Nutrient Evaluation for Sod Production within the Lake Okeechobee Watershed Open Access

Florida has the most land area in turfgrass sod production in the USA. Demand for sod continues as Florida’s population has increased to be the third highest in the country in 2022. This study documents fertilizer inputs on sod farms in the Lake Okeechobee watershed, and specifically nitrogen (N) and phosphorous (P) exports via sod harvesting. Since the land occupied by those sod farms is within the Lake Okeechobee watershed basin, reducing N and P fertilizer inputs and increasing the removal of those nutrients through sod, could have an ultimate net positive effect of lessening the nutrient load into Lake Okeechobee and subsequent discharges into hydrologically-linked ecosystems such as the Everglades. In addition, the discharge of nutrient-rich water into coastal waters from Lake Okeechobee has been implicated in influencing Red Tide occurrence (i.e., harmful algal blooms). Thus, demonstrated reductions in fertilizer inputs and increased N and P exports away from fragile ecosystems could support the sod production industry’s contribution to natural resource conservation and sustainable land management practices.

Lake Okeechobee is the largest freshwater lake in Florida, and is the eighth largest freshwater lake in the U.S. Lake Okeechobee, also known as Florida’s “inland sea”, covers 1,900 square kilometers (730 square miles), but actually is a shallow lake with an average depth of 2.7 m (9 ft). The Lake Okeechobee Protection Act requires that proposed changes in land use will not result in increased phosphorous (P) loading over that of existing land uses (SFWMD, 2007).

In recent years, interest in converting farm or fallow land to commercial turfgrass sod production has increased in the Lake Okeechobee watershed, located in south Florida (personal communication, Turfgrass Producers of Florida; https://floridaturf.com/). Based on previous studies, sod farms in Florida are net exporters of P from the ecosystem (Graetz et al. 1991, 2002). Those studies determined that P is exported from the sod farm in both plant tissue and in soil contained within each harvested sod piece (Graetz et al. 1991, 2002). For example, a minimum of 25 kg P per ha (21 lb P per acre) was exported from bahiagrass (Paspalum notatum Flüggé) harvested from a sand/muck soil (Graetz et al. 1991, 2002). Two assumptions are in this value: (1) 70% of the sod field is harvested, with the remaining 30% left for propagation to reestablish the turfgrass stand, and (2) once established the production cycle is 10.5 months yielding 1.14 harvests per year (Graetz et al. 1991, 2002). Those reports further stated the information sampling and data compiling methods were limited, although currently these parameters are used by the South Florida Water Management District (West Palm Beach, FL) regulatory staff to evaluate the P load change related to land use change.

Furthermore, a total maximum daily load (TMDL) for nitrogen (N) in the tributaries of Lake Okeechobee has been developed (SFWMD, 2007) which has increased the level of interest of N concentration in land for the surrounding counties within the Lake Okeechobee watershed basin. However, no study has analyzed either P or N soil pools and processes within sod farms in that basin even though sod farming practices and sod farming land use areas have been adapted or changed (Cisar et al. 1992, Satterwaite et al. 2009). Therefore, the objective of this study was to reevaluate the P and N status of sod farms in the Lake Okeechobee watershed and to determine if those sod farms are net exporters of P and N from the ecosystem.

A sod farmer survey was conducted in the Lake Okeechobee watershed during January 2009. Survey questions (Table 1) were developed with input from members of the South Florida Water Management District (H. Zhao and R. Boney, West Palm Beach, FL), Florida Department of Agriculture and Consumer Services (Linda Crane; Tallahassee, FL) and the University of Florida’s Ft. Lauderdale Research and Education Center (J. Cisar, M. McMillan, P. McGroary, and G. Snyder; Davie, FL). Twenty sod producers in the Lake Okeechobee watershed were selected at random, contacted, and asked to participate in the “Sod N and P Fertilizer Practice Survey”. Sod farms that agreed to participate in the survey were visited in-person by J. Cisar, G. Snyder, and P. McGroary to expedite the collection of accurate data as well as to inform the sod producers of the context of the survey and the possible future use of the results. All respondents that participated in the survey chose to remain anonymous (i.e., no name attached to the completed survey form). Of note, without being offered anonymity, the sod producers would not have participated in the survey. The survey questions, while not covering all agronomic and economic aspects of sod production, attempted to assess practices utilized by the sod production industry in south Florida that are relevant to the objective of this study.

