Experimental design.

Twenty-four mixed landscape plots (3.05 m × 3.66 m) were established in a randomized complete block design at the University of Florida–Institute of Food and Agricultural Sciences (IFAS) Gulf Coast Research and Education Center in Wimauma, FL, to simulate new residential landscapes. All vegetation was removed from the site before plot construction. The entire research area was prepared at a 2% grade (as is typically required by construction codes) and compacted (D_b range: 1.7 to 1.9 g·cm⁻³) using a small plate compactor (Wacker Neuson, Munich, Germany). Individual landscape plots were constructed inside water-sealed treated wooden boxes. Within each plot, the compacted field soil (Zolfo fine sand; sandy, siliceous, hyperthermic Oxyaquic Alorthods) (USDA-NRCS, 2004) was then buried under 1.13 m³ of uncompacted soil fill material. Three fill materials were mixed in equal parts to simulate a “top soil” material that would be applied during residential construction. The three fill soil material sources included: a subsoil fill containing construction material and other debris; a clean topsoil material (St. Johns fine sand; sandy, siliceous, hyperthermic Typic Alaquod) obtained from depth of 30 to 60 cm (Hills Dirt Pit, LLC., Riverview, FL), and a clean subsoil fill (St. Johns fine sand) fill obtained from a depth of 122 to 213 cm (Hills Dirt Pit, LLC.).

Composted dairy manure solids (compost; Agrigy, Palm Harbor, FL) were applied as an organic soil amendment at a rate of 508 m³·ha⁻¹ (5-cm depth, equaling ≈256 Mg·ha⁻¹) in combination with two mechanical soil treatments (shallow tillage and aeration) for a total of five soil management treatments: 1) tillage only; 2) compost only; 3) compost + tillage; 4) aeration only; and 5) compost + aeration. Chemical analysis of the compost indicated that the material had a pH of 6.59, an EC of 1.02 dS·m⁻¹, and a total carbon to N (C:N) ratio of 13.6. Plots that received compost applications received total N and EPA 3050 digestible P at a rate of ≈3277 and 1385 kg·ha⁻¹, respectively. In plots receiving the tillage treatment, soil (or soil + compost) was turned to a depth of 10 to 15 cm using a counterrotating tines tiller (Sears Brands, LLC, Hoffman Estates, IL). In plots receiving the aeration treatment, soil aeration plugs were mechanically removed using a core aerator (Billy Goat Industries, Inc., Lee's Summit, MO). An untreated control plot (no tillage or organic amendment) was included as the sixth soil treatment.

Once soil treatments were applied, each plot was split across the contour and 5.58 m² of the plot was planted with Stenotaphrum secundatum (Walter) Kuntze turfgrass; the remaining 5.58 m² was planted with ornamental plants. Ornamental plants species, selected to represent species commonly installed in Florida urban landscapes, included: Galphimia glauca Cav., Rhaphiolepis indica (L.) Lindl. ex Ker Gawl., Ilex cornuta ‘Burfordi’ Lindl. & Paxton, and Liriope muscari (Decne.) L. H. Bailey. In all landscape plots, turfgrass was fertilized at a total N rate of 220 kg·ha⁻¹ based on current University of Florida–IFAS recommendations for South Florida (moderate maintenance) (Sartain, 2007): complete turf fertilizer (26N–0.9P–9.1K; Lesco Professional Turf Fertilizer, Sebring, FL) at an N rate of 48.8 kg·ha⁻¹ per application in February and October, polymer-coated urea (42N–0.0P–0.0K; Harrell's Professional Fertilizer Solutions, Lakeland, FL) as a slow-release N source at and N rate of 48.8 kg·ha⁻¹ per application in May and August, urea (46N–0.0P–0.0K; Potash Corp., Northbrook, IL) as a soluble N source at an N rate of 24.4 kg·ha⁻¹ in April, and 6.34 L·ha⁻¹ of ferrous sulfate (Sunniland Corporation, Sanford, FL) in July. Ornamental plants were fertilized every 3 months with urea (40N–0.0P–0.0K) at an N rate of 24.4 kg·ha⁻¹ per application based on University of Florida–IFAS recommendations for established woody ornamentals grown in the landscape (Knox et al., 2002). The nutrient content of added compost was not considered when fertilizing turfgrass or ornamentals because compost was applied based on horticultural recommendations to improve soil conditions (Urban, 2008) rather than as a nutrient source.

