«A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Crop, Soil, and Environmental Sciences ...»
a high density CAFO production area (National Agricultural Statistics Service, 2006).
The four counties that compose NWA produce 28% of state poultry production on only 6% of total state land area (National Agricultural Statistics Service, 2006). Applying excessive amounts of PL in small regional areas causes water quality concerns and court action in one watershed was used to limit PL applications and reduce P in overland flow (DeLaune et al., 2006). Water quality concerns in NWA also causes the 205 Mg of wet BS produced daily from the five major municipalities of NWA to be disposed in landfills versus land applied where nutrients would be recycled (Northwest Arkansas Conservation Authority, 2003). Costs of handling organic sources of nutrients are large and inhibit the use of these materials far from the source, exacerbating disposal problems in NWA.
High economic values of the turfgrass industry coupled with a recent surge in demand for organic products may help offset costs of transporting organic waste nutrients over long distances. In Arkansas alone, the sod, golf course and landscape maintenance industry had $327 million in economic output in 2002 (Haydu et al., 2006). Recent trends also indicated that market shares for organic fertilizers outpaced traditional fertilizers by 15% for American homeowners (MarketWatch, 2007). Considerable amounts of PL and BS could be applied to nutrient-deficient soils if even a small percentage of the United State's 12.8 million ha of turfgrass was utilized (Lindsey, 2005). A surge in organic product demand coupled with huge land areas may allow for a synergistic market between consumer's need for fertilizer and society's desire for organic nutrient waste utilization if environmentally safe products are manufactured.
applications as it reduces nutrient and sediment loads in runoff and may export nutrients from a watershed if sold as sod. Vietor and coworkers (2002) indicated that up to 77% of applied P was removed with sod when harvested; thereby lowering P on original application fields. High plant densities of turfgrass also retard nutrient and sediment loss compared to less dense stands of pasturelands, where PL is commonly land applied (Gross et al., 1991; Linde et al., 1995). Overall, properly managed turfgrass was shown to minimize nutrient runoff and serve as a good sink for excess nutrients due to dense stands and natural resistance to water runoff (Gross et al., 1990).
Several researchers indicated that turf applications of organic fertilizers lowered P and N losses in runoff compared to inorganic fertilizers. For example, a study by Gaudreau and coworkers (2002) showed that DRP losses were 44% less with PL compared to inorganic P fertilizers; however, nutrient release was limited and turf quality was compromised when using PL alone. These researchers suggested a co-application of inorganic fertilizers to fully meet turfgrass needs for quality maintenance. Research by Edwards and Daniel (1994) demonstrated that PL had lower runoff DRP and inorganic N concentrations than inorganic fertilizer, attributable to higher soluble nutrient fractions in inorganic fertilizers compared to PL. However, runoff from PL treatments posed a higher water quality threat than inorganic fertilizer in regards to chemical oxygen demand and total suspended solids. McLeod and Hegg (1984) showed that municipal BS produced lower N concentrations in runoff than either PL or inorganic fertilizer on tall fescue (Festuca arundinacea L. Shreb.) pastures.
Agglomeration (granular combination of fine powder materials and liquid binding
the fresh product into a material that may be easier to store, transport and apply than PL and BS in the unprocessed forms (Wolfe et. al., 2002). Additives incorporated during the agglomeration process may further enhance product value and environmental stability.
For instance, binding agents are commonly used in the pelletizing industry to change the product's physical and nutrient release characteristics (Hinkle and Rosenthal, 1991; Tabil et. al., 1997; Veverka and Hinkle, 2001). Inorganic nutrients may be incorporated to increase plant-available N and P for quick plant uptake or to increase overall nutrient analysis to decrease shipping and application costs per unit of nutrient (Hamilton and Sims, 1995). Nitrification inhibitors may be added to reduce N loss during the growing season, especially in systems with abundant macropores and continuous irrigation such as turfgrass (Thomas and Phillips, 1979; Barton and Colmer, 2006). However, the high heat agglomeration process may increase DRP and inorganic N in organic-fertilizers making nutrients more susceptible to losses during rainfall simulations than unprocessed PL and BS, but also more available for plant uptake (Haggard et al., 2005; Ojeda et al., 2006;
Toor et al., 2007).
Numerous research projects have established the potential for nutrient loss by inorganic and organic fertilizer sources subjected to simulated and natural rainfall.
