the environment have been widely documented; however, there is also the potential for some negative effects associated with the increased practice of urban agriculture. One of the most significant environmental concerns with any agricultural system is the transport of excess nutrients and other agriculture-associated contaminants into waterways (Berka et al. 2001; Hart et al. 2004; King and Torbert 2007; Kleinman et al. 2011). The use of conventional or manufactured fertilizer has been attributed to increased rates of nutrient runoff, especially when applied before periods of increased precipitation (King and Torbert 2007). The same applies to outdoor urban agriculture (Bachman et al. 2016; Lusk et al. 2020), especially in areas where management is switching from no or low fertilizer use to greater fertilizer use for crop production (Bachman et al. 2016; Janke et al. 2017; Spence et al. 2012), and in areas such as parking lots that lack surrounding vegetation (Hale et al. 2015; Shetty et al. 2019). In addition, the use and overuse of nutrient sources has been shown to contribute to nutrient runoff from both green roofs (Czemiel Berndtsson 2010; Mitchell et al. 2017; Toland et al. 2012) and ground-level systems (Cameira et al. 2014; Dewaelheyns et al. 2013; Huang et al. 2006; Salomon et al. 2020; Shrestha et al. 2020; Small et al. 2019; Wielemaker et al. 2018, 2019). These substances can alter and affect the water quality of runoff negatively, which can lead to the impairment or degradation of nearby aquatic systems as well as potential health hazards (Berndtsson et al. 2009).
Nitrogen (N) and phosphorus (P) are the nutrients commonly found in fertilizers most associated with increased aquatic plant or algal growth and eutrophication risks (Anderson et al. 2002; Conley et al. 2009; Correll 1998; Smith and Schindler 2009). Nitrogen, which supports protein synthesis, and P, which is needed for DNA, RNA, and energy transfer, are needed by both terrestrial and aquatic plants (Conley et al. 2009). In excess, however, the presence of N and P can accelerate the growth of aquatic plants and harmful cyanobacteria (Conley et al. 2009). Anthropogenic eutrophication and dead zones are the number-one problem facing aquatic ecosystems globally, and can affect all types of aquatic systems (Kleinman et al. 2011; Smith and Schindler 2009). Nitrogen in reactive forms such as ammonium
Important physicochemical properties of water, such as pH, electrical conductivity, color, and turbidity, help to explain the quality of runoff water and its possible effect on aquatic life cycles (Rameshkumar et al. 2019; Whittinghill et al. 2016). Use of different growing media, sources of fertilizer, and crop management practices may have an effect on these criteria. A pH that is too high or too low can kill many aquatic species, affect hatching and survival rates, and stress the entire aquatic animal system (Freda 1987). Most aquatic animals prefer a pH of 6 to 9 (Collier et al. 1990). Changing pH levels (high or low) may also facilitate the solubilization of numerous harmful heavy metals, hence increasing the risk of absorption by aquatic organisms (Gensemer et al. 2018). Electrical conductivity is a measure of the salinity of water. Increasing the salt level in freshwater aquatic systems may increase the cost of water treatments for human consumption, reduce freshwater diversity, alter ecosystem function, and, ultimately, affect socioeconomic well-being by altering the goods and services of the freshwater aquatic system (Cañedo-Argüelles et al. 2016). Water color, measured on the platinum/cobalt (Pt/Co) scale, is usually used to analyze the pollution level in wastewater. Water with a yellow tint has more watercolor on the Pt/Co scale and is considered more polluted. In general, such color in water is a result of the humic and fulvic fractions of dissolved organic compounds (Bennett and Drikas 1993). Turbidity is a water-quality parameter that measures the optical clarity of water (Davies-Colley and Smith 2001). Increased turbidity means less solar radiation penetration. High turbidity influences aquatic life through a reduction in photosynthesis and dissolved oxygen, affects movement resulting from poor visualization, and kills fish directly or reduces their growth (Sader 2017). Turbidity could also be a good factor to predict aquatic life diversity and richness (Figueroa-Pico et al. 2020).
There has been extensive research over many decades on the effects of conventional farming practices on excess nutrient input and pesticide contamination, and the associated impacts of these contaminants on the water quality of various aquatic ecosystems (Berka et al. 2001; Elrashidi et al. 2005; Gaudreau et al. 2002; Hart et al. 2004; Heathwaite et al. 1998; King and Torbert 2007; Kleinman et al. 2011; Liu et al. 2014; McLeod and Hegg 1984). Until recently, there was a lack of research on monitoring these same effects from commonly used urban agricultural systems at ground level. Research has focused more on the use of green roof systems as opposed to soil-based urban agriculture systems used at ground level (Beck et al. 2011; Czemiel Berndtsson 2010; Emilsson et al. 2007; Getter and Rowe 2006; Kok et al. 2013; Toland et al. 2012; Whittinghill et al. 2016). At ground-level urban sites, where use of soil fertility testing and nutrient application recordkeeping can be limited (Small et al. 2019; Whittinghill and Sarr, 2021; Wielemaker et al. 2019; Witzling et al. 2011), overapplication of nutrients is especially common. This has been tied to nutrient buildup in the soil (Abdulkadir et al. 2013; Cameira et al. 2014; de Barros Sylvestre et al. 2019; Dewaelheyns et al. 2013; Huang et al. 2006; Salomon et al. 2020; Shrestha et al. 2020; Small et al. 2019; Witzling et al. 2011) and an increase in the nutrient concentration of runoff water (Huang et al. 2006; Jackson et al. 1994; Shrestha et al. 2020). Excessive nutrient losses can also be attributed to a preference for compost as a nutrient source (Cameira et al. 2014; Dewaelheyns et al. 2013; Metson and Bennett 2015; Small et al. 2019; Wielemaker et al. 2019); the lower fertilizer nutrient equivalencies for composts the year of application, with continued nutrient release in subsequent years; and a difference in the availability of nutrients from compost (Maltris-Landry et al. 2016; Mikkelsen and Hartz 2008; Wielemaker et al. 2019). It has been suggested that these nutrient inefficiencies in urban agriculture could represent a significant component of the global P budget, if urban agriculture were scaled up to its full potential (Small et al. 2019). It is, therefore, imperative to understand more fully the effects of nutrient runoff from ground-level urban agriculture systems on water quality. Our research compared water-quality variables from four different nutrient sources applied to raised beds and container gardens.
The container gardens used in this experiment were constructed from small plastic wading pools. This growing system has been used on rooftops (Hell’s Kitchen Farm Project 2020) and has received increased social media attention (Michaels 2021; Pinterest 2021). These small plastic pools are readily available during the growing season and can be purchased at a fairly low cost. When comparing cost per area, small plastic pools cost about $12 for 1 m2 of planting area, compared with almost $70 for raised beds with sides (Durham et al. 2018), ∼$40 to $80 for nursery pots (Hummert International 2021), and as much as $127 for other commercially available planters (Lowe’s 2021). This makes the pool growing system a low-cost option suitable for urban areas without soil or where soil is contaminated and considered unsafe for growing. However, this growing system has received very little scientific attention, especially in terms of how it may affect yields and nutrient leaching when compared with raised beds and other in-ground production systems. Including the small plastic pool growing system in our research is a first step in addressing this knowledge gap.
Materials and Methods
Data collection for this study began in 2018 and continued into 2019 at the Harold R. Benson Research and Demonstration Farm in Frankfort, KY, USA, using two different ground-level production systems that are commonly used for urban agriculture: raised beds and containers. For the construction of these systems, all the materials were purchased from Lowe’s Co., Inc. (Mooresville, NC, USA). All lumber used was pressure-treated with EcolifeTM (Viance, LLC, Charlotte, NC, USA). The raised beds and containers were constructed in May 2018 over a 156.1-m2 area and treated with one of four different fertilizer treatments. Both the raised beds and containers were elevated 0.6 m off the ground on 1.22- × 1.22-m platforms. The upright supports were constructed using 10.16- × 10.16-cm lumber. The deck of the platform was constructed with 5.08- × 10.16-cm lumber for the frame and two joists, then was topped with severe-weather common, square southern yellow pine plywood sheeting. Sixteen of these elevated platforms were topped with raised beds; the remaining 16 were topped with containers. Both raised beds and pool containers were fitted with gutters and downspouts connected to a 7.6-L bucket (Fig. 1).
The gutter and downspout with bucket attachment for water collection pictured on a raised-bed platform.
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
The raised beds measured 1.22 × 1.22 m, with a depth of 30.48 cm, and were made using 5.08- × 10.16-cm and 5.08- × 15.24-cm lumber. They were then lined with black Smartpond® nylon mesh pond liner (West Palm Beach, FL, USA) to act as a barrier, then filled to a depth of 20.32 cm with premium shredded topsoil (Table 1). To collect runoff from the raised beds and avoid losing soil, a 5.08-cm gap covered with Phifer Super Solar charcoal fiberglass replacement screen (Phifer, Inc., Tuscaloosa, AL, USA) was left in the front of each bed. Covered gutters were then attached to the front of each bed at this gap.