The entire research plot area was equipped with a spray irrigation system, which allowed for individual landscape plots to be irrigated, as needed, based on University of Florida–IFAS recommendations (Zazueta et al., 2005). Plots were watered daily for 30 d after planting to allow for establishment of turf and ornamental plant material. Irrigation frequency was then reduced to 2 d per week based on typical watering restrictions for landscape irrigation that would be mandated in times of drought (South Florida Water Management District, 2010; St. Johns River Water Management District, 2010). Irrigation was applied for 51 min (irrigation controller run time for two irrigation events per week at a 0.13 cm·h⁻¹ application rate, assuming system efficiency of 80% and considering effective rainfall) per plot on Mondays and Thursdays starting at 0300 hr and ending at ≈0900 hr.

Soil physical and chemical properties.

Soil physical and chemical properties were measured before tillage, aeration, and compost treatments were applied and then repeated every 3 months [0, 13, 27, 40, and 52 weeks after treatment (WAT)] for a period of 1 year. Soil D_b was measured on a single soil core collected from the turf and ornamental beds (two samples per plot) at 0 to 10 cm using the core method (Blake and Hartge, 1986). Ten to 15 soil cores were collected in a random pattern from the turf areas and ornamentals beds in each landscape plot using a soil probe at a depth of 0 to 10 cm and 10 to 20 cm and mixed together to form two composite samples per plot (one per depth). Composite soil samples were then air-dried at room temperature (25 ± 2 °C) and sieved to pass a 2-mm screen. Soil pH (1:2 soil to deionized water ratio), EC (1:2 soil to deionized water ratio), and OM (loss on ignition) were determined by standard methods of the University of Florida–IFAS Extension Soil Testing Laboratory (Mylavarapu, 2009). Soil moisture content at field capacity was determined by the method described in Tan (1996) and particle size was determined by the hydrometer method (Bouyoucos, 1962). Composite soil samples were extracted using the Mehlich-1 solution (1:4 ratio of soil to 0.0125 M H₂SO₄ + 0.05 M HCl) (Mylavarapu, 2009) and analyzed for P and K by inductively coupled plasma atomic emission spectroscopy (ICP-AES).

Plant growth, root growth, and tissue analysis.

Plant growth measurements and tissue nutrient content were determined at 13 and 40 WAT to evaluate the effect of soil tillage or compost amendment on the establishment and growth of ornamental plants and turfgrass. Growth index (GI) was used as a quantitative indicator of ornamental plant growth rate and to compare the size of the plants grown in the different soil treatments. GI for each plant was calculated as: GI (m³) = H × W1 × W2; where H is the plant height (m), W1 is the widest width (m), and W2 is the width perpendicular to the widest width (m) (Scheiber et al., 2007). Turfgrass was mowed on an as-needed basis with most mowing events occurring during the summer months. Turf clippings were collected during mowing events at 13 and 40 WAT to determine clipping dry weight based on the method outlined by Ervin and Koski (2001) with some modifications. A 0.46-m wide section from the center of each plot was mowed to a height of 5.7 cm. The clippings were collected from a bag attached to the mower after every plot and then dried to a constant mass at 105 °C and weighed. Ornamental and turf tissue samples were collected by randomly sampling ≈40 to 50 leaves or blades of grass from each plot at 13 and 40 WAT. Plant tissue samples were dried at 105 °C and digested using the standard method of the University of Florida–IFAS Extension Soil Testing Laboratory (Mylavarapu, 2009) and analyzed for total Kjeldahl N (TKN), and total P and K by ICP-AES.

Shrub root cross-sectional area was used to compare the influence of soil treatments on root growth of shrub species (i.e., G. glauca, R. indica, and I. cornuta) based on the methods outlined by Gilman et al. (2010). Shrubs from all landscape plots were dug from the ground and soil was removed from the root balls using a high-pressure water spray. A caliper was used to measure the diameter of the 15 largest roots growing at a soil depth of 0 to 10 cm that measured greater than 1 cm. The cross-sectional area of each root was then calculated from root diameter as: CSA (cm²) = (½ × root diameter)² × π.

Data analysis.

