Pine Bark Substrates Amended with Parboiled Rice Hulls: Physical Properties and Growth of Container-grown Spirea during Long-term Nursery Production

The decline in the availability of pine (Pinus taeda L.) bark (PB) supplies and increasing prices have caused concerns in the nursery industry. Research was conducted to evaluate the effect of parboiled rice (Oryza sativa L.) hulls (PBH) as a substrate amendment to PB-based container substrates on the growth of Spiraea ·bumalda L. ‘Anthony Waterer’ and to examine the changes in physical properties of the substrates during long-term production cycles under outdoor nursery conditions. Six substrates were formulated by blending PB with 0%, 20%, 40%, 60%, 80%, or 100% PBH (by volume). Substrate composition affected plant growth components evaluated, generally decreasing growth as the amount of PBH increased. However, amending PB with up to 40% PBH did not result in a significant decrease in plant growth or increase the volume or frequency of irrigation for container-grown spirea. Physical properties of substrates amended with PBH improved over time. Based on these results, PB-based substrates amended with up to 40% PBH retained physical properties that were generally within current guidelines for nursery container substrates after one (25 weeks) and two (70 weeks) growing seasons. The most common components of soilless container media used by the nursery industry in the United States are bark from loblolly pine (Pinus taeda L.) and douglas fir [Pseudotsuga menziesii (Mirb.) Franco]. Loblolly PB is widely used by growers on the East Coast, in the Midwest, and in the southern regions of the United States and douglas fir bark is commonly used on the West Coast. Within the last decade, the nursery industry has faced a steady decline in the availability of PB as well as higher costs because of an increase in demand for alternative uses (e.g., heating fuel), a decline in log harvest, and an increase in freight costs (Haynes, 2003; Lu et al., 2006). A greater shortage and inferior quality of PB are expected as a result of the increasing demand for wood-based materials to be used as biofuels (Day, 2009). The industry is interested in alternative, economical, and sustainable container substrates that are able to provide adequate growing conditions for nursery production. To address this issue, a long list of bark alternatives has been evaluated, including but not limited to pine trees/wood (Boyer et al., 2008; Fain et al., 2008; Jackson, 2008; Jackson et al., 2009; Wright and Browder, 2005; Wright et al., 2006), recycled paper (Craig and Cole, 2000), composted turkey litter (Tyler et al., 1993), cotton gin waste (Cole et al., 2005; Jackson et al., 2005; Owings, 1993), and sewage sludge (Guerrero et al., 2002). Other new materials currently being evaluated as substitutes for PB in nursery production include some fast-growing herbaceous crops such as switchgrass, willow, corn, and bamboo (Boyer et al., 2010). Another possible alternative is rice (Oryza sativa L.) hull. Rice hulls are a relatively underused and sustainable container substrate, which are normally considered a waste byproduct of the rice milling and processing industry (Lovelace and Kuczmarski, 1992). Large quantities of rice hulls are produced annually in the United States, especially in the southern and western states. Numerous studies have been conducted evaluating different forms of rice hulls as alternative substrates in propagation, greenhouse, and nursery production. Rice hulls are available in a variety of forms, including fresh, aged, carbonized, composted, burnt, and parboiled (Buck, 2008). Fresh rice hulls are typically avoided as container substrates because of residual rice and/or weed seed. Parboiled rice hulls are produced by steaming and drying rice hulls after the milling process. This results in a lightweight and consistent product that is free of viable weed and/or rice seed (Evans and Gachukia, 2004). Another advantage in using PBH as a horticultural substrate amendment is the low decomposition rate during the typical production cycle of nursery crops. Despite being an organic compound, rice hulls consist mainly of lignin, cutin, and insoluble silica, providing a slow breakdown of particles and therefore making PBH an appropriate substrate for longterm crop production (Juliano et al., 1987). Einert and Guidry (1975) published some of the earliest work on the use of fresh and composted rice hulls as an amendment for the soil-based container production of woody ornamentals. Although statistical analyses were not included, either form of rice hulls appeared to be a suitable media amendment based on mortality and growth data for Pfitzer juniper [Juniperus ·pfitzeriana (L.) Späth]. Laiche and Nash (1990) evaluated the effect of composted rice hulls on the growth of three woody plants (Rhododendron indicum L., Ilex crenata Thunb., and Juniperus