Analysis results of the premium shredded topsoil as provided by the University of Kentucky Soil Testing Laboratory.
Blue, plastic, round wading/kiddie pools (Summer Waves, purchased at Lowe’s, Mooresville, NC, USA) were used for the container plots. These pools measured 114 cm in both length and width, and were filled to a depth of 17.78 cm with premium shredded topsoil (Table 1), which was the maximum amount of soil these containers could hold. To allow for water drainage, two 1.27-cm holes were drilled at the front of each pool, even with the bottom of the pool. The holes were attached to gutters using Eastman 1.27-cm polyvinylchloride clear vinyl tubing (Eastman Chemical Co., Kingsport, TN, USA). To improve drainage after installation, a small amount of Rooflite® drain media wrapped in Rooflite® separation fabric (Skyland, LLC, Landenberg, PA, USA) was added to each plot over the drainage holes.
Following a randomized complete block design with modifications to reduce edge effects, which was necessary because of varying topography and shading within the study area, four fertilizer treatments were replicated four times in each of the two systems. The four fertilizer types used were conventional (10N–10P–10K Twin Pines® All Purpose Fertilizer, Know, IN, USA), organic (Espoma® 3N–4P–6K Tomato Tone, 12N–0P–0K Blood Meal, and 4N–12P–0K Bone Meal, Milleville, NJ, USA), 14.42 kg/m2 compost (0.1N–0.1P–0.1K; Michigan Peat Garden Magic® Compost and Manure, Houston, TX, USA), and 7.21 kg/m2 compost + organic fertilizer (0.1N–0.1P–0.1K; Michigan Peat Garden Magic® Compost and Manure, and Espoma® 3N–4P–6K Tomato Tone and 12N–0P–0K Blood Meal). The fertilizers and compost used are readily available to urban farmers and growers in the region. The target nutrient application rates for each of these fertilizer treatments were 19.61 g/m2 N, 16 g/m2 phosphorus pentoxide, and 16 g/m2 potassium oxide, as recommended for greens by the College of Agricultural and Environmental Sciences at the University of Georgia (Athens, GA, USA). The actual nutrients supplied by each treatment are listed in Table 2. The conventional fertilizer treatment was applied to supply the recommended amount of N. The organic fertilizer treatment used the three listed fertilizers to supply the recommended amounts of all three nutrients. The nutrient contributions of the low-compost treatment were estimated to be 6.59 g/m2 N, and 7.21 g/m2 P and potassium (K) based on the listed product nutrient content, so small amounts of all three organic fertilizers were used to supply the remaining nutrients to meet the full nutrient recommendation. The compost-only treatment was estimated to supply 14.42 g/m2 P and K in the first year after application, based on the listed product nutrient content. Fertilizer was applied before each planting, whereas compost was applied at the beginning of each growing season (Table 3).
Nutrients added by compost at the beginning of the growing season and by fertilizers with each planting in each nutrient management treatment.
The nutrient management, planting, and harvesting timeline with calendar dates and days after compost addition for the 2 years of the study for both the container and raised-bed growing systems.
Crops planted included seven types of greens, which were planted in succession in each plot of every treatment: Lactuca sativa (Encore lettuce mix), Eruca sativa (Astro arugula), Brassica rapa (Mizuna Asian greens), Brassica juncea (Red giant mustard greens), Beta vulgaris (Fordhook giant Swiss chard), Brassica napus (Red Russian kale), and Spinacia oleracea (Covair spinach). Some of the plants listed were grown in 2018 but not 2019, or vice versa (Table 3). Planting started later than expected in 2018 because of plot construction, so lettuce was not planted, and the arugula crop was planted later than expected. In 2019, Swiss chard was added during the warmest part of the summer because of its greater heat tolerance. A drought was also experienced in Aug, Sep, and Oct 2019, which slowed crop growth, so spinach was not planted. The same planting density, 386 plants/m2, was used for each crop in all treatments. This planting density was derived from a variety of sources, including local Extension recommendations, other Extension recommendations for the production of baby greens, and the Johnny’s Selected Seed website (https://www.johnnyseeds.com). Plants were irrigated with drip irrigation to supply 2.5 cm of water per week, including rain.
Runoff samples were collected from each plot once a month, weather permitting, from Jul 2018 through Feb 2020. Water-quality analysis was not performed for 4 separate months during the study period: Dec 2018, and Mar, Sep, and Nov 2019. No water was collected during these months because there was either no rain (Sep 2019) or insufficient rain, or larger rainstorms occurred when samples could not be collected. When storms early in the month did not result in enough runoff water, further attempts were made to collect water later in the month. Samples of 250 mL each were collected from each plot with a functional gutter and downspout, and were transported to Kentucky State University’s Aquaculture Research Center for laboratory analysis. The pH values were measured using a FisherbrandTM Accumet™ AP110 Meter Kit (Thermo Fisher Scientific, Inc., Waltham, MA, USA). Conductivity values were acquired using an Oakton® Con 6+ Meter Kit (Vernon Hills, IL, USA). Color was analyzed using a LaMotte® Smart3 Colorimeter (Lamotte Co., Chestertown, MD, USA), and turbidity was measured using a LaMotte® 2020 we turbidity meter. Nitrate-N,
Occasional issues with downspout or collection bucket failure occurred and resulted in lower than necessary water collection volumes. In these cases, either no analysis was performed or some of the analyses were performed and were chosen based on water volume requirements to maximize the number of analyses that could be performed. Water collection system failures increased in the second study year as downspouts aged. Laboratory equipment failure resulted in the inability to test accurately for K in runoff samples from most container plots in Sep 2018. Nitrate-N could not be tested for any plots in Jul 2019 because the laboratory was shorthanded and samples needed to be stored for longer than recommended for that analysis. This, combined with occasional downspout failure or low runoff volume, resulted in missing data points for some of the plots in each growing system and each fertilizer treatment at least once. Because sampling took place on different days after compost addition (DAC) in study years 1 and 2, comparisons between years were limited to samplings that took place within 7 d of each other when there was a significant interaction between DAC and study year, and the study year main effect as appropriate. There were four pairs of sampling dates that were within 7 d of each other: 61 and 67, 145 and 152, 184 and 186, and 281 and 286 (Table 4). Average air temperature and total precipitation data were obtained from the Franklin County MESONET weather station located at the Harold R. Benson Research and Demonstration Farm, 250 m from the experimental site. Climatic normal air temperature and precipitation data based on a time period from 1981 to 2010 were obtained from the National Oceanographic and Atmospheric Administration National Climate Data Center (2020) for the weather station at the Capital City Airport in Frankfort, KY, USA.
Water sampling dates and their corresponding days after compost addition for the 2 years of the study.
Statistical analysis was performed in R (version 1.2.5001; R Foundation for Statistical Computing, Vienna, Austria). Data that did not conform to a normal distribution were transformed using cube root (conductivity), log10 (color, K) and inverse (turbidity,
Results
Weather data.
In total, 1341 mm of rainfall occurred during the study period in 2018 and 1483 mm in 2019. Precipitation during the study period in 2018 was greatest in August (236 mm) and lowest in June (135 mm) (Fig. 2A). Precipitation in 2018 was more than the climatological normal during all months of the study period. In 2019, October had the greatest rainfall, with a total of 232 mm, which was 16% of the annual total. July, August, and September received lower than normal rainfall in 2019, with just less than 76 mm total monthly rainfall for July and August. There was a drought in Kentucky in 2019 from the last week of August to early October. There were only 10 mm of precipitation between 28 Aug and 6 Oct. Temperature patterns in 2018 and 2019 were similar to each other and to the climatological normal temperatures (Fig. 2A).
(A) Monthly total precipitation and average temperature for the period of study between May 2018 and December 2019, and the 30-year (1981–2010) climatological normal (National Oceanic and Atmospheric Administration, National Climate Data Center 2020). (B) Total precipitation between sampling times for the first (2018) and second (2019) years of the study, where d: is the number of days between that sampling and the previous sampling.
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
Physical and chemical characteristics.
Model simplification was performed to remove nonsignificant interactions with an alpha level of 0.05. All three-way interactions were not significant for pH, conductivity, color, and turbidity (Supplemental Table S1). All two-way interactions were also not significant for pH, color, and turbidity (Supplemental Table S1). Only significant interactions and main effects are discussed further.