The experiment was designed as a randomized complete block split-plot design with four blocks and six soil treatments (main plot) in each block. Half of each plot was planted with ornamental plants and the other half was planted with turfgrass (subplot) as described previously. The soil treatments were assigned randomly within each block. Soil properties were analyzed using the PROC MIXED procedure in SAS with soil treatment as a fixed effect and block as a random effect (SAS Institute, 2003). Plant cover was added as a fixed effect for analysis of D_b samples, because separate samples were collected from turf and ornamental beds. Plant GI was analyzed separately for each shrub species using the PROC MIXED procedure in SAS with soil treatment as a fixed effect and block as a random effect (SAS Institute, 2003); initial GI (measured at 0 WAT) was included in the model as a covariate to account for variation in initial plant size. Soil and plant GI data were analyzed separately for each sample collection date when significant treatment × date interaction or date effects were noted. Root cross-sectional area was analyzed using the PROC GLM procedure in SAS (SAS Institute, 2003). All data were checked for normality by examining histogram and normality plots of the conditional residuals (generated by the command plots = residual panel). All pairwise comparisons were completed using the Tukey honestly significant difference test with a significance level of α = 0.05.

Soil physical and chemical properties.

Initial soil samples were divided between the topsoil fill (0 to 10 cm) and the native field soil (10 to 20 cm). Soil particle size analysis indicated that the texture classification of the topsoil fill was loamy sand and the native field soil was sand. The pH was 7.5 and 6.5 and the EC was 0.30 and 0.49 dS·m⁻¹ for the topsoil fill and the field soil, respectively. Initial Mehlich-1 nutrient content analysis indicated that the field soil had lower nutrient concentrations, with the exception of P, than the topsoil fill. Mehlich-1 P and K concentrations were 145 and 20.2 mg·kg⁻¹ in the topsoil fill and 77.6 and 9.7 mg·kg⁻¹ in the native field soil. Based on soil test results, only applications of N and K would be recommended for these soils (Kidder et al., 1998).

There were no significant changes in soil particle size distribution as a result of soil treatment throughout the study; the soils were predominantly sand (mean = 88%) with very little silt (mean = 5%) or clay (mean = 6%) (data not shown). A significant soil treatment × vegetative cover interaction on soil D_b was evident; soil D_b generally followed the trend: compost-amended soils planted with ornamental plants < compost-amended soils planted with turf < unamended (control) or tilled/aerated soils planted with turf ≈ unamended (control) or tilled/aerated soils planted with ornamental plants (Table 1). Multiple researchers have reported decreased D_b as a result of incorporating compost (derived from materials including yard waste, biosolids, brewery byproducts, chicken manure, horse/cow manure, papermill byproducts, mushroom substrate) into field soils (Curtis and Claassen, 2009; Landschoot and McNitt, 1994). While we reported a decrease in soil D_b, the D_b of the unamended soils was less than the 1.8 g·cm⁻¹ threshold for root restriction for a sandy soil (Hanks and Lewandowski, 2003), suggesting that D_b was unlikely to influence plant growth in our study. The significant vegetative cover × soil treatment effect on soil D_b is likely related to physiological differences in the root systems of woody ornamentals versus turfgrass (i.e., woody versus fibrous). We hypothesize that enhanced growth of woody roots in the compost-amended soils created more macropores, resulting in decreased bulk density. Although turf root growth was probably also enhanced by application of compost, the fibrous nature of the root systems made turf roots less effective at increasing soil porosity and, thereby, decreasing bulk density.

Table 1.

Bulk density (D_b) of fill soil samples (n = 4) collected from 0- to 10-cm depth in simulated residential landscape plots receiving compost, shallow tillage, and/or aeration treatments averaged over five sampling dates.

View Table

Addition of compost (incorporated or as a top-dress) significantly increased the soil field moisture capacity compared with unamended soils at all sampling dates (Table 2). We found that compost application improved soil field moisture capacity, thereby increasing the volume of plant-available water in the soil. Pandey and Shukla (2006) reported that the application of yard trimming compost to soils at a commercial vegetable farm at a rate of 100 Mg·ha⁻¹ increased the soil moisture content compared with soils to which no compost was applied. Similarly, Curtis and Claassen (2009) found that plant-available water was increased after the addition of composted yard waste to coarse-textured field soils at a rate of 540 Mg·ha⁻¹ (on a dry mass basis; equivalent to 25% by vol.). However, the authors noted that there was no difference in plant-available water between fine-textured field soils amended with compost and unamended soils.

Table 2.

Selected physical and chemical properties of fill soil samples (n = 4) collected from 0- to 10-cm depth in simulated residential landscape plots receiving compost, shallow tillage, and/or aeration treatments at five sampling dates.