horizontalis Moench.) in containers. Their results demonstrated plant growth in organic components of 100% composted rice hulls or 50% composted rice hulls:50% bark compared favorably with the growth obtained using 100% PB. Lovelace and Kuczmarski (1992) reported that aged rice hulls compared favorably to 100% PB in cost and performance when used as a component of a blend including PB, rice hulls, and sand (2:2:1 by volume) for a variety of woody ornamentals. Baiyeri (2005) demonstrated that when using composted rice hulls amended with poultry manure (3:1 by volume), sucker plantlets from five banana genotypes generally resulted in more vigorous suckers than when sawdust and poultry manure (3:1 by volume) or rice hulls, sawdust, and poultry manure (1.5:1.5:1 by volume) were used at the nursery stage of production. Fresh (Einert, 1972; Papafotiou et al., 2001; Sambo et al., 2008), carbonized (Kämpf and Jung, 1991; Tatum and Winter, 1997), parboiled (Evans and Gachukia, 2007), and ground parboiled (Buck and Evans, 2010) rice hulls have been evaluated as substrates on a number of greenhouse crops. Dueitt and Newman (1994) determined that fresh rice hulls hold more water than aged rice hulls in greenhouse media for seedlings of Tagetes erecta L. and Limonium suworowii (Reg.) Kuntze. Evans and Gachukia (2007) reported that the large particles of PBH provide adequate drainage and aeration in peat-based substrates. More recently, Buck and Evans (2010) revealed that given its physical properties, ground PBH can be used as a suitable replacement for up to 40% peatmoss to grow greenhouse crops. A preferred container substrate should provide stable plant support, a reservoir for nutrients and water to the root system, and adequate gas exchange (Nelson, 2003). Bunt (1988) stated that the most important physical properties of containerized substrates are dry bulk density (DBD; g cm), air-filled pore space (AS; %), water-holding capacity Received for publication 31 Jan. 2011. Accepted for publication 15 Mar. 2011. This research was funded in part by Riceland Foods, Inc. Stuttgart, AR. Pine bark provided by Sun Gro Horticulture, Pine Bluff, AR. Assistance from Douglas Karcher, David Hensley, and Leonel Espinoza is gratefully acknowledged. Former graduate student. Professor. To whom reprint requests should be addressed; e-mail [email protected]. 784 HORTSCIENCE VOL. 46(5) MAY 2011 (WHC; %), and total porosity (TP; %). Particle size distribution is also considered a fundamental characteristic in the physical properties of substrates because the shape, size, and density of the individual particles as well as the proportion of the different particle sizes will determine the proportion of air and water in a substrate (Handreck and Black, 2002). Physical properties of substrates in containers will change over time as a result of the decomposition of the components caused by physical and biological degradation. The breakdown of the components generally results in lower AS, higher water retention, and increased weight of solids in the container (Ingram et al., 2003). Substrate shrinkage is also an important factor to consider when looking at the changes in physical properties over time. Nash and Pokorny (1990) defined shrinkage as the loss of bulk volume of substrates in containers, which can be attributed to the settling of fine particles into the macropores located between coarse particles. Organic matter decomposition as well as physical breakdown will also cause the substrate to shrink as the particles become smaller and fit closer together (Ingram et al., 2003). Erosion can also affect substrate shrinkage in that the substrate particles can be washed out of the container after intense rainfalls and/or irrigation depending on the particle sizes (Ingram et al., 2003). Recommended ranges for the commercial production of nursery crops include: DBD (0.19 to 0.7 g cm), TP (50% to 85%), AS (10% to 30%), and WHC (45% to 65%) (Yeager et al., 2007). Plant production management can be expected to be less intensive if substrates are maintained within these suggested physical property ranges (Bilderback et al., 2005); however, growers’ production techniques can influence the outcomes. Research is needed on the effect of PBH as an amendment for PB-based container substrates in the long-term production of ornamental plants. Therefore, the objectives of the study were to compare the changes in physical properties as well as the plant growth responses for the PB substrates amended with various ratios of PBH and to characterize how the amount of PBH affected physical properties for production of container-grown shrubs over long-term crop cycles. Materials and Methods Substrate formulation. A preliminary experiment was conducted to determine the lime rate required to adjust PB (Sun Gro Horticulture, Pine Bluff, AR; pH 5.7) to a pH similar to PBH (Riceland Foods, Stuttgart, AR; pH 6.2 to 6.6). Small