The conventional fertilizer plots had the least runoff pH (7.51 ± 0.04), but were only significantly less than the compost-only plots (7.67 ± 0.04) (Fig. 3A). There were significant differences in runoff pH across sampling times (Fig. 3B), with a slight increase from the start of the experiment to 131 DAC, followed by a slight decrease to the end of the study year. Runoff water pH was least at 286 DAC (7.11 ± 0.04) and 328 DAC (7.12 ± 0.10), and greatest at 131 DAC (8.19 ± 0.03). There were significant differences between the first and second study year for two of the four pairs of sampling dates: 61 and 67 DAC (7.99 ± 0.06 vs. 7.67 ± 0.07), and 281 and 286 DAC (7.62 ± 0.04 vs. 7.10 ± 0.04). In both cases, the first study year was significantly greater than the second study year. The runoff collected from the container system had significantly lower pH values (mean ± SE, 7.53 ± 0.03) in comparison with the runoff collected from the raised-bed system (7.62 ± 0.03) (Fig. 3C).
The pH of runoff water for each (A) nutrient management treatment, (B) sampling time in days after compost addition for study years 1 and 2, and (C) growing system. US Environmental Protection Agency (2022a) freshwater minimum and human health maximum pH (US Environmental Protection Agency 2022b) thresholds are included as solid and dashed lines, respectively. Letters denote significant differences between growing systems, among nutrient management treatments, or among sampling times. Max = maximum; Min = minimum.
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
There were no significant differences among fertilizer treatments in the second year of the study, but in the first year of the study, conductivity was greatest in runoff from the conventional fertilizer treatment (608 ± 73 μS/cm), but was only significantly greater than the compost-only treatment (277 ± 26 μS/cm) (Fig. 4A). Conductivity differed significantly by sampling time (Fig. 4B). During the first study year, conductivity was significantly greater during the growing season (61, 84, 131, and 152 DAC) than during the winter months. During the first study year, this pattern was not as visible, when conductivity was greater than the winter months only for the first (4 DAC) and third (67 DAC) samplings. Runoff water conductivity was greater in the first year of the study than the second year of the study in both growing systems when averaged across nutrient management treatment and sampling time (Fig. 4C), in all nutrient management treatments when averaged across growing system and sampling time (Fig. 4A), and for all four pairs of sampling dates when averaged across growing system and nutrient management treatment (Fig. 4B). In the first year of the study, the raised beds (539 ± 42 μS/cm) had significantly greater conductivity than the container plots (350 ± 33 μS/cm), but not during the second year of the study (217 ± 28 and 162 ± 23 μS/cm, respectively) (Fig. 4C).
The conductivity of runoff water for each (A) nutrient management treatment, (B) sampling time in days after compost addition for study years 1 and 2, and (C) growing system. Letters denote significant differences among means. Uppercase letters denote differences between growing systems and among nutrient management treatments within study year; lowercase letters denote significant differences between study years within growing system and nutrient management treatment.
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
There was no significant difference in color among the nutrient management treatments (Fig. 5A). There were significant differences among sampling times for color, with a slight decrease in color with time from compost addition (Fig. 5B), but color was greatest in runoff from the conventional fertilizer treatment (742 ± 114 Pt/Co) and least in the compost-only treatment (453 ± 56 Pt/Co). The greatest color was seen at 44, 55, 84, and 281 DAC (1739 ± 329, 1585 ± 293, 1022 ± 79, and 1684 ± 197 Pt/Co, respectively), which were all significantly greater than the least color seen at 145, 243, 286, and 328 DAC (83 ± 18, 126 ± 40, 75 ± 18, and 144 ± 38 Pt/Co, respectively). The only pair of sampling dates that had color significantly different between the 2 study years was 281 and 286 DAC, for which study year 1 was significantly greater than study year 2 (Fig. 5B). Runoff water color was significantly greater from the raised beds (645 ± 66 Pt/Co) than from the containers (464 ± >53 Pt/Co) (Fig. 5C).
The color of runoff water for each (A) nutrient management treatment, (B) sampling time in days after compost addition for study years 1 and 2, and (C) growing system. US Environmental Protection Agency (1986) domestic water supply maximum thresholds are included as dashed lines. Letters denote significant differences between growing systems, among nutrient management treatments, or among sampling times. Max = maximum; PtCo = platinum/cobalt scale.
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
Turbidity of runoff water from the conventional fertilizer treatment [31 ± 5 nephelometric turbidity unit (NTU)] was significantly greater than the organic fertilizer treatment (17 ± 3 NTU), but there were no other significant differences among nutrient management treatments (Fig. 6A). Turbidity followed the same pattern as color over time, with the greatest samples at 44, 55, 84, 253, and 281 DAC (50 ± 9, 40 ± 7, 23 ± 2, 42 ± 9, and 91 ± 11 NTU, respectively), and the lowest at 145 DAC (1 ± 0), followed closely by 243 DAC (6 ± 3 NTU) (Fig. 6B). Three out of four paired sampling dates had significant differences between study years: 145 and 152, 184 and 186, 281 and 286. For all three, the samples from year 1 had greater turbidity than the samples from year 2 (Fig. 6C). Turbidity of runoff water from the raised beds (23 ± 2 NTU) was significantly greater than that of the containers (17 ± 2 NTU) (Fig. 6C).
The turbidity of runoff water for each (A) nutrient management treatment, (B) sampling time in days after compost addition for study years 1 and 2, and (C) growing system. Letters denote significant differences between growing systems, among nutrient management treatments, or among sampling times. NTU = nephelometric turbidity unit.
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
Macronutrients.
Model simplification was performed to remove nonsignificant interactions with an alpha level of 0.5. All three-way interactions were not significant for
Within the container plots averaged across study year and sample time, the compost + organic fertilizer treatment had the greatest
Nitrate-nitrogen concentration from runoff from each growing system and nutrient management treatment combination averaged across sampling times and study years for (A) the entire range of the data and (B) the range containing the majority of the observations. Letters denote significant differences among means. Uppercase letters denote differences among nutrient management treatments within growing systems; lowercase letters denote significant differences between growing systems within nutrient management treatment. US Environmental Protection Agency (2022b) drinking water maximum and national pollutant discharge elimination system stormwater discharge (US Environmental Protection Agency 2015) thresholds are included as solid and dashed lines, respectively. Max = maximum.
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
Nitrate-nitrogen concentration from runoff from each nutrient management treatment and study year combination averaged across growing systems and sampling times for (A) the entire range of the data and (B) the range containing the majority of the observations. Letters denote significant differences among means. Uppercase letters denote differences among nutrient management treatments within growing systems; lowercase letters denote significant differences between growing systems within nutrient management treatment. US Environmental Protection Agency (2022b) drinking water maximum and national pollutant discharge elimination system stormwater discharges maximum (US Environmental Protection Agency 2015) thresholds are included as solid and dashed lines, respectively. Max = maximum.
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
Nitrate-nitrogen concentrations from each sampling time in days after compost addition for study years 1 and 2 for (A) the entire range of the data and (B) the range containing the majority of the observations. Letters denote significant differences among means. US Environmental Protection Agency (2022b) drinking water maximum and national pollutant discharge elimination system stormwater discharges maximum (US Environmental Protection Agency 2015) thresholds are included as solid and dashed lines, respectively. Max = maximum.
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
There were no significant differences among fertilizer treatments for most sampling times, except 61, 84, and 186 DAC (Fig. 10A–D). In all three cases, the conventional fertilizer treatment had the greatest average
Distribution of mean values of ammonia from runoff from each sampling time in days after compost addition for study years 1 and 2 for the (A) conventional fertilizer, (B) organic fertilizer, (C) low-compost + organic fertilizer, and (D) compost-only nutrient management treatments, and (E) the two growing systems (containers and raised beds). US Environmental Protection Agency (2013) acute and chronic exposure thresholds are included as solid and dashed lines, respectively. Letters and asterisks denote significant differences among means. Uppercase letters denote differences among nutrient management treatments within sampling times; lowercase letters denote differences among sampling times within nutrient management treatment. Asterisks denote significant differences between study years within nutrient management treatment for paired sampling dates, with two asterisks indicating the lower mean.