The N and P soil analyses were conducted at University of Florida’s Belle Glade Research and Education Center (Belle Glade, FL). During February 2009, five sod samples of 90.3 cm² (14 in²) × 2.54 cm (1 in) depth were randomly collected from each turfgrass species grown on each sod farm. All individual soil samples were oven-dried at 60 C (140 F) for four days, oven-dried weight recorded, then ground with a Wiley Mill (Model 110.3, Thomas Scientific; Swedesboro, NJ) to homogenize the sample for laboratory determination of total N and total P. Total N was calculated by adding total Kjedahl N plus nitrate plus nitrite together. Total P was measured by digesting the sample to measure for actual P. The methods used for determination of total N and total P were EPA 351.2, 353.2, and 365.4 (Sparks et al. 1996). All laboratory analysis results were recalculated to account for percent moisture for each soil sample. All N and P data means were calculated from five replications as a completely random experimental design (Mead et al. 2003).

Twenty sod producers were contacted within the Lake Okeechobee watershed, and ten agreed to participate in the survey for a 50% positive response rate. During January and February 2009, those sod farms were visited in-person to inform the sod producers about the purpose and context of the survey, and to expedite the collection of accurate data. All respondents chose to remain anonymous (i.e., no name attached to completed survey form), however, the sod farms that completed the survey were located in Glades, Highlands, Martin, Okeechobee, and Palm Beach counties (Fig. 1).

A total land area of 9,499 ha (23,472 acres) was utilized for sod production among the ten farms that completed the survey (Fig. 2). Bahiagrass (Paspalum notatum Flüggé) was grown on the largest land area at 5,463 ha (13,500 acres), followed-by St. Augustinegrass (Stenotaphrum secundatum (Walt.) Kuntze) on 3,726 ha (9,208 acres), bermudagrass (Cynodon dactylon (L.) Pers. × Cynodon transvaalensis Burtt-Davy) on 188 ha (465 acres), and zoysiagrass (Zoysia japonica Steud. ‘Empire’) on 121 ha (299 acres) (Fig. 2). Of note, bahiagrass was grown on only two sod farms. Bahiagrass represented 58% of total land use for sod production, following by 39% for St. Augustinegrass, 2% for bermudagrass, and 1% for zoysiagrass (Fig. 2).

Among the ten sod farms surveyed representing 9,499 ha (23,472 acres) of sod land use, sod was produced on a total of 5,852 ha (14,461 acres) of sand, and 3,647 ha (9,011 acres) of muck soil (Fig. 3). Thus, 62% of all sod production occurred on sand, and 38% on muck soil (Fig. 3). The sand textural class was attributed to ≥ 85% sand particle size content, of which the majority was the fine sand-size fraction, and typically contained a range of 2 to 8% organic matter on those sod farms. Muck soil is defined as an organic soil (i.e., histosol), and the muck soil on those sod farms typically contained a range of 60 to 90% highly decomposed organic matter that is mostly black in color. Muck soils typically are found in Florida’s Everglades Agricultural Area, and also in the low-lying fields surrounding Lake Placid and Lake Istokpoga in Central Florida (Williams 2008).

Sod farm size was defined as small (0-201 ha; 0-499 acres), medium (202-404 ha; 500-999 acres), large (405-809 ha; 1,000-1,999 acres), and very large (≥ 810 ha; ≥ 2,000 acres). Among the ten sod farms surveyed, 98% of sod production occurred on very large (73%) and large (25%) farms, and the remaining sod produced on small (2%) farms (Table 2). Among the total of 3,726 ha (9,208 acres) of St. Augustinegrass sod produced, 61% was on large farms, 39% very large farms, and < 1% small farms (Table 2). With bahiagrass, 100% of the 5,463 ha (13,500 acres) of sod was produced on very large farms (Table 2). Among the total of 188 ha (465 acres) of bermudagrass sod grown, 68% was on small farms and 32% on large farms (Table 2). With the 121 ha (299 acres) of zoysiagrass sod produced, 56% was on small farms, 26% on large farms, and 18% on very large farms (Table 2).

With St. Augustinegrass, 95% of sod was produced on muck (3,541 ha, 8,750 acres) and 5% produced on sand (185 ha, 458 acres) (Table 2). All bahiagrass sod (5,463 ha, 13,500 acres) was produced on sand (Table 2). With bermudagrass sod, 68% was produced on sand (127 ha, 315 acres) and 32% was produced on muck (61 ha, 150 acres) (Table 2). With zoysiagrass, 63% of sod was produced on sand (76 ha, 188 acres) and 37% on muck (45 ha, 111 acres) (Table 2). Of note, the determination of which turfgrass species is produced on sand or muck is dictated by the soil type that exists at each individual sod farm.