batches of composted PB were amended with pelletized dolomitic lime (M.K. Minerals, Manhattan, KS) at rates of 0, 1.8, 3.6, 5.3, 7.1, and 8.9 kg m and were allowed to age for 2 weeks before measuring the pH. Based on results from this experiment, a lime rate of 8.9 kg m was pre-plant-incorporated to all PB. Six media substrates were formulated by blending PBH with pH-adjusted PB (pH 6.4). Individual blends with 0%, 20%, 40%, 60%, 80%, or 100% PBH (by volume) were mixed in a Mitchell Ellis 1-cubic yard soil mixer (Mitchell Ellis, Semmes, AL) on 14 Apr. 2009 and 24 Mar. 2010. Osmocote Plus (15N-3.9P-10K; 8–9 month; O.M. Scotts Horticulture Products, Marysville, OH) was preplant-incorporated at a rate of 7.1 kg m. Containers were weighed (± 1 g) using an electronic balance as they were filled to minimize variation among containers. On 14 Apr. 2009, spirea (Spiraea ·bumalda L. ‘Anthony Waterer’) liners (average 16-cm tall) were potted in #2 plastic containers [Classic 600, Nursery Supplies Inc., Chambersburg, PA; 22.2-cm (h)· 21.6-cm (top diameter) · 17.5-cm (bottom diameter)] and #5 plastic containers [Classic 2000, Nursery Supplies Inc.; 28.2-cm (h) · 29.2-cm (top diameter) · 24.5-cm (bottom diameter)]. Additional containers were filled with similar substrates but were left unplanted (fallow containers) to compare changes in substrate physical properties with and without plants. Fallow containers were managed similarly as the containers with plants for the duration of the study. The two studies with the different container sizes were placed on an outdoor (no cover) gravel area at the University of Arkansas Horticulture Research Farm (UAHRF) in Fayetteville, AR (lat. 36 06# N, long. 94 10# W; hardiness zone 6). One week after planting (WAP), all containers were treated with preemergent herbicide (Pendulum 2G, pendimethalin; BASF Corp., Research Triangle Park, NC) at a rate of 224 kg ha. The study using #2 containers with and without the different plant species was replicated on 1 Apr. 2010 to evaluate growing season differences. The study using #5 containers remained in the field for a 70-week period to monitor changes in physical properties as well as the plant growth differences from the individual substrates over longer production cycles. During the growing season, containers were spaced as needed to alleviate canopy overlap. Containers were placed pot-to-pot during the winter. On 13 Apr. 2010, plants grown in #5 containers were re-fertilized (top-dressed) at a rate of 5.3 kg m with the same fertilizer initially applied. Irrigation. The irrigation method varied according to the size of the containers. Irrigation for plants growing in #2 containers was determined based on the container capacity (CC) of the individual substrates. During the first year of the study, the amount of available water was established in a preliminary experiment using tomato (Lycopersicon esculentum Mill.) plants planted on 24 Feb. 2009 following the methods of Argo and Biernbaum (1994). The preliminary study was conducted in a climate-controlled glass greenhouse at the University of Arkansas. Plants were grown in #2 containers with the same substrate blends used in the main experiment. On 2 May 2009, all plants were watered thoroughly and then allowed to sit for 1 h to establish CC weight. After this, no additional water was supplied to these plants. The tomato plants were monitored twice daily to observe the first signs of visible wilt and containers with plants were weighed. The difference between the CC and wilt weight was used as an estimate of the total amount of available water held in the root media after irrigation. Concurrent with the first year of the main study, a second preliminary experiment was conducted using spirea. Plants received overhead irrigation for the duration of the experiment, after which, following the same methodology previously described, the amount of available water for each container substrate was determined. The calculated value for each substrate was used as the target weight in the irrigation procedures for the #2 containers in 2010. From the date of planting until 15 May 2009 and 6 May 2010, plants were handwatered so the liners could get established in their containers, after which irrigation was treated independently for each root substrate treatment. Each day the decision to irrigate the substrate treatments was determined gravimetrically when the average weight of three pots reached a target weight based on a loss of more than 50% of the available water as determined in the respective preliminary experiment. Sufficient water was applied for each substrate to return containers to CC providing a 20% leaching fraction (LF). The indicator pots were weighed before and after irrigation to determine the amount of water absorbed by the root medium. The volume of water was recorded every time water was applied. Fallow #2 containers were overhead-irrigated