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
There were few differences among fertilizer treatments within study year and growing system. In general, the conventional fertilizer treatment had the greatest runoff water P concentrations of all nutrient management treatments. The conventional fertilizer treatment was significantly greater than all other nutrient management treatments (range, 0.061 ± 0.07 to 0.70 ± 0.16 and 0.47 ± 0.07 to 0.68 ± 0.13 mg/L, respectively) in the containers in study year 1 and the raised beds in study year 2 (2.37 ± 0.37 and 3.36 ± 0.76 mg/L, respectively) (Fig. 11). For the raised beds in study year 1, the conventional fertilizer treatment (2.39 ± 0.75 mg/L) was only significantly greater than the organic fertilizer and compost-only treatments (0.62 ± 0.13 and 0.89 ± 0.24 mg/L, respectively). At 61 DAC, the conventional fertilizer treatment (6.28 ± 2.27 mg/L) was significantly greater than the organic fertilizer treatment (0.70 ± 0.28 mg/L) (Fig. 12). At 67 and 281 DAC, the conventional fertilizer treatment (1.88 ± 0.49 and 2.59 ± 0.49 mg/L, respectively) was significantly greater than the conventional fertilizer treatment (0.30 ± 0.08 and 0.40 ± 0.08 mg/L, respectively) (Fig. 12). No other significant differences among nutrient management treatments were found. There are no significant differences in runoff P content over time for the organic fertilizer treatment (Fig. 12B). For the conventional fertilizer and compost + organic fertilizer treatments, there were two peaks in runoff content: at 61 and 84 DAC, respectively, and at 186 and 281 DAC, respectively (Fig. 12A and C). The compost-only treatment appears to have only the first peak at 84 DAC, with lower and consistent P levels after 131 DAC (Fig. 12D). There was a significant difference between study years for the conventional fertilizer treatment in container plots, where the first study year was significantly greater than the second (Fig. 11). There were no other significant differences in runoff P concentration between study years within any other nutrient management treatment–growing system combination or for any paired sampling times (Fig. 12). No significant differences in P were found between growing systems within any nutrient management treatment in the first study year, and there was only a significant difference between growing systems for the conventional fertilizer treatment in the second study year (Fig. 11). In this case, the raised bed had greater runoff water P concentration than the container.
Distribution of phosphorous concentrations from runoff from four nutrient management strategies within each growing system for study years (A) 1 and (B) 2. Letters and asterisks denote significant differences among means. Uppercase letters denote differences between growing systems within nutrient management treatments; lowercase letters denote differences among nutrient management treatments within growing system. Asterisks denote significant differences between study years within nutrient management treatment and growing system, with two asterisks indicating the lower mean. US Environmental Protection Agency (2015) stormwater discharge maximum, and threshold to prevent the development of biological nuisances and to control eutrophication (US Environmental Protection Agency 1986) are included as solid and dashed lines, respectively.
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
Distribution of phosphorous concentrations from runoff for each sampling time in days after compost addition for the (A) conventional fertilizer, (B) organic fertilizer, (C) compost + organic fertilizer, and (D) compost-only nutrient management treatments. Letters denote significant differences among means. Uppercase letters denote differences among nutrient management treatments within sampling time; lowercase letters denote differences among sampling times within nutrient management treatments. US Environmental Protection Agency (2015) stormwater discharge maximum, and threshold to prevent the development of biological nuisances and to control eutrophication (US Environmental Protection Agency 1986) are included as solid and dashed lines, respectively.
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
Within the raised beds, the conventional fertilizer treatment also had the greatest K in runoff of all nutrient management treatments (Fig. 13A). It was significantly greater than the compost + organic fertilizer (21.1 ± 2.5 mg/L) and compost-only treatments (17.0 ± 2.7 mg/L), but not the organic fertilizer treatment (32.3 ± 5.4 mg/L). There were no significant differences among nutrient management treatments within the container system (range, 12.6 ± 2.1 to 28.7 ± 6.8 mg/L). There were significant differences among sampling times, with greater average K and greater variability in K in runoff and during the growing season (Fig. 13B). Runoff K was greatest at 84 DAC (65.2 ± 5.3 mg/L) and least at 248 DAC (4.0 ± 1.8 mg/L). There was a significant difference between study years for one pair of sampling dates: 145 and 152 DAC. Runoff K at 152 DAC (study year 1) (51.2 ± 7.1 mg/L) was among the greatest of all sampling dates, and 145 DAC (study year 2) (27.8 ± 17.1 mg/L) was significantly less (Fig. 13B). There was only a significant difference in K in runoff between container plots and raised-bed systems in the conventional fertilizer treatment, where K in runoff from the raised bed (69.6 ± 12.6 mg/L) was significantly greater than the container plots (28.7 ± 6.8 mg/L) (Fig. 13A).
Distribution of potassium concentration from runoff from (A) each growing system and nutrient management treatment combination, and (B) each sampling time in days after compost addition. Letters denote significant differences among means. Lowercase letters denote differences among nutrient management treatments within growing system (A) or among sampling times (B). Uppercase letters denote differences between growing systems within nutrient management treatment (A) or study years for pairs’ sampling dates (B).
Citation: HortScience 58, 8; 10.21273/HORTSCI17215-23
Discussion
Nutrient management treatment.
The small difference between the pH of runoff water from the different treatments (7.51 vs. 7.67) and a lack of difference in the proportion of samples from any fertilizer treatment that do not meet the US Environmental Protection Agency pH water-quality standards for freshwater (US Environmental Protection Agency 2022a) and human health (US Environmental Protection Agency 2022b) (χ2 = 0.279, P = 0.965) suggest that the statistically significant difference (Fig. 3B) is not meaningful from a water-quality standpoint. Variability among the samples may be obscuring the patterns seen in other studies. Applications of manure-based composts increase the pH of soils (Adugna 2018; Beochat et al. 2013; Bowden et al. 2007; Giannakis et al. 2014; Gilley and Eghball 2002). Use of compost also increases soil pH above that of soils that have received inorganic fertilizers (Bowden et al. 2007; Dewaelheyns et al. 2013). Some evidence of a dose response to compost has been found (Adugna 2018; Giannakis et al. 2014), although that was not seen here.
Although not always significantly different from the other nutrient management treatments, the conventional fertilizer treatment was greatest for conductivity (Fig. 4A), color (Fig. 5A), and turbidity (Fig. 6A). The conductivity levels found in our study appear similar to those found in other studies in urban settings. Whittinghill et al. (2016) observed conductivity values up to 473 μS/cm from an agricultural green roof. In some cases, values observed by our study exceeded this, although most were less than 500 μS/cm. Toor et al. (2017) monitored runoff from a residential neighborhood and found conductivity values sometimes twice as high (3640 μS/cm) as those observed in our study. Although agriculture can increase runoff conductivity over some urban landscapes (Whittinghill et al. 2016), results suggest that these effects may be no worse than those of some other land uses (Toor et al. 2017).
Similar results for color have been reported in runoff from green roofs (Berghage et al. 2009) and container production (Hoskins et al. 2014). This suggests that application of chemical fertilizers could lead to a yellow to brown color of runoff water, which when mixed with proximal water sources could produce “objectionable color” for aesthetic purposes. The US Environmental Protection Agency (1986) criteria for water quality lists this objectionable color at a threshold of 75 Pt/Co for sources of domestic water supply. All of the average color values observed in our study exceed the 75 Pt/Co threshold, regardless of fertilizer treatment (Fig. 5A), and there were no differences in the proportion of samples that surpassed this threshold among the fertilizer treatments (χ2 = 3.027, P = 0.387). Color and turbidity are also listed as important for the depth of the compensation point for photosynthetic activity in a waterbody, which should not be reduced by more than 10% (US Environmental Protection Agency 1986). It is unclear how color and turbidity values observed in this study would affect this measure in downstream water bodies, but there is a correlation (Pearson’s R = 0.8149, P < 0.001) between high color and high turbidity in samples collected in our study. This correlation and the connection to a known water-quality issue could warrant further study of the effects of the color and turbidity of runoff water from urban agriculture. The greater turbidity of runoff observed here than in a previous study on a rooftop farm (Whittinghill et al. 2016) suggests that the growing media influence turbidity. In our study, a lot of silt or clay was observed settling in some of the water collection sample bottles. The water in these samples was more cloudy visibly than that of other samples and took longer to filter before other analyses could be performed.
In general, the conventional fertilizer treatment also contained greater amounts of
In other studies, the type of fertilizer applied to agricultural land has been shown to have a direct impact on the amount of
The levels of
Phosphorus levels in runoff from the organic fertilizer, compost + organic fertilizer, and compost-only treatments in our study are similar to those found in other studies on container production (Pershey et al. 2015), residential communities (Toor et al. 2017), other urban runoff (Li et al. 2015), and green roofs (Matlock and Rowe 2017; Whittinghill et al. 2016). Phosphorus concentrations in runoff from the conventional fertilizer treatment in our study were greater than many of those reported, except for those reported by Toor et al. (2017), when reclaimed water was used (average, 12.5 mg/L). These differences can be attributed to two causes. First, research in traditional agricultural settings has shown that the form of nutrient applied also influences nutrient losses. Heathwaite et al. (1998), for example, found that grassland areas receiving inorganic fertilizers had greater P loss than the grassland areas receiving solid cattle manure and liquid cattle slurry. Although the N and P applied per hectare was more in areas treated with solid cattle manure than in areas treated with inorganic fertilizer, the P loss was greater from areas treated with inorganic fertilizer. A similar result was found by Gaudreau et al. (2002), where runoff loss of dissolved P from the manure fertilizer was found to be 44% less than the runoff from the inorganic fertilizer at equal P application rates in turfgrass plots. This was thought to be because of the less soluble and less transportable nature of manure nutrients.