Harvest events per year (i.e., harvest = sod cut and removed) varied by turfgrass species (Table 3). Zoysiagrass and St. Augustinegrass typically is harvested once per year, bermudagrass twice per year, and bahiagrass once every four years (Table 3). Thus, only bermudagrass was capable of two harvests per year, resulting in a six-month production cycle. Zoysiagrass and St. Augustinegrass harvest time frequently depends upon sod market demand and agronomic production practices, but typically results in one harvest per year over a 12-month production cycle. Bahiagrass harvest cycle was much longer at three to four years, thus reflecting the dual use of bahiagrass as both a pasture and as a low resource input turfgrass.

Total harvested land area from highest to lowest was bahiagrass, St. Augustinegrass, bermudagrass, and zoysiagrass (Table 3). Actual percent of land area harvested, however, ranged from 86 to 87% each for zoysiagrass, St. Augustinegrass, and bermudagrass, and 65% for bahiagrass (Table 3). Actual land area harvested was highest for bahiagrass at 3,551 ha (8,775 acres), followed-by St. Augustinegrass at 3,205 ha (7,919 acres), bermudagrass at 164 ha (405 acres), and zoysiagrass at 104 ha (257 acres) (Table 3). Of note, sod producers attributed reduced land areas for bahiagrass harvest to more land needed for reestablishment and replanting, longer production cycles with less agronomic inputs, greater insect pest damage, invasive weed infestations, and submerged or wet areas that could not be harvested due to being grown on muck soils.

Harvested sod thickness was determined by measuring five random pieces of freshly cut sod per site (i.e., field or fields surveyed at each sod farm). The thickness of harvested or cut sod (i.e., soil or root zone depth removed with harvested sod) from all surveyed farms was approximately 2.54 cm (1.0 in), which is a commonly accepted practice with sod production for the purposes of minimizing the weight and optimizing the quality of sod during transport and delivery (Cockerham 2008). Of note, sod farmers in the Lake Okeechobee watershed often will determine or measure thickness-of-cut based on individual cored sod pieces. The actual sod-cutting blade typically measures 6.4 mm (0.25 in) in thickness. Also, with all turfgrass species produced for sod in the Lake Okeechobee watershed, the mowing height ranged from 3.8 to 10.1 cm (1.5 to 4 in) depending on market demand and customer expectations and was typically mowed at those heights-of-cut prior to harvest.

When a hectare (or acre) of warm-season turfgrass sod is harvested, only a portion of that land area is actually cut and removed from the field. Thus, vegetative “strips” or “ribbons” throughout each field are left unharvested to allow for regeneration of the next crop (Cockerham, 2008). For example, among the ten sod farms surveyed, 74% of total harvested land area (6,848 of 9,296 ha; 16,923 of 22,972 acres) for all turfgrass species was cut, removed, and exported from the farm site (Table 3). Recent improvements to maximize sod yield include satellite-guided steering technology that allows for more precise cuts with less waste and leaving narrower ribbons (i.e., narrow strips of turf remaining after harvest to establish into the next crop). Besides the need for vegetative turfgrass strips or ribbons to remain on the harvested field for reestablishment, limits to harvested acreage include excessively wet areas, weed infestations, loss of turfgrass stand density due to insect pests, land area occupied by ditches and roads, and uneven land surfaces and slopes.

All surveyed farms responded affirmatively that they adhere to best management practices and guidelines for sod production, referring to ‘Water Quality/Quantity Best Management Practices for Florida Sod’ (Bartnick et al. 2008) as their resource. From this survey, nine-out-of-ten or 90% of sod farms surveyed had net exports of N (Table 4). One farm producing bahiagrass on sand and zoysiagrass and St. Augustinegrass on muck soils applied more N than exported (Table 4). With P fertilizer, ten-out-of-ten or 100% of sod farms surveyed had net exports of P (Table 5). Within a few sod farms on muck soil, more P was probably applied than needed (Table 5).

The N sources utilized varied by turfgrass species and sod farms, and included ammonium nitrate, ammonium sulfate, diammonium phosphate, monoammonium phosphate, sulfur coated urea, and sewage sludge (Table 4). The most commonly used N source was ammonium sulfate presumably due to low cost and its ability to alleviate high soil pH found in many Florida soils that have appreciable deposits of calcium carbonate (Table 4). For all turfgrass species produced on both sand and muck soil among all sod farms and total land area, the total N applied ranged from 21.8 to 176.2 kg·N ha⁻¹ (48.00 to 388.50 lb·N acre⁻¹) (Table 4). The wide range of N applications reflect the difference between sand and muck (histosol) soils in Florida where it was demonstrated that N fertilizer applications did not negatively affect sod harvest cycles, sod quality, or sod strength on histosols in the Everglades Agricultural Area (Cisar et al. 1992). Conversely, sand root zones retain few nutrients and sod production requires frequent applications of fertilizer including N compared to sod production on muck soils.

