Materials and methods

The substrates evaluated in this study included SLC, OM, and PEAT. Development of the SLC is described by Herring et al. (2018). The OM comprised aged pine bark fines, peat, soil, perlite, and worm castings (Just Natural Organic Potting Mix; Jolly Gardener). The PEAT was a conventional substrate comprising peat with aged bark fines, perlite, vermiculite, dolomitic limestone, gypsum, and a wetting agent (Pro-Line C/P). Three species, basil, chives, and dill, were grown in Raleigh, NC (lat. 35.78°N, long. 78.64°W), with a planting date of 8 June 2016. The experiment consisted of the three substrate treatments (SLC, OM, and PEAT) and three species with six replications in a randomized complete block design. Black, plastic, 72-cell inserts (1-5/9 inch cell top diameter × 2-1/3 inches deep, 3-3/5 inch³ maximum dry volume; Landmark Plastic Co., Akron, OH) were filled with the designated transplant substrate and planted with three seeds of each species per cell. Trays were randomized in the greenhouse (85/65 °F day/night) with natural irradiance and photoperiod under fog (CoolNet Pro Fogger 0303420LL-B; M.L. Irrigation System; Laurens, SC) applied for 8 s every 8 min for germination. Clear plastic sheeting was pulled around and over the bench. Plastic sheeting was removed 12 d after planting and germination was measured. Cells were then thinned to one seedling. Plants were irrigated by hand each day throughout the remainder of the study.

The substrates were evaluated and compared through chemical and physical property analyses, and growth. Physical property analyses including total porosity, airspace, container capacity, bulk density, available water, unavailable water, and particle size distribution were conducted in the Horticultural Substrates Laboratory, Department of Horticultural Science, North Carolina State University, Raleigh. Three replications of each substrate were packed into ≈21-1/5-inch³ cylindrical aluminum rings (3 × 3 inches) and used to determine total porosity, airspace, container capacity, and bulk density per procedures outlined in Tyler et al. (1993). Each substrate was packed into 6-3/20-inch³ cylindrical aluminum rings (3 × 7/8 inches), with five replications per modified procedures of Bilderback and Johnson (1982) and used to determine unavailable water following procedures described in Klute (1986). Available water was calculated as container capacity minus unavailable water. To determine particle size distribution, three 100-g samples of each substrate were dried at 105 °C for 48 h and placed in a sieve shaker (Ro-tap Shaker model B; W.S. Tyler, Mentor, OH) fitted with seven sieves, 6.3, 2, 0.71, 0.5, 0.25, and 0.106 mm for 5 min. The sample from each sieve was weighed, and particle size was expressed as a percentage of the total weight of the sample.

Beginning 29 d after sowing (DAS), substrate pH and EC were measured weekly from planted cells using 1:2 (by volume) substrate to water extracts. Before extraction, cell packs were watered and allowed to drain for 3 h to establish container capacity within the substrate. For each substrate sample, 400 mL of substrate was removed from the tray, thoroughly mixed with 800 mL of distilled water and allowed to sit for 20 min. The solution was then strained through filter paper (grade 1, 185 mm, catalog no. 1001-185; Whatman, Houston, TX) to remove solids. Extract solution EC and pH were measured using a combination EC/pH meter (HI 8424; Hannah Instruments, Ann Arbor, MI). After EC and pH measurements, substrate solution samples were submitted to the North Carolina Department of Agriculture and Consumer Services Agronomy Division, Raleigh for inorganic nitrogen (IN-N), urea, P, K, Zn, Cu, calcium (Ca), magnesium (Mg), sulfur (S), boron (B), iron (Fe), and manganese (Mn). Inorganic nitrogen fraction concentrations included nitrate nitrogen plus nitrite nitrogen (NO₃-N + NO₂-N) and ammonium nitrogen (NH₃-N + NH₄-N). Organic nitrogen fraction concentration included urea. Nitrate nitrogen was determined on ≈10-mL homogenized sample, filtered using acid-washed filter paper (Laboratory Filtration Group, Houston, TX) by nitrate hydrazine reduction (Kempers and Luft, 1988; Skalar Analytical, 1995a); ammonia was determined by a modified Berthelot reaction (adapted from Krom 1980; Skalar Analytical 1995b). The urea concentration was determined with the diacetyl monoxime thiosemicarbazide colorimetric method (Sullivan and Havlin, 1991; Skalar Analytical, 1995c) with an auto-flow spectrophotometric analyzer (San++ Segmented Flow Auto-Analyzer; Skalar Analytical, Breda, The Netherlands). Total concentrations of P, K, Ca, Mg, S, Fe, Mn, Zn, Cu, B, and sodium (Na) were determined on ≈10-mL homogenized sample, filtered using acid-washed filter paper (Laboratory Filtration Group) with inductively coupled plasma–optical emission spectrometry [ICP-OES (Spectro Arcos EOP; Spectro Analytical, Mahwah, NJ)] [Donohue and Aho 1992; adapted from U.S. Environmental Protection Agency (USEPA), 2001]. Chloride concentration was determined on ≈10-mL homogenized sample, filtered using acid-washed filter paper (Laboratory Filtration Group) by the thiocyanate displacement method (Skalar Analytical, 1995d; Zall et al., 1956) with an auto-flow spectrophotometric analyzer (San++ Segmented Flow Auto-Analyzer, Skalar Analytical).