Second, mean P (Fig. 11) and K (Fig. 13A) in runoff from plots using conventional fertilizer could be explained, in part, by how the fertilizer was applied. The conventional fertilizer was applied to supply adequate N. In a 10N–10P–10K fertilizer with crops that require more N than P or K, this means that P and K will be oversupplied. In this case, 3.61 g/m2 of extra P and K were applied to the conventional fertilizer plots. Three organic fertilizers were used to supply the recommended amounts of N, P, and K for both the organic fertilizer and compost + organic fertilizer treatments. The application method used for the conventional fertilizer appears to have had a greater effect on P in runoff than K. Despite the overapplication of K in the conventional fertilizer treatment, concentrations of K in runoff water from that treatment were not greater than that of the organic fertilizer treatment in either growing system or the compost + organic fertilizer treatment in the container growing system. One study on growing media for containers suggests that excess K may be retained in the media cation exchange capacity and swapped out for other anions such as calcium and magnesium (Hoskins et al. 2014). These other cations were not measured in our study, but could account for the differences in the effect that overapplication of P and K had on runoff water content of those nutrients. Future studies including measurements of these nutrients in runoff and soils could provide a greater understanding of the processes at play.
The compost-only treatment undersupplied P and K by about 1.58 g/m2 when comparing the nutrient recommendations to the nutrients supplied by the compost. Despite this, there is no statistical difference between the amounts of P in runoff from this treatment and the organic fertilizer treatment or the compost + organic fertilizer treatment, or between the K in runoff in the two compost treatments. This underapplication of P and K also had no consistent effect on crop yields (data not shown). Other studies that measured nutrient leaching from systems using different amounts of compost did observe a dose response to the fertilizer (Shrestha et al. 2020; Small et al. 2019). The amounts of nutrients supplied by compost and possibly the amount of compost used in both studies was, however, greater than in our study. Unlike our study, the amount of P supplied by Shrestha et al. (2020) was also greater than that supplied by the synthetic fertilizer treatment in that study, which corresponded to greater P leaching as well. Small et al. (2019) also found that manure-based composts had a greater increase than municipal compost.
Sampling time.
Although some studies have found a relationship between rainfall and the values of various water-quality metrics (Hoskins et al. 2014; Teemusk and Mander 2007), this was not observed in our study. Correlations between pH (R2 = 0.0304, P = 0.0001), conductivity (R2 = 0.0041, P = 0.1542), color (R2 = 0.0037, P = 0.1841), turbidity (R2 = 0.0003, P = 0.7127),
For the other variables, peaks in the measured variables do not always correspond to peaks in precipitation. There are numerous examples of this mismatch. The 145- and 286-DAC samples, for example, have similar amounts of precipitation between samplings to the 281-DAC sample, but did not result in high color or turbidity readings (Figs. 2B, 4C, and 5C). Similarly, sampling times with equal or greater precipitation between samplings (e.g., 55 and 131 DAC) (Fig. 2B) do not have spikes in
Another trend in differences among sampling times seems to relate to samples taken during the growing season and samples taken closer to compost or fertilizer application. Conductivity in the first study year is greater and more variable during the growing season, but did not exhibit a relationship to compost or fertilizer application. Several sample timings with greater average conductivity took place long after fertilizers were applied; the 131- and 152-DAC samples were taken 25 and 19 d, respectively, after the previous fertilizer application (Table 3). In the second year of the study, samples taken soon after fertilizer application also had relatively low conductivity; the 55- and 105-DAC samples were taken 1 and 6 d, respectively, after the previous fertilizer applications (Table 3). The link between color and turbidity, and the addition of compost and fertilizers appears to be stronger. High runoff color just after compost addition (4 d), with a general trend of decreasing color until just before compost reapplication, suggests that materials being flushed out of the system over time are affecting color. The lack of effect of compost treatment suggests the source is not compost alone. There are a couple of peaks that could also be associated with fertilizer addition, at 55 and 84 DAC (Fig. 5B), which were 1 and 2 d, respectively, after the previous fertilizer application (Table 3). A similar pattern was seen in turbidity, with greater values early in the study year (Fig. 6B). Other studies have reported a relationship between turbidity and temperatures, with greater turbidities found during the winter months when temperatures are lower (Islam et al. 2006; Liu et al. 2007).
General trends were observed in nutrient runoff, with
Ammonium exhibited a different pattern, with greater levels after the growing season (Fig. 10A–D). Ammonium levels in runoff water exceeded US Environmental Protection Agency quality standards for acute (χ2 = 14.532, P = 0.001), but not chronic (χ2 = 0.018, P = 0.894) exposures more during the winter months than the growing season (US Environmental Protection Agency 2013). This greater
Growing system.
For all water-quality metrics for which there was a significant difference between the two growing systems, the raised beds had greater values than the containers. For many water-quality metrics, this difference was present in only one fertilizer treatment or study year. These differences also did not always translate to meaningful differences in water quality. For pH, both averages were close to neutral and within US Environmental Protection Agency water-quality standards for freshwater (US Environmental Protection Agency 2022a) and human health (US Environmental Protection Agency 2022b). When examining the total range of values measured, the container system had a significantly greater proportion of samples (χ2 = 8.971, P = 0.003) that did not meet US Environmental Protection Agency guidelines limiting pH between 6.5 and 9 (US Environmental Protection Agency 2022a). Eight samples read less than 6.5 and three read more than 9, but no samples from the raised beds read outside these thresholds (Fig. 3C). The range of values from both systems for conductivity (Fig. 4C), color (Fig. 5C), and turbidity (Fig. 6C) was similar. There was also no significant difference between the two growing systems in terms of the number of samples that exceeded the US Environmental Protection Agency 75-Pt/Co color threshold (χ2 = 0.307, P = 0.579) for the domestic water supply (US Environmental Protection Agency 1986).
The raised beds had a greater mean runoff of
Conclusion
The impacts of nutrient management systems in urban agriculture have important implications for the environmental sustainability of urban agricultural practices and urban centers. This will be especially important given the current increase in urban agriculture. During this study, compost fertilizer was also found to have fewer peaks in
Expansion of urban agriculture will likely include novel production systems, including low-cost found-object containers such as the kiddie pool containers used in our study. The kiddie pool containers used in our study did not create substantially different runoff water quality than the raised beds, suggesting they are a good candidate for continued use in urban agriculture and for future research. One area for future research on kiddie pool containers suggested by the results of our study is drainage: how to create adequate drainage to minimize waterlogging but still prevent excessive water loss.
The high levels of
References Cited
Abdulkadir A, Leffelaar PA, Agbenin JO, Giller KE. 2013. Nutrient flows and balances in urban and peri-urban agroecosystems of Kano, Nigeria. Nutr Cycl Agroecosyst. 95:231–254. https://doi.org/10.1007/s10705-013-9560-2.
Ackerman K, Conard M, Culligan P, Plunz R, Sutto MP, Whittinghill L. 2014. Sustainable food systems for future cities: The potential of urban agriculture. Econ Soc Rev. 45(2):189–206.
Adugna G. 2018. A review on impact of compost on soil properties, water use, and crop productivity. Acad Res J Agric Sci Res. 4(3):93–104. https://doi.org/10.1016/j.jenvman.2020.110209.
Anderson DM, Gilbert PM, Burkholder JM. 2002. Harmful algal blooms and eutrophication: Nutrientsources, composition, and consequences. Estuaries. 25(4):704–726. https://doi.org/10.1007/BF02804901.
Angotti T. 2015. Urban agriculture: Long-term strategy or impossible dream? Lessons from Prospect Farm in Brooklyn, New York. Public Health. 129(4):336–341. https://doi.org/10.1016/j.puhe.2014.12.008.
Bachman M, Inamdar S, Barton S, Duke JM, Tallamy D, Bruck J. 2016. A comparative assessment of runoff nitrogen from turf, forest, meadow, and mixed landuse watersheds. J Am Water Resour Assoc. 52(2):397–408. https://doi.org/10.1111/1752-1688.12395.
Bahamonde A. 2019. Mental health through the art of gardening. J Therap Hortic. 29(2):27–44.
Barbosa GL, Almeida Gadelha FD, Kublik N, Proctor A, Reichelm L, Weissinger E, Gregory MW, Halden RU. 2015. Comparison of land, water, and energy requirements of lettuce grown using hydroponic vs. conventional agricultural methods. Int J Environ Res Public Health. 12(6):6879–6891. https://doi.org/10.3390/ijerph120606879.