View Table

As expected, the application of compost increased the soil OM content compared with unamended soils through 40 WAT (Table 2). These results are consistent with many other studies that report increases in soil OM after the addition of compost materials (Curtis and Claassen, 2009; Ingelmo et al., 1998; Landschoot and McNitt, 1994; Wright et al., 2007). Manures and manure-based compost typically contain lower amounts of lignin and higher amounts of cellulose (and other more labile C forms) than composts derived from woody plant materials (Casale et al., 1995; Litvany and Ozores-Hampton, 2002), suggesting that C in these composted manures would be a readily available energy source for soil microbes. Therefore, we hypothesize that the low lignin content of the composted dairy manure solids used in our study, coupled with the warm, wet conditions that are common in Florida, allowed for rapid oxidation of C after the compost was applied to the soil. As a result, we reported no treatment differences on soil OM content after 40 WAT (Table 2).

The soil pH and EC were affected by the application of compost and, in some cases, tillage treatments. Soils that received compost additions had a lower soil pH (mean pH = 7.29) than unamended soils (pH = 7.70); this trend persisted from 13 through 52 WAT (Table 2). The decrease in soil pH after application of compost was the result of the addition of compost with a lower pH than reported in the fill soils (compost pH = 6.59; fill soil pH = 7.50) or the production of H₂CO₃ during microbial oxidation of compost derived C (Scharenbroch, 2009). Other researchers have documented that the pH of compost can influence soil pH when compost is applied to the soil (Eghball, 1999, 2002; Scharenbroch, 2009; Weindorf et al., 2006). The decrease in pH after addition of compost to our soils may improve plant growth and quality because of enhanced plant availability of micronutrients (e.g., iron, manganese, copper). However, the pH of the composted soils in our study remained higher than the target pH for S. secundatum and woody ornamentals (6.5 and 6.0, respectively) (Kidder et al., 1998).

In general, soils receiving compost exhibited significantly higher EC than soils receiving no compost additions through 27 WAT (Table 2). Johnson et al. (2006) showed that amending soils with composted dairy manure increased soil EC when the compost application rate exceeded 99 m³·ha⁻¹; the compost application rate in our experiment was 508 m³·ha⁻¹. Stamatiadis et al. (1999) also reported an increase in soil EC after application of compost as a result of the presence of salts (other than nitrates) in the compost material. The increase in soil EC in our study was a result of the compost having a higher EC (1.02 dS·m⁻¹) than the fill soil (0.30 dS·m⁻¹). However, the application of this compost did not increase soil EC to levels (greater than 3 dS·m⁻³) that would be detrimental to even the most salt-sensitive landscape plants (Miyamoto et al., 2004). After 27 WAT, compost-amended soils no longer had higher EC than the soils receiving no compost, which was probably a result of the added salts leaching downward through the soil profile with heavy rain events. Additionally, Mehlich-1 sodium (Na) concentrations in the compost-amended soils decreased from 66.9 mg·kg⁻¹ at 0 WAT to 23.6 mg·kg⁻¹ (not significantly different from unamended soils) by 13 WAT. These soil test Na concentrations were below values that would impact plant growth (E.A. Hanlon, personal communication).

The nutrient content of soil was also affected by the addition of compost. Concentrations of Mehlich-1 P were generally higher in composted soils (Table 2); however, all soils in our study had very high Mehlich-1 P concentrations and would require no additional P fertilizer (Kidder et al., 1998). Wright et al. (2007) reported an increase in NH₄OAc-EDTA-extractable P in soils with increasing compost application rates. Gilley and Eghball (2002) also found that soil test (Bray 1) P at 0 to 5 and 5 to 15 cm was significantly greater after 4 years of corn production when composted beef manure was applied based on crop N requirements. Warman et al. (2009) found that an application of municipal solid waste compost applied to field soil at an N rate of 400 kg·ha⁻¹ resulted in greater Mehlich-3-extractable P and K compared with an unamended field soil. Similarly, we also showed an increase in the concentration of Mehlich-1 K concentrations when compost was applied (Table 2). Mehlich-1 soil test interpretations for Florida indicated that the soil test K in the compost amended soils was high (61 to 125 mg·kg⁻¹; no fertilizer recommended) (Kidder et al., 1998) through 13 WAT as a result of high K content in the compost. As a result, Mehlich-1 K concentrations were sufficient for growth of turf and ornamental plants throughout much of the study (Kidder et al., 1998). However, Mehlich-1 K concentrations in compost-amended soils had declined by 52 WAT, suggesting that K was absorbed by plant roots or leached downward into the soil profile (Table 2).

Plant growth, root growth, and tissue analysis.