Plant heights, taken from the root collar to the tip of the shoot, were taken at harvest. Harvest dates were determined when each species reached developed transplant stage (43 d for basil and 49 d each for chives and dill). Shoots (stems and leaves) were dried at 62 °C for 24 h. After drying, samples were weighed.

All variables were subjected to analysis of variance procedures using the general linear model procedure in SAS (version 9.4; SAS Institute, Cary, NC) and P value was considered significant at ≤0.05 (SAS Institute, 2016). All means were separated with Tukey’s honestly significant difference means separation test (P ≤ 0.05) where appropriate. Where the substrate × species interaction was nonsignificant, main effects are discussed.

Results and discussion

Regardless of composition, the three substrates did not differ in their particle size distributions (data not shown). Substrates had an average of 4% (6.3 mm), 32% (2 mm), 31% (0.71 mm), 12% (0.5 mm), 12% (0.25 mm), 7% (0.11 mm), and 2% (<0.11 mm). Substrate physical properties varied, except for total porosity, which was similar for all substrates (Table 1). Airspace was greatest for SLC and lowest for OM and PEAT. The opposite occurred for container capacity and available water as OM and PEAT were greatest and SLC was lowest. The SLC did not differ from OM and PEAT in unavailable water, but unavailable water values were greater for OM than PEAT. Bulk density was greatest for OM and SLC. Although root growth was not measured in this study, more prolific root growth throughout the plug was observed with SLC compared with OM and PEAT possibly because of the greater airspace in SLC as shown in Fig. 1. Composts produced from other agricultural wastes, and cow and poultry manures, have also produced suitable physical properties for production of transplants (Bustamante et al., 2008).

Table 1.Initial physical properties of three transplant substrates that were used to grow basil, dill, and chives without additional fertilizer in a greenhouse trial [85/65 °F (29.4/18.3 °C) day/night] with natural irradiance and photoperiod.

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

Days after seeding did not interact with substrate for pH and EC when looking at the substrate solution supporting growth of each species (data not shown). The SLC substrate had a greater pH and EC than OM and PEAT for each species (Table 2). Substrate solution pH did not change substantially over time, whereas EC decreased from 0.24 to 0.14 mS·cm⁻¹. Both substrate pH and EC were within acceptable ranges (Bunt, 1988; Raviv et al., 1986).

Table 2.Influence of substrate type and days after seeding (DAS) on substrate pH and electrical conductivity (EC) on greenhouse-produced basil, chives, and dill transplants. Substrate 1:2 (v/v) extractions, starting at 29 DAS, were measured.

View Fullscreen

For basil and chives, the DAS × substrate interaction was significant for IN-N, NH₄-N, NO₃-N, P, Mg, Mn, and B (data not shown). Swine lagoon compost maintained greater IN-N concentrations than OM and PEAT throughout the 50 DAS as shown in Table 3. Inorganic nitrogen concentration decreased from ≥10 to ≥2 ppm, whereas OM and PEAT had ≥2 ppm at 29 DAS, and these concentrations decreased to ≤2 ppm. Almost all of the IN-N was in the NO₃-N form with SLC throughout which is indicative of a stable compost (California Compost Quality Council, 2001). The SLC substrate also maintained greater concentrations of P, Mg, Fe, Mn, and B than OM or PEAT over time for both basil and chives (Table 3). Phosphorus and Mg concentrations remained greater in SLC than OM and PEAT throughout the study; however, no transplants in SLC showed signs of nutrient toxicity for either nutrient (P. Herring, personal observation). For dill, the DAS × substrate interaction was largely nonsignificant; however, SLC also resulted in greater concentrations of IN-N, NO₃-N, P, Ca, Mg, Fe, Zn, Cu, and B than OM or PEAT, whereas substrate did not affect NH₄-N or K concentrations (Table 4 or data not shown).