Turfgrass cultivar and cropping system also were important factors for N application with zoysiagrass and St. Augustinegrass needing greater N inputs than bahiagrass (Table 4). In certain sod farms, bahiagrass is grown as a forage grass as well as for sod with little N applied if the primary use is as forage grass (Table 4). However, for sod-focused bahiagrass production on sand root zones, the N applications are far greater per sod production cycle (Table 4).

Laboratory analysis revealed total N contained in the soil harvested as sod ranged from 0.000034 to 0.002834 kg N per 0.454 kg harvested sod (0.000075 to 0.006249 lb N per lb of harvested sod) (Table 4). For all ten sod farms, the combined total amount of N removed in harvested sod was 2,315.37 kg·N ha⁻¹ (5,104.52 lb·N acre⁻¹) (Table 4).

The net N exported with harvested sod (i.e., total N removed in the soil through harvested sod minus total N applied as fertilizer) ranged from (15.07 to 167.06 kg·N ha⁻¹ (33.24 to 368.32 lb·N acre⁻¹) (Table 4). For all ten sod farms, the combined total amount of net N exported was 1,085.58 kg·N ha⁻¹ (2,393.31 lb·N acre⁻¹) (Table 4). Although overall more N was exported from all 10 farms, there were three farms that exported less than applied, suggesting an opportunity for greater N management efficiency (Table 4).

The P sources utilized also varied by turfgrass species and sod farms, and included diammonium phosphate, monoammonium phosphate, sewage sludge, and triple superphosphate (Table 5). The most used P source was diammonium phosphate (Table 5). For all turfgrass species produced on sand and muck soil among all sod farms, total P applied ranged from 0 to 52.14 kg·P ha⁻¹ (0 to 114.96 lb·P acre⁻¹) (Table 5). Research on P fertilization for St. Augustinegrass sod production on histosols in south Florida indicated that P applications may not be necessary and should be applied only if soil test indices verify insufficient P levels (Cisar et al. 1992). Some farms did not apply P fertilizer, but the vast majority of sod producers did, thus suggesting opportunities to reduce P fertilizer inputs (Table 5).

Laboratory analysis revealed total P contained in the soil harvested as sod ranged from 0.000029 to 0.000581 kg P per 0.454 kg harvested sod (0.000065 to 0.001282 lb P per lb of harvested sod) (Table 5). For all ten sod farms, the combined total amount of P removed in harvested sod was 611.16 kg·P ha⁻¹ (1,347.38 lb·P acre⁻¹) (Table 5).

The net P exported with harvested sod (i.e., total P removed in the soil through harvested sod minus total P applied as fertilizer) ranged from 7.70 to 48.65 kg·P ha⁻¹ (16.98 to 107.27 lb·P acre⁻¹) (Table 5). For all ten sod farms, the combined total amount of net P exported was 239.76 kg·P ha⁻¹ (528.60 lb·P acre⁻¹) (Table 5). Survey results indicated that five farms apply more P than exported (Table 5).

Of note, N and P content within leaf, stolon, and rhizome tissues of all four turfgrass species were not analyzed from the harvested sod collected during this survey. The N and P concentration of oven-dried plant tissues for the four turfgrass species would range from 1 to 2% N and 0.1 to 0.2% P (Cisar et al. 1992). Therefore, N and P content in plant tissues was considered to be negligible and therefore not included in this survey.

In conclusion, sod production is a major agricultural enterprise in the Lake Okeechobee watershed. The net positive export of N and P recorded from the ten surveyed sod farms in the region suggest that sod production provides a route for removing these two nutrients of impairment from this fragile hydrologically-linked ecosystem. Nevertheless, the observations from some participants having more N and P input than export illustrate the opportunities for extension education to improve N and P fertilization practices and export efficiency for sod producers in Florida. Highlighting scientifically-based research that provides guidelines for lowering fertilizer inputs, basing fertilizer application timings and rates on soil and/or turfgrass tissue testing benchmarks, and using slow-release nutrient sources may reduce the potential for adverse environmental impacts as well as being cost-effective strategies for sod producers in Florida.