The response of plants to the application of compost was species-dependent. Although compost and tillage treatments had no effect on plant GI at 13 WAT (Table 3), one or more of the soil treatments that included compost increased the GI of G. glauca, I. cornuta, and L. musicari by 40 WAT (Table 3). The application of compost also increased root growth of G. glauca and I. cornuta. Mean total cross-sectional area of G. glauca roots was 578, 681, and 809 cm² in composted soils (compost only, compost + aeration, and compost + tillage, respectively) compared with 206 cm² in the unamended control soil. Similarly, total cross-sectional area of I. cornuta roots was 706, 597, and 578 cm² in composted soils (compost only, compost + aeration, and compost + tillage, respectively), compared with 339 cm² in the unamended soils. In contrast, soil treatments did not affect the GI (Table 3) or root growth (data not shown) of R. indica. In general, soils amended with composted dairy manure solids resulted in larger plants than unamended soils. Shoot and root growth differences between species were attributed to differences in plant growth habits and nutritional needs. The dry mass of S. secundatum clippings was greater from compost-amended soils compared with uncomposted soils at 13 WAT (Table 4). Although there were no significant soil treatment effects at 40 WAT, our results suggest that adding compost may help turf to establish more quickly and produce greater biomass compared with unamended soils (Table 4).

Table 3.

Growth index (m³) and plant tissue nutrient content (g·kg⁻¹) for landscape ornamentals (n = 8) grown in sandy fill soils receiving compost, shallow tillage, and/or aeration treatments in simulated residential landscape plots at 13 and 40 weeks after treatment.

View Table

Table 4.

Dry mass and nutrient concentrations (n = 4) of Stenotaphrum secundatum clippings collected from turf grown in sandy soils receiving compost, shallow tillage, and/or aeration treatments in simulated landscape plots at 13 and 40 WAT.

View Table

Several researchers have reported an increase in shoot and root growth of ornamental landscape plants or turf after the application of composts. For example, a study by Rivenshield (2003) found that additions of food waste compost to compacted urban soil increased plant vigor and growth of Acer saccharum Marshall and Acer saccharinum L. trees. Curtis and Claassen (2009) reported an increase in biomass of ornamental grasses [10.8 times greater for Elymus multisetus (J. G. Sm.) Burtt Davy and 1.6 times greater for Nassella pulchra (Hitchc.) Barkworth] compared with control or tillage alone treatments when yard waste compost was applied to disturbed soils formed from lahar or sandstone parent material, respectively. Similarly, Caravaca et al. (2003) reported an increase in shoot biomass of 120% and 360% for Pistacia lentiscus L. and Retama sphaerocarpa (L.) Boiss. shrubs, respectively, planted into a degraded silt-loam soil amended with composted urban residue. Scharenbroch (2009) also reported an increase in shoot and root growth across plant species and organic amendment types when organic amendments were added to soil.

There are several possible explanations for the improved growth of ornamentals and turf reported in our study. For example, the increase in field capacity (and corresponding increase in plant-available water) may have impacted root and shoot growth. However, because plant water stress was not measured in this study, we cannot definitively say whether an increase in water-holding capacity of the soil was responsible for the increase in plant growth. Alternatively, it is possible that plant growth was enhanced as a result of increased soil fertility attributable to the addition of significant amounts of plant nutrients in the compost (particularly N that was mineralized as a result of soil microbial activity stimulated by compost application). This theory is supported by our results showing an increase in tissue TKN when plants were grown in soils receiving compost (+ inorganic N) compared with those grown in unamended soils (inorganic N only) (Tables 3 and 4). The concentration of TKN in the tissue of ornamentals and turf grown in soils receiving composted was higher than for plants grown in uncomposted soils that received only inorganic sources of N (Tables 3 and 4). Based on published concentrations of adequate tissue N for woody ornamentals (20 to 25 mg·kg⁻¹) (Yeager, 2010) and S. secundatum (20 to 30 mg·kg⁻¹) (Sartain, 2008), our results suggest a plant response as a result of the mineralization of compost N (perhaps resulting from enhanced soil microbial activity).

Although there was a more pronounced effect of compost on tissue P content of G. glauca (13 WAT) and I. cornuta (40 WAT) than the other ornamental plant species (Tables 3 and 4), the overall trends indicate that tissue P concentrations were sufficient for plants grown with or without compost [1.5 to 5.0 mg·kg⁻¹ for woody ornamentals and S. secundatum (Sartain, 2008; Yeager, 2010)]. Soil treatments had little consistent effect on levels of tissue total K at any time for any shrub species, but tissue levels were above reported sufficiency ranges for all treatments (data not shown). Therefore, a response to P and K added in the compost treatments was unlikely.