However, limited or no research is available on the environmental integrity of using processed PL and BS fertilizers in combination with additives even though this is a multimillion dollar industry. The objective of our study was to evaluate the effect of three binding agents, a nitrification inhibitor, urea, and dried BS on runoff water quality from PL granular fertilizers. We also wanted to evaluate the risks of using these N-fortified
Rainfall simulations were conducted on a Captina silt loam (Fine-silty, siliceous, active, mesic Typic Fragiudult) at the Arkansas Agricultural Research and Extension Center in Fayetteville (36°4'N; 94°9'W). Plots were established on bermudagrass (Cynodon dactylon L. 'Princess-77') maintained similar to a golf fairway in regards to irrigation, mowing, pesticide applications, and fertility. Plot slopes averaged 4% and background soil characteristics are presented in Table 5.1.
Nitrogen-fortified PL and BS granular fertilizers, BS, Milorganite, fresh PL, and urea + TSP treatments were applied on a TP basis at a rate of 20 kg P ha"1. Resulting rates of inorganic N, organic N, TN, DRP, and total solids applied are presented in Table 5.1.
Fertilizer applications were made by evenly spreading the material over the plot area by hand on the day rainfall simulations were conducted.
Rainfall simulation plots were established according to the National Phosphorus Project Protocol (2008) using a portable rainfall simulator (Humphry et al., 2002). Plots were 1.5 x 2.0 m and used aluminum edging to partition runoff into a flow collector on the downward slope end. Water used to simulate rainfall was passed through a series of exchange resins (cation-anion-cation) to simulate natural rainfall (i.e., low buffering capacity and pH ~4). All water used for simulations was analyzed and determined to have
A non-acidified sub-sample was filtered through 0.45 urn pore filter paper for DRP analysis. Dissolved reactive P was quantified colorimetrically using a spectrophotometer (Hitachi U2000; Hitachi Scientific Instruments, Mountain View, CA 94043) (Method 365.2; USEPA, 2007). An unfiltered sub-sample was acidified using concentrated H2SO4 (1 drop per 10 mL sample = pH ~ 2) for NH4-N and TN analysis.
Ammonium-N was analyzed colorimetrically using flow injection analysis (Quickchem 8500; Lachat Instruments, Loveland, CO 80539) (Method 350.1; USEPA, 2007). Total N was digested using K2S2O8 and NaOH in an autoclave heated to 110°C for 30 min and quantified by flow injection analysis using cadmium reduction (Quickchem 8500) (Method 4500-N C ; Eaton et. al, 2005). The remaining unfiltered and non-acidified runoff sample was analyzed for NO3-N, TP, total dissolved solids, total solids, EC, pH, and turbidity. Nitrate-N was analyzed with ion chromatography (DX100; Dionex Corp., Sunnyvale, CA 94088) (Method 300.0; USEPA, 2007). Total P was digested using concentrated HNO3 and H2SO4 as described by Pote and Daniel (2000) and quantified using a SPECTRO CIROS inductively coupled argon plasma (ICAP) spectrometer (SPECTRO Analytical Institute, Kleve, Germany). Total dissolved solids were determined by filtering 100 mL water through a glass fiber filter (Type A/E; Pall Life Sciences, Ann Arbor, MI 48103) under vacuum and drying filtrate at 98°C until water disappeared and then drying at 180°C for 1 h until a constant weight was obtained (Method 2540 C ; Eaton et. al., 2005). Total solids were found by drying 100 mL
h at 105°C (Method 2540 B.; Eaton et al, 2005). Electrical conductivity and pH were measured immediately after the runoff event concluded with portable pH (Thermo Orion 230; Thermo Fisher Scientific, Inc., Waltham, MA 02454) and EC meters (Orion 130A;
Thermo Fisher Scientific, Inc). Turbidity was measured in Nephelometric turbidity units (NTU) using a turbidimeter (Hach 2100A; Hach Company, Loveland, CO 80539) (Method 180.1; USEPA, 2007).
Soil samples (10 cm depth) were randomly taken outside of plots from each replication prior to rainfall simulations to test background soil properties in Table 5.1.