Beck DA, Johnson GR, Spolek GA. 2011. Amending greenroof soil with biochar to affect runoff water quantity and quality. Environ Pollut. 159(8–9):2111–2118. https://doi.org/10.1016/j.envpol.2011.01.022.
Bennett LE, Drikas M. 1993. The evaluation of color in natural waters. Water Res. 27(7):1209–1218.
Beochat CL, Santos JAG, Accioly AMA. 2013. Net mineralization nitrogen and soil chemical changes with application of organic wastes with ‘Fermented Bokashi Compost’. Acta Sci. 35(2):257–264. https://doi.org/10.4025/actasciagron.v35i2.15133.
Berghage RD, Beattie D, Jarrett AR, Thuring C, Razaei F, O’Connor T, O’Connor TP. 2009. Green roofs for stormwater runoff control. US Environmental Protection Agency, Washington, DC, EP A/600/R-09/026.
Berka C, Schreier H, Hall K. 2001. Linking water quality with agricultural intensification in a rural watershed. Water Air Soil Pollut. 127(1–4):389–401. https://doi.org/10.1023/A:1005233005364.
Berndtsson JC, Bengtsson L, Jinno K. 2009. Runoff water quality from intensive and extensive vegetated roofs. Ecol Eng. 35(3):369–380. https://doi.org/10.1016/j.ecoleng.2008.09.020.
Bowden C, Spargp J, Evanylo G. 2007. Mineralization and N fertilizer equivalent value of composts as assessed by tall fescue (Festuca arundinacea). Compost Sci Util. 15(2):111–118. https://doi.org/10.1080/1065657X.2007.10702320.
Broschat TK. 1995. Nitrate, phosphate, and potassium leaching from container-grown plants fertilized by several methods. HortScience. 30(1):74–77. https://doi.org/10.21273/HORTSCI.30.1.74.
Cameira MR, Tedesco S, Leitão TE. 2014. Water and nitrogen budgets under different production systems in Lisbon urban farming. Biosyst Eng. 125:65–79. https://doi.org/10.1016/j.biosystemseng.2014.06.020.
Cañedo-Argüelles M, Hawkins CP, Kefford BJ, Schäfer RB, Dyack BJ, Brucet S, Buchwalter D, Dunlop J, Frör O, Lazorchak J, Coring E, Fernandez HR, Goodfellow W, Achem ALG, Hatfield-Dodds S, Karimov BK, Mensah P, Olson JR, Piscart C, Prat N, Ponsá S, Schulz CJ, Timpano AJ. 2016. Saving freshwater from salts. Science. 351(6276):914–916.
Chadwick DR, Chen S. 2002. Manures, p 57–82. In: Haygarth PM, Jarvis SC (eds). Agriculture, hydrology and water quality. CABI Publishing, Wallingford, UK. 10.1079/9780851995458.0057.
Collier KJ, Ball OJ, Graesser AK, Main MR, Winterbourn MJ. 1990. Do organic and anthropogenic acidity have similar effects on aquatic fauna? Oikos. 59:33–38. https://doi.org/10.2307/3545119.
Colman A. 2017. Community and home gardening develop lifelong healthy habits. Parks Recreat. 52(8):34–35.
Conley DJ, Paerl HW, Howarth RW, Boesch DF, Seitzinger SP, Havens KE, Lancelot C, Likens GE. 2009. Controlling eutrophication: Nitrogen and phosphorus. Science. 323(5917):1014–1015. https://doi.org/10.1126/science.1167755.
Correll DL. 1998. The role of phosphorus in the eutrophication of receiving waters: A review. J Environ Qual. 27(2):261–266. https://doi.org/10.2134/jeq1998.00472425002700020004x.
Czemiel Berndtsson J. 2010. Green roof performance towards management of runoff water quantity and quality: A review. Ecol Eng. 36(4):351–360. https://doi.org/10.1016/j.ecoleng.2009.12.014.
Davies-Colley RJ, Smith DG. 2001. Turbidity, suspended sediment, and water clarity: A review. J Am Water Resour Assoc. 37:1085–1101. https://doi.org/10.1111/j.1752-1688.2001.tb03624.x.
de Barros Sylvestre T, Braos LB, Filho FB, da Cruz MCP, Ferreira ME. 2019. Mineral nitrogen fertilization effects on lettuce crop yield and nitrogen leaching. Sci Hortic. 255:153–160. https://doi.org/10.1016/j.scienta.2019.05.032.
Dewaelheyns V, Elsen A, Vandendriessche H, Gulinck H. 2013. Garden management and soil fertility in Flemish domestic gardens. Landsc Urban Plan. 116:25–35. https://doi.org/10.1016/j.landurbplan.2013.03.010.
Durham R, Lee B, Osborne A. 2018. Gardening in small spaces. University of Kentucky Cooperative Extension Service ID 248.
Elrashidi MA, Mays MD, Fares A, Seybold CA, Harder JL, Peaslee SD, VanNeste P. 2005. Loss of nitrate-nitrogen by runoff and leaching for agricultural watersheds. Soil Sci. 170(12):969–984.
Emilsson T, Berndtsson JC, Mattsson JE, Rolf K. 2007. Effect of using conventional and controlled release fertilizer on nutrient runoff from various vegetated roof systems. Ecol Eng. 29(3):260–271. https://doi.org/10.1016/J.ECOLENG.2006.01.001.
Figueroa-Pico J, Carpio AJ, Tortosa FS. 2020. Turbidity: A key factor in the estimation of fish species richness and abundance in the rocky reefs of Ecuador. Ecol Indic. 111:106021. https://doi.org/10.1016/j.ecolind.2019.106021.
Freda J. 1987. The influence of acidic pond water on amphibians: A review, p 439–450. In: Martin HC (ed). Acidic precipitation. Springer, Dordrecht, The Netherlands.
Gaudreau JE, Vietor DM, White RH, Provin TL, Munster CL. 2002. Response of turf and quality of water runoff to manure and fertilizer. J Environ Qual. 31(4):1316–1322. https://doi.org/10.2134/jeq2002.1316.
Gensemer RW, Gondek JC, Rodriquez PH, Arbildua JJ, Stubblefield WA, Cardwell AS, Santore RC, Ryan AC, Adams WJ, Nordheim E. 2018. Evaluating the effects of pH, hardness, and dissolved organic carbon on the toxicity of aluminum to freshwater aquatic organisms under circumneutral conditions. Environ Toxicol Chem. 37:49–60. https://doi.org/10.1002/etc.3920.
Getter KL, Rowe DB. 2006. The role of extensive green roofs in sustainable development. HortScience. 41(5):1276–1285. https://doi.org/10.21273/HORTSCI.41.5.1276.
Ghose R, Pettygrove M. 2014. Urban community gardens as spaces of citizenship. Antipode. 46(4):1092–1112. https://doi.org/10.1111/anti.12077.
Giannakis GV, Kourgialas NN, Paranychianakis NV, Nikolaidis NP, Kalogerakis N. 2014. Effects of municap solid waste compost on soil properties and vegetables growth. Compost Sci Util. 22:116–131. https://doi.org/10.1016/j.chemosphere.2011.08.025.
Gilley J, Eghball B. 2002. Residual effects of compost and fertilizer applications on nutrients in runoff. Trans ASAE. 45(6):1905–1910. https://doi.org/10.13031/2013.11441.
Goldstein B, Hauschild M, Fernandez J, Birkved M. 2016. Testing the environmental performance of urban agriculture as a food supply in northern climates. J Clean Prod. 135:984–994. https://doi.org/10.1016/j.jclepro.2016.07.004.
Gunia A. 2020. How coronavirus is exposing the world’s fragile food supply chain—and could leave millions hungry. TIME Magazine, 1–8. https://time.com/5820381/coronavirus-food-shortages-hunger/. [accessed 24 May 2023 ].
Hale RL, Turnbull L, Earl SR, Childers DL, Grimm NB. 2015. Stormwater infrastructure controls runoff and dissolved material export from arid urban watersheds. Ecosystems (NY). 18(1):62–75. https://doi.org/10.1007/s10021-014-9812-2.
Hart MR, Quin BF, Nguyen ML. 2004. Phosphorus runoff from agricultural land and direct fertilizer effects: A review. J Environ Qual. 33(6):1954–1972. https://doi.org/10.2134/jeq2004.1954.
Heathwaite AL, Griffiths P, Parkinson RJ. 1998. Nitrogen and phosphorus in runoff from grassland with buffer strips following application of fertilizers and manures. Soil Use Manage. 14(3):142–148. https://doi.org/10.1111/j.1475-2743.1998.tb00140.x.
Hell’s Kitchen Farm Project. 2020. https://www.hkfp.org/. [accessed 3 Nov 2021].
Hoskins TC, Owens JS, Felds JS, Altland JE, Easton ZM, Niemiera AX. 2014. Solute transport through a pine bark-based substrate under saturated and unsaturated conditions. J Am Soc Hortic Sci. 139(6):634–641. https://doi.org/10.21273/JASHS.139.6.634.