whenever containers with spirea plants received water. Number 5 containers were overhead-irrigated as needed depending on the weather conditions for the length of the experiment. Sufficient water was applied for each substrate to return containers to CC providing a 20% LF. The municipal water source (Lowell, AR) for the UAHRF had a pH that ranged from 6.5 to 8.5 and an alkalinity of 47 ppm CaCO3. Plants grown in #5 containers were harvested on 19 Aug. 2010 (70 WAP). Plants grown in #2 containers were harvested on 8 Oct. 2009 and 29 Sept. 2010 (25 WAP). The average daily temperatures from April to October for the first and the second years of the study were 20 and 22 C. Physical properties: initial. Immediately after the blending process, three random samples of each substrate were collected for further analysis. Total porosity (v/v), AS (v/v), WHC (v/v), and DBD (w/v), were determined by using air-dried substrate samples. Samples were rewetted to a moisture level of 50% (w/w) and allowed to equilibrate to attain moisture uniformity. Three replicate samples of each substrate were then packed into 7.6-cm tall by 7.6-cm i.d. aluminum cylinder porometers. Physical properties were determined following the NCSU Porometer methods as described by Fonteno et al. (1995). Wet bulk density (WBD) was calculated by dividing the wet weight of each substrate sample by the core volume. Substrate shrinkage was HORTSCIENCE VOL. 46(5) MAY 2011 785 determined by measuring the difference in substrate height (centimeters from the top of the container to the substrate surface measured in four locations per container) at 1 WAP and again at harvest. Shrinkage is reported as a percent decrease from its original height. Initial physical properties for each experiment were analyzed separately. Variables for the study using #5 containers were subject to mean separation among substrates. For the study using #2 containers, the same variables were analyzed as a 6 · 2 factorial with six substrates and two years (2009 and 2010) in a completely randomized design. Results were subjected to the analysis of variance (ANOVA) procedure and means were separated using Tukey’s honestly significance difference (HSD). Physical properties: final. Root medium samples were collected from each container by manually separating the roots from the substrate. Three replicate samples of each substrate were obtained by mixing the substrate from two containers together. Each replicate sample was used for physical property measurements following the same procedures used for initial measurements. The process of removing substrates from the roots and further preparing them for analysis involved very vigorous handling of the substrates. It was therefore surprising to find in substrates with 40% or more PBH, ‘‘aggregates’’ of PBH particles. However, while preparing substrates for physical properties analysis, we used substrate samples with no ‘‘aggregates.’’ Samples of ‘‘aggregates’’ analyzed under a dissecting microscope revealed what looked like fungal hyphae. Samples were plated by a plant pathologist in the Department of Plant Pathology at the University of Arkansas and determined to belong to the genera Mucor, Trichoderma, and Fusarium. Although samples from undisturbed substrate-filled containers were not used to determine final physical properties, the analysis of the substrates that remained in the containers for the duration of the studies were used as an indicator of how the physical properties of the components changed over time compared with how they were initially. The final physical properties of the substrates discussed in this article are not representative of the physical properties in the containers after 70 and 25 weeks under nursery production conditions. Final physical properties (including shrinkage) for #5 containers were analyzed as a 6 · 2 factorial with six substrates and two planting methods (with plants or fallow containers). Final physical properties for #2 containers were analyzed as a 6 · 2 · 2 factorial with six substrates, two planting methods, and 2 years (2009 and 2010). Both studies consisted of three substrate replications per treatment combination (n = 3) arranged in a completely randomized design. Data were subjected to ANOVA and means were separated by Tukey’s HSD. Because initial data were not specific for a given planting method, significance for the change in physical properties over time was based on a 95% confidence interval (CI). The change over time was obtained by subtracting the final value from its respective initial value. If a CI did not overlap with zero, then a significant change over time was considered. Plant measurement and evaluation. Initial plant width (average of the plant width measured in two directions) and height was recorded at 1 WAP. Final plant width and height were recorded before harvesting and a growth index (GI) calculated. This index was calculated