Soil samples were air dried and sieved through a 2-mm screen to remove non-soil fragments and to reduce aggregates. Water-extractable P was determined by shaking 2 g soil with 20 g 0.01 MCaC^ for 1 h at 200 evolutions per minute (epm), filtering through
0.45um membrane filter, and acidifying to pH~2 by adding two drops concentrated HC1 (Kuo, 1996). Total P was determined with an HCIO4 digestion as described by Kuo (1996). Soil DRP and TP were quantified using a SPECTRO CIROS ICAP. Inorganic N fractions were extracted by shaking 10 g soil with 100 g 2 MKC1 for 1 h at 200 epm (Mulvaney, 1996). Soil solution was filtered through filter paper (Whatman 42; Whatman pic, Middlesex, UK) and analyzed using a Sanplus System continuous flow analyzer (SKALAR Inc., Norcross, GA 30071). Total N was estimated by dry combustion (LECO CN2000; LECO Corp., St. Joseph, MI 49085) (Colombo and Giazzi, 1982). Organic N was determined by subtracting inorganic N fractions from TN. Soil pH and EC were determined on a 1:2 ratio (soil:water, w:w) using double-deionized water (Thomas, 1996).
and BS granular fertilizers, PL, BS, Milorganite, and inorganic fertilizers as previously discussed for soils (Hoskins et al., 2003). Total P concentrations were determined by digesting fertilizers with concentrated HNO3 and H2O2 and quantifying with a SPECTRQ CIROS ICAP (Hoskins et. al., 2003).
Nutrient and solid loads were calculated by multiplying runoff water concentrations (mg L"1) by runoff volume (L ha"1) to achieve kg ha"1. Organic N was found by subtracting inorganic N fractions from TN present in runoff samples. Each fraction of N (NH4-N, NO3-N and organic N), P (DRP) or solids (dissolved) is presented as a percentage of TN, TP or total solids and was calculated by dividing the fraction's load by total N or P load. Percent fertilizer loss was calculated by dividing nutrient loads in runoff water by amounts applied in Table 5.1.
Rainfall simulations were established as a randomized complete block (RCB) design with fertilizer source as a fixed effect. All treatments were replicated 3 times with replication serving as a random effect. Two statistical analyses were conducted. Initially, data was analyzed as a RCB design with each fertilizer treatment serving independently of another (Gomez and Gomez, 1984). The 17 treatments included the 12 N-fortified PL and BS fertilizers, fresh PL, dried municipal BS, Milorganite, urea + TSP, and a nofertilizer control. The purpose of this analysis was to discuss potential benefits and risks of using the 12 N-fortified PL and BS fertilizers compared to unprocessed PL and BS, inorganic fertilizer and the commonly used turf organic fertilizer Milorganite. Secondly,
factorial arrangement of fertilizers with and without BS, with and without DCD, and bound with one of 3 binding agents [lignosulfonate, urea formaldehyde or water] totaling 12 different fertilizer combinations (Gomez and Gomez, 1984). This analysis was primarily used to distinguish impacts of using different fertilizer ingredient combinations on runoff water quality. For both designs, data were analyzed using analysis of variance (PROC GLM) with SAS software (SAS Institute, 2003). Means were separated using Fisher's protected least significant difference tests (LSD) using a significance level of p0.\0 that was established a priori.
Total P runoff water concentrations, loads and percentage TP applied lost were significantly affected by fertilizer sources (Table 5.2). The urea + TSP treatment had higher runoff water concentration, load and percentage of TP applied lost (28.8 mg P L",
4.9 kg P ha"1 and 24.7% TP loss, respectively) than any N-fortified PL and BS granular fertilizer or organic fertilizer treatment (Table 5.2). Edwards and Daniel (1994) showed similar results in their rainfall simulations on a Captina silt loam. In their study, inorganic P fertilizer lost over 40% more TP than PL alone in the initial rainfall following fertilizer application. Biosolids (2.4%) had similar TP loss to no-fertilizer control treatments indicating that very little applied P was dissolved in runoff water (Table 5.2).
Milorganite, a municipal BS derived fertilizer from Milwaukee, WI, had statistically similar TP loss (4.0%) as dried BS (2.4%) used in our granular formulations from Stuttgart, AR. Fresh PL generally had similar TP runoff water concentrations, loads and
fertilizer treatments without BS (Table 5.2). Similar TP losses to unprocessed PL indicate that grinding, heat and pressure from the granulation process did not exacerbate TP runoff risks.