Huang B, Shi X, Yu D, Öborn I, Blombäck K, Pagella TF, Wang H, Sun W, Sinclair FJ. 2006. Environmental assessment of small-scale vegetable farming systems in peri-urban areas of the Yangtze River Delta region, China. Agric Ecosyst Environ. 112:391–402. https://doi.org/10.1016/j.agee.2005.08.037.
Hummert International. 2021. Round containers. https://www.hummert.com/containers-round/. [accessed 3 Nov 2021].
Islam MS, Ueda H, Tanaka M. 2006. Spatial and seasonal variations in copepod communities related to turbidity maximum along the Chikugo estuarine gradient in the upper Ariake Bay, Japan. Estuar Coast Shelf Sci. 68(1–2):113–126. https://doi.org/10.1016/j.ecss.2006.02.002.
Jackson LE, Stivers LJ, Warden BT, Tanji KK. 1994. Crop nitrogen utilization and soil nitrate loss in a lettuce field. Fert Res. 37:93–105. https://doi.org/10.1007/BF00748550.
Jackson BE, Wright RD, Seiler JR. 2009. Changes in chemical and physical properties of pine tree substrate and pine bark during long-term nursery crop production. HortScience. 44(3):791–799. https://doi.org/10.21273/HORTSCI.44.3.791.
Janke BD, Finlay JC, Hobbie SE. 2017. Trees and streets as drivers of urban stormwater nutrient pollution. Environ Sci Technol. 51(17):9569–9579. https://doi.org/10.1021/acs.est.7b02225.
King KW, Torbert HA. 2007. Nitrate and ammonium losses from surface-applied organic and inorganic fertilizers. J Agric Sci. 145(4):385–393. https://doi.org/10.1017/S0021859607006946.
Kleinman PJ, Sharpley AN, McDowell RW, Flaten DN, Buda AR, Tao L, Bergstrom L, Zhu Q. 2011. Managing agricultural phosphorus for water quality protection: Principles for progress. Plant Soil. 349(1–2):169–182. https://doi.org/10.1007/s11104-011-0832-9.
Knizhnik HL. 2012. The environmental benefits of urban agriculture on unused, impermeable and semi-permeable spaces in major cities with a focus on Philadelphia, PA. (Masters thesis). University of Pennsylvania, Philadelphia, PA, USA.
Kok KH, Sidek LM, Abidin MRZ, Basri H, Muda ZC, Beddu S. 2013. Evaluation of green roof as green technology for urban stormwater quantity and quality controls. OP conference series: Earth and environmental science. IOP Conf Ser: Earth Envrion Sci. 16:012045. https://doi.org/10.1088/1755-1315/16/1/012045.
Kramer SB, Reganold JP, Glover JD, Bohannan BJ, Mooney HA. 2006. Reduced nitrate leaching and enhanced denitrifier activity and efficiency in organically fertilized soils. Proc Natl Acad Sci USA. 103(12):4522–4527. https://doi.org/10.1073/pnas.0600359103.
Kuta S. 2021. These pandemic food shortages caught everyone by surprise: Here’s how they happened. https://www.foodandwine.com/news/pandemic-food-shortages. [accessed 2 Nov 21].
Li D, Wan J, Ma Y, Wang Y, Huang M, Chen Y. 2015. Stormwater runoff pollutant loading distributions and their correlation with rainfall and catchment characteristics in a rapidly industrialized city. PLoS One. 10(3):e0118776. https://doi.org/10.1371/journal.pone.0118776.
Lin BB, Philpott SM, Jha S. 2015. The future of urban agriculture and biodiversity-ecosystem services: Challenges and next steps. Basic Appl Ecol. 16(3):189–201. https://doi.org/10.1016/j.baae.2015.01.005.
Liu W, Liu Y, Mannaerts CM, Wu G. 2007. Monitoring variation of water turbidity and related environmental factors in Poyang Lake National Nature Reserve, China. Proc. SPIE 6754, Geoinformatics 2007. Geospatial Information Technology and Applications. 67541H. https://doi.org/10.1117/12.764879.
Liu R, Wang J, Shi J, Chen Y, Sun C, Zhang P, Shen Z. 2014. Runoff characteristics and nutrient loss mechanism from plain farmland under simulated rainfall conditions. Sci Total Environ. 468:1069–1077. https://doi.org/10.1016/j.scitotenv.2013.09.035.
Lowe’s. 2021. Lawn and garden: Plastic pots and planters. https://www.lowes.com/pl/Plastic–Pots-planters-Planters-stands-window-boxes-Plants-planters-Lawn-garden/4294612569?refinement=4294965728. [accessed 3 Nov 2021].
Lusk MG, Toor GS, Inglett PW. 2020. Organic nitrogen in residential stormwater runoff: Implications for stormwater management in urban watersheds. Sci Total Environ. 707:135962. https://doi.org/10.1016/j.scitotenv.2019.135962.
Maltais-Landry G, Scow K, Brennan E, Torbert E, Vitousek P. 2016. Higher flexibility in input N:P ratios results in more balanced phosphorus budgets in two long-term experimental agroecosystems. Agric Ecosyst Environ. 223:197–210. https://doi.org/10.1016/j.agee.2016.03.007.
Matlock JM, Rowe DB. 2017. Does compost selection impact green roof substrate performance? Measuring physical properties, plant development, and runoff water quality. Compost Sci Util. 25(4):231–241. https://doi.org/10.1080/1065657X.2017.1295887.
McClintock N, Cooper J, Khandeshi S. 2013. Assessing the potential contribution of vacant land to urban vegetable production and consumption in Oakland, California. Landsc Urban Plan. 111:46–58. https://doi.org/10.1016/j.landurbplan.2012.12.009.
McDougall R, Kristiansen P, Rader R. 2019. Small-scale urban agriculture results in high yields but requires judicious management of inputs to achieve sustainability. Proc Natl Acad Sci USA. 116(1):129–134. https://doi.org/10.1073/pnas.1809707115.
McDougall R, Rader R, Kristiansen P. 2020. Urban agriculture could provide 15% of food supply to Sydney, Australia, under expanded land use scenarios. Land Use Policy. 94:104554. https://doi.org/10.1016/j.landusepol.2020.104554.
McGril S. 2021. Food shortages, pandemic supply chain disruptions create challenges for schools. NBC Universal Media. https://www.nbcconnecticut.com/news/local/food-shortages-pandemic-supply-chain-disruptions-create-challenges-for-schools/2593710/. [accessed 2 Nov 2021].
McLeod RV, Hegg RO. 1984. Pasture runoff water quality from application of inorganic and organic nitrogen sources. J Environ Qual. 13(1):122–126. https://doi.org/10.2134/jeq1984.00472425001300010022x.
Meenar M, Hoover BM. 2012. Community food security via urban agriculture: Understanding people, place, economy, and accessibility from a food justice perspective. J Agric Food Syst Community Dev. 3(1):143–160. https://doi.org/10.5304/jafscd.2012.031.013.
Metson GS, Bennett EM. 2015. Phosphorus cycling in Montreal’s food and urban agriculture systems. PLoS One. 10(3):e0120726. https://doi.org/10.1371/journal.pone.0120726.
Michaels K. 2021. How to make a garden planter from a plastic kiddie pool. https://www.thespruce.com/make-a-kiddie-pool-into-a-garden-planter-848239. [accessed 24 Nov 2021].
Mikkelsen R, Hartz TK. 2008. Nitrogen sources of organic crop production. Better Crops Plant Food. 92(4):16–19.
Mitchell ME, Matter SF, Durtsche RD, Buffam I. 2017. Elevated phosphorus: Dynamics during four years of green roof development. Urban Ecosyst. 20(5):1121–1133. https://doi.org/10.1007/s11252-017-0664-3.
Mok HF, Williamson VG, Grove JR, Burry K, Barker SF, Hamilton AJ. 2014. Strawberry fields forever? Urban agriculture in developed countries: A review. Agron Sustain Dev. 34(1):21–43. https://doi.org/10.1007/s13593-013-0156-7.
Monroe L. 2015. Horticulture therapy improves the body, mind and spirit. J Therap Hortic. 25(2):33–39.
National Oceanic and Atmospheric Administration National Climate Data Center. 2020. Normals daily station details Frankfort Capital City Airport, KY US. http://www.ncdc.noaa.gov/cdo-web/datasets. [accessed 17 Jun 2020].
Nogeire-McRae T, Ryan EP, Jablonski BBR, Carolan M, Arathi HS, Brown CS, Saki HH, McKeen S, Lapansky E, Schipanski ME. 2018. The role of urban agriculture in a secure, healthy, and sustainable food system. Bioscience. 68(10):748–759. https://doi.org/10.1093/biosci/biy071.