by the formula p · h · r, where h is shoot height and r is calculated by multiplying 1⁄2 times the mean of two diameter measurements taken at a 90 angle from each other. At harvest, the stems were cut at the substrate surface and dried in a forced-air oven (40 C) for 96 h and recorded as shoot dry weight (SDW). Subsequently, substrate was separated from the roots and the roots dried in a forced-air oven (40 C) for 96 h and recorded as root dry weight (RDW). Plant growth (GI, SDW, and RDW) was analyzed by plant species in a perfect Latin square design with each of the substrates appearing once in each row and once in each column for a total of 36 containers. Two growing seasons (2009 and 2010) were analyzed for plants growing in #2 containers. Six single plant replications were used for each substrate. Data were analyzed using a multiple comparison of means at a = 0.05 with Tukey’s HSD and the ANOVA procedure. All data were analyzed with JMP 8 (SAS Institute, Inc., Cary, NC). Results and Discussion No. 5 containers Substrate physical properties. Significance for main effects and the interaction among factors related to shrinkage, TP, AS, WHC, and DBD are shown in Table 1. Shrinkage was greater in fallow containers (32 cm) than in containers with plants (14 cm). The lower substrate shrinkage observed when plants were present has already been reported for wood-based substrates by Jackson et al. (2009), who suggested it is likely a result of plant roots filling the substrate voids created by decomposition and thus preventing the loss of volume by microbial degradation. Total porosity in the initial blends was greatest in substrates with 40% to 100% PBH and TP in those same blends was above the recommended range (50% to 85%; Yeager et al., 2007) (Table 2). In substrates with 40% to 80% PBH, TP significantly decreased over the 70 weeks regardless of the planting method. A significantly lower TP was observed in blends containing 20% PBH when plants were present compared with fallow containers. Otherwise, the presence of plants did not alter the TP of the substrates. At 70 WAP, TP in fallow containers with up to 40% PBH and containers with plants with up to 80% PBH fell within the range suggested by Yeager et al. (2007). In the initial blends, AS increased as the percentage of PBH increased (Table 2). Overall, AS decreased significantly over time. Except for substrate with 80% PBH, AS over the 70 weeks of the experiment was unaffected by planting method. The suggested range for AS in container substrates is 10% to 30% (Yeager et al., 2007). Initially, all substrates that contained PBH in the blends had AS percentages above the acceptable range. However, at 70 WAP, fallow containers with up to 60% PBH and containers with plants with up to 40% PBH fell within the recommended range. In the initial blends, WHC decreased as the percentage of PBH increased (Table 2); that trend was generally unaffected after 70 weeks regardless of the planting method. In general, WHC significantly increased over time. Fallow containers with 60% and 80% PBH resulted in greater WHC than containers with plants; otherwise, the presence of plants did not alter WHC for the different substrates. The suggested range for WHC of substrates used in containers is 45% to 65% (Yeager et al., 2007). Initially, substrates with 40% or more PBH had WHC percentages below that range; however, at 70 WAP, fallow containers with up to 80% PBH and containers with plants with up to 60% PBH fell within the recommended range, suggesting an increase in WHC over time with the addition of PBH to the blends. Dry bulk density in the initial blends decreased as the percentage of PBH increased (Table 2). Dry bulk density significantly decreased over time for substrates with 0% and 20% PBH regardless of the planting method. Except for fallow containers with 20% PBH, DBD for each substrate was unaffected by planting method. The ideal DBD range is 0.19 to 0.70 g cm (Yeager et al., 2007). By this standard, initial substrates with 40% or more PBH had a DBD below the ideal range. At 70 WAP, only substrates with no addition of PBH fell within the lower margin of the recommended range. Although WBD is not commonly reported as a physical property for container production, growers typically handle containers when wet and could benefit from knowing WBD Table 1. Analysis of variance for physical properties of pine bark substrates amended with parboiled rice hulls 70 weeks after planting in #5 containers with and without (fallow) spirea plants. Source Shrinkage TP AS WHC DBD Planting method *** *** NS ** NS Substrate NS *** *** *** *** Planting method · substrate NS * *** *** ** Planting method = with plants or fallow containers. TP = total porosity; AS = air-filled pore space; WHC = water-holding capacity; DBD = dry bulk density. ***, **, *, NS indicate statistical significance at the 0.001, 0.01, and 0.05 P level and non-significant,