Table 3.Effect of substrate and days after seeding (DAS) on nutrient substrate concentration for basil and chives for the duration of a trial in which herbs were grown in three different substrates to test their capability to produce transplants from seed to optimal size without the use of additional fertilizer. Data collection began 29 DAS with 1:2 (v/v) extractions and were continued weekly until harvest of the species.^z

View Fullscreen

Table 4.Effect of substrate and days after seeding (DAS) on substrate nutrient concentration for dill. Data collection began 29 DAS with 1:2 (v/v) extractions and were continued weekly until harvest of the species.

View Fullscreen

Specific heavy metal concerns with SLC are Cu and Zn, which are given as feed supplements to increase growth performance in young pigs (Jacela et al., 2010). Before composting, the swine lagoon solids had Zn levels that slightly exceeded the USEPA target range of ≤2800 mg·L⁻¹ with a level of 3043 mg·L⁻¹ (Herring et al., 2018). Mixing the swine lagoon solids with peanut hulls [15:85 (by volume) SLS:peanut hull] and composting lowered Zn levels to 871 mg·L⁻¹ and below. Copper in the swine lagoon solids were within the EPA limit (1500 mg·L⁻¹) before composting, nevertheless, composting also lowered Cu levels (Herring et al., 2018). Using the compost developed by Herring et al. (2018) in this study, the SLC maintained greater Zn and Cu (Zn average = 0.04 ppm; Cu average = 0.01 ppm) concentrations than OM (Zn average = 0.02 ppm; Cu average = 0.01 ppm) and PEAT (Zn average = 0.03 ppm; Cu average = 0.01 ppm); however, all substrates had acceptable levels (Zn ≤0.5 ppm; Cu ≤3.0 ppm) (Misra, 1995; Panou-Filotheou et al., 2001).

Perhaps, the nutrient of greatest environmental concern with animal wastes is P (Jongbloed and Lenis, 1998; Sharpley et al., 1994). Substrate solution concentrations of P in SLC averaged ≈700 times greater than OM and PEAT across the species (Tables 3 and 4). Phosphorus in the substrate solution of SLC-grown transplants decreased over time from 47.43 ppm (basil) at 29 DAS to 10.57 ppm (chives) at 50 DAS for all species (Table 3). Although no P toxicity symptoms were observed, P remediation (Penn et al., 2001) of runoff water leaving a greenhouse using the SLC substrate is recommended.

Transplant growth was evaluated at germination and at maturity (Table 5). Germination of each species was similar (50% to 60%), regardless of substrate and appears to be in the expected range of chives and dill (Maynard and Hochmuth, 2007). The SLC and PEAT supported germination slightly better for dill; however, the SLC and OM resulted in slightly lower germination for chives. Height for basil, chives, and dill transplants was significantly greater with transplants produced in SLC than in OM and PEAT; however, there was no significant difference between SLC and PEAT for basil (Table 5). Harvest dry weights were also significantly greater for basil and dill when grown in SLC when compared with OM and PEAT. Dry weight of chives was similar regardless of substrate.

Table 5.Effect of substrate on germination of basil, chives, and dill and plant growth at transplant harvest. Germination counts of each species were taken at 12 d after seeding. Height was measured from the root collar to the tip of the shoot on the day of harvest. Dry weights were taken of the shoots (stems and leaves).

View Fullscreen

These results suggest that SLC can be used alone as a transplant substrate to produce basil, chives, and dill transplants of similar quality to commercially available peat-based products with no fertilizer applied. Research is continually adding to this field, as compost is an attractive option as a peat substitute (DeKalb et al., 2014; Kahn et al., 2005; O’Brien and Barker, 1996).

Units

View Fullscreen