Nugent R. 2002. The impact of urban agriculture on the household and local economies, p 67–97. In: Bakker N, Dubbeling M, Gündel S, Sabel-Koschella U, de Zeeuw, H (eds). Growing cities, growing food: Urban agriculture on the policy agenda: A reader on urban agriculture. Deutsche Stiftung fur Internationale Entwicklung, Bonn, Germany.
Ohly H, Gentry S, Wigglesworth R, Bethel A, Lovell R, Garside R. 2016. A systematic review of the health and well-being impacts of school gardening: Synthesis of quantitative and qualitative evidence. BMC Public Health. 16(1):1–36. https://doi.org/10.1186/s12889-016-2941-0.
Opitz I, Berges R, Piorr A, Krikser T. 2016. Contributing to food security in urban areas: Differences between urban agriculture and peri-urban agriculture in the Global North. Agric Human Values. 33(2):341–358. https://doi.org/10.1007/s10460-015-9610-2.
Pershey NA, Cregg BM, Andresen JA, Fernandez RT. 2015. Irrigating based on daily water use reduces nursery runoff volume and nutrient load without reducing growth of four conifers. HortScience. 50(10):1553–1561. https://doi.org/10.21273/HORTSCI.50.10.1553.
Pinterest. 2021. How to turn a plastic kiddie pool into a garden planter. https://www.pinterest.com/pin/409475791089553063/. [accessed 24 Nov 2021].
Rameshkumar S, Radhakrishnan K, Aanand S, Rajaram R. 2019. Influence of physicochemical water quality on aquatic macrophyte diversity in seasonal wetlands. Appl Water Sci. 9:12. https://doi.org/10.1007/s13201-018-0888-2.
Sader M. 2017. Turbidity measurement: A simple, effective indicator of water quality change. OTT Hydromet. http://aqualab.com.au/wp-content/uploads/2017/10/Application-Note-Turbidity-AUS.pdf. [accessed May 24, 2023].
Salomon MJ, Watts-Williams SJ, McLaughlin MJ, Cavagnaro TR. 2020. Urban soil health: A city-wide survey of chemical and biological properties of urban agriculture soils. J Clean Prod. 275:122900. https://doi.org/10.1016/j.jclepro.2020.122900.
Shetty NH, Hu R, Mailloux BJ, Hsueh DY, McGillis WR, Wang M, Chandran K, Culligan PJ. 2019. Studying the effect of bioswales on nutrient pollution in urban combined sewer systems. Sci Total Environ. 665:944–958. https://doi.org/10.1016/j.scitotenv.2019.02.121.
Shrestha P, Small GE, Kay A. 2020. Quantifying nutrient recovery efficiency and loss from compost-based urban agriculture. PLoS One. 15(4):e0230996. https://doi.org/10.1371/journal.pone.0230996.
Shuman LM. 2002. Phosphorus and nitrate nitrogen in runoff following fertilizer application to turfgrass. J Environ Qual. 31(5):1710–1715. https://doi.org/10.2134/jeq2002.1710.
Small GE, Osborne S, Shrestha P, Kay A. 2019. Measuring the fate of compost-derived phosphorus in native soil below urban gardens. Int J Environ Res Public Health. 16:3998. https://doi.org/10.3390/ijerph16203998.
Smith DR, Owens PR, Leytem AB, Warnemuende EA. 2007. Nutrient losses from manure and fertilizer applications as impacted by time to first runoff event. Environ Pollut. 147(1):131–137. https://doi.org/10.1016/j.envpol.2006.08.021.
Smith VH, Schindler DW. 2009. Eutrophication science: Where do we go from here? Trends Ecol Evol. 24(4):201–207. https://doi.org/10.1016/j.tree.2008.11.009.
Spence PL, Osmond DL, Childres W, Heitman JL, Robarge WP. 2012. Effects of lawn maintenance on nutrient losses via overland flow during natural rainfall events. J Am Water Resour Assoc. 48(5):909–924. https://doi.org/10.1111/j.1752-1688.2012.00658.x.
Teemusk A, Mander U. 2007. Rainwater runoff quantity and quality performance from a greenroof: The effects of short-term events. Ecol Eng. 30(3):271–277. https://doi.org/10.1016/j.ecoleng.2007.01.009.
Toland DC, Haggard BE, Boyer ME. 2012. Evaluation of nutrient concentrations in runoff water from green roofs, conventional roofs, and urban streams. Trans ASABE. 55(1):99–106. https://doi.org/10.13031/2013.41258.
Toor GS, Occhipinti ML, Yang YY, Majcherek T, Haver D, Oki L. 2017. Managing urban runoff in residential neighborhoods: Nitrogen and phosphorus in lawn irrigation driven runoff. PLoS One. 12(6):e0179151. https://doi.org/10.1371/journal.pone.0179151.
US Environmental Protection Agency. 1986. Quality Criteria for Water. EPA 440/5-86-001. https://www.epa.gov/sites/default/files/2018-10/documents/quality-criteria-water-1986.pdf. [accessed 9 Mar 2022].
US Environmental Protection Agency. 2013. Aquatic life ambient water quality criteria for ammonia-freshwater. https://www.epa.gov/sites/default/files/2015-08/documents/aquatic-life-ambient-water-quality-criteria-for-ammonia-freshwater-2013.pdf. [accessed 10 Mar 2022].
US Environmental Protection Agency. 2015. United States Environmental Protection Agency (EPA) national pollutant discharge elimination system (NPSES) multi-sector general permit for stormwater discharges associated with industrial activity (MSGP). https://www.epa.gov/sites/default/files/2015-10/documents/msgp2015_finalpermit.pdf. [accessed 9 Mar 2022].
US Environmental Protection Agency. 2022a. National recommended water quality criteria: Aquatic life criteria table. https://www.epa.gov/wqc/national-recommended-water-quality-criteria-aquatic-life-criteria-table. [accessed 9 Mar 2022].
US Environmental Protection Agency. 2022b. National recommended water quality criteria: Human health criteria table. https://www.epa.gov/wqc/national-recommended-water-quality-criteria-human-health-criteria-table. [accessed 9 Mar 2022].
van Averbeke W. 2007. Urban farming in the informal settlements of Atteridgeville, Pretoria, South Africa. Water SA. 33(3):337–342. https://doi.org/10.4314/wsa.v33i3.49112.
Van Tuijl E, Hospers GJ, Van Den Berg L. 2018. Opportunities and challenges of urban agriculture for sustainable city development. Eur Spatial Res Policy. 25(2):5–22. https://doi.org/10.18778/1231-1952.25.2.01.
Whittinghill LJ, Hsueh D, Culligan P, Plunz R. 2016. Stormwater performance of a full scale rooftop farm: Runoff water quality. Ecol Eng. 91:195–206. https://doi.org/10.1016/j.ecoleng.2016.01.047.
Whittinghill LJ, Sarr S. 2021. Sustainable urban agriculture: A case study of Louisville, Kentucky’s largest city. Urban Science. 5:92. https://doi.org/10.3390/urbansci5040092.
Wielemaker R, Oenema O, Zeeman G, Weijma J. 2019. Fertile cities: Nutrient management practices in urban agriculture. Sci Total Environ. 668:1277–1288. https://doi.org/10.1016/j.scitotenv.2019.02.424.
Wielemaker RC, Weijma J, Zeemana G. 2018. Harvest to harvest: Recovering nutrients with new sanitation systems for reuse in urban agriculture. Resour Conserv Recycling. 128:426–437. https://doi.org/10.1016/j.resconrec.2016.09.015.
Witzling L, Wander M, Phillips E. 2011. Testing and educating on urban soil lead: A case of Chicago community gardens. J Agric Food Syst Community Dev. 1(2):167–186. https://doi.org/10.5304/jafscd.2010.012.015.
Yadzi MN, Sample DJ, Scott D, Owen JS, Ketabchy M, Alamdari N. 2019. Water quality characterization of storm and irrigation runoff from a container nursery. Sci Total Environ. 667:166–178. https://doi.org/10.1016/j.scitotenv.2019.02.326.
Yang J, Yang Z, Zou J. 2012. Effects of rainfall and fertilizer types on nitrogen and phosphorus concentrations in surface runoff from subtropical tea fields in Zhejiang, China. Nutr Cycl Agroecosyst. 93(3):297–307. https://doi.org/10.1007/s10705-012-9517-x.
Supplemental Table S1. Analysis of variance results for the interactions of growing system, nutrient management treatment, sampling time, and study year for pH, conductivity, and color.
Supplemental Table S2. Analysis of variance results for the interactions of growing system, nutrient management treatment, sampling time, and study year for nitrate- and ammonia-nitrogen, total phosphorus, and potassium.