Experimental plot setup.

‘Braeburn’ apple [Malus sylvestris (L.) Mill var. domestica (Borkh.) Mansf.] on M.9 root stock was planted Mar. 2000 at 1.2 m (within row) × 2.6-m (between row) spacing (3205 trees/ha) on a coarse-textured soil in a commercial orchard in Naches, WA, and subsequently trained to a post and wire system. The soil was a Cleman very fine sandy loam (coarse-loamy, mixed, superactive, mesic Torrifluventic Haploxeroll). Soil fertility analyses before planting indicated optimum soil pH, K, calcium (Ca), magnesium (Mg), P, low salinity values, above optimum soil NO₃-N, and slightly below optimum soil boron (B). The soil was fumigated with Vapam (metam sodium, active ingredient sodium methyldithiocarbamate) (AMVAC, Los Angeles, CA) in fall of 1999, but analyses conducted in 2003 indicated that the replant disease complex was affecting tree vigor at the site despite the earlier fumigation (Mark Mazzola, personal communication). The trees were established and maintained for the first five growing seasons with eight different irrigation and soil management treatments designed to encourage initial tree establishment and performance. A randomized complete block design with six replicate plots of each treatment was established in three long orchard rows. Each plot consisted of four measurement trees bordered at each end by a guard tree. Pertinent to this report was a factorial combination of four treatments involving presence or absence of mulch with or without soil amendment, all of which were under the same irrigation treatment described subsequently. The four treatments considered here included: 1) a control with neither mulch nor amendment; 2) alfalfa hay mulch applied at a rate ≈50 t·ha⁻¹ in May 2000 within a 1.5-m wide strip centered on the row. The mulch was maintained until the 2004 growing season by periodic applications of additional alfalfa each spring to maintain a 10-cm depth; 3) a minimally CDS amendment applied at 45 t·ha⁻¹ in May 2000 within the tree row as for the mulch; and 4) a combined mulch and amendment treatment established in May 2000 by applying the CDS beneath the alfalfa mulch. The CDS was obtained from a dairy in the Yakima area. Properties of alfalfa hay and the CDS are presented in Table 1.

Table 1.

Concentrations of major nutrients in composted dairy manure solids (CDS) amendment and alfalfa hay mulch and estimated total initial nutrient inputs based on bulk application rates of 45 Mg·ha⁻¹ and 50 Mg·ha⁻¹of CDS and alfalfa at planting, respectively.

View Table

For all treatments, irrigation was applied through two 4-L·h⁻¹ pressure-compensating emitters located at 0.3 m either side of each tree within the row. Irrigation quantity was applied daily in response to the previous day’s evapotranspiration demand automatically scheduled by atmometer (Parchomchuk et al., 1996). In Apr. 2000, ammonium polyphosphate (10N–15P–0K) was fertigated to apply 20 g P per tree over a 3-d period (30 min of fertigation per day). Calcium nitrate (15.5N–0P–0K) was fertigated daily to apply 20 g N per tree from 17 May to 7 Sept. in 2000 and annually, 2001–04, over an 8-week period starting mid-May. Starting in 2002 and annually until 2004, 0.17 g B per tree was fertigated over a 4-week period starting in mid-May. Cultural management of the orchard block was undertaken by the cooperating orchardist and followed standard production practices for the region as described in the Crop Production Guide for Tree Fruits in Washington State (<http://pubs.wsu.edu>).

Plant responses.

Composite samples of 30 leaves from the midportion of extension shoots of the current year’s growth were collected from each of the four measurement trees for each experimental plot annually from 2000 to 2004. All samples were oven-dried at 65 °C and ground in a stainless steel mill. In 2000 and 2001, a 250-mg subsample was digested for 0.75 h on a block digester at 350 °C in a H₂SO₄ solution containing K₂SO₄ and HgO. Nitrogen in the digest was determined through the formation of an ammonium–salicylate complex and P was determined through the formation of a phosphomolybdenum blue complex (Technicon Autoanalyzer II Industrial Method No. 334-74 A/A; Technicron, Elmsford, NY). From 2002 to 2004, leaf N was determined using the LECO FP-528 (LECO Corporation, St. Joseph, MI) combustion analyzer. From 2002 to 2004, leaf P was determined on 0.5-g samples, dry-ashed at 525 °C, and dissolved in 1.2 M HCl before P determination by inductively coupled argon plasma (ICP) spectrophotometry. Ca, Mg, and K were determined on the samples by atomic absorption spectrophotometry and B by ICP spectrophotometry.

Total shoot growth was determined for each measurement tree at harvest in 2000 and 2001. Trunk diameter at 30 cm above the graft union was measured annually in October and used to calculate annual trunk cross-sectional area (TCSA).

Annual yield and the number of fruit were measured for each experimental plot and replicate from 2001 to 2004 in early October at commercial harvest. Average fruit weight for each plot was calculated from these numbers. A randomly selected 10 apple sample from the midcanopy location for each experimental plot was also evaluated annually for fruit quality characteristics. Flesh firmness was determined with a Baullaf (Lake City Technical Products Ltd., Kelowna, British Columbia, Canada) penetrometer (11.1-mm diameter tip). An additional random sample of 10 fruit from each plot was collected for mineral analyses. Samples were rinsed under distilled water and then air-dried. Chemical analyses were conducted on a composite of opposite, unpeeled quarters from each apple minus stem tissue and seeds. In 2000 and 2001, fruit tissue was blended with 1.5 times their weight of distilled water. A 150-mL subsample was further homogenized with a high-speed tissue homogenizer. A weighed 9-mL subsample of homogenized slurry was digested in 5.4 mL of concentrated H₂SO₄ containing NaSO₄ (1.8 g), copper (0.36 mL 25% CuSO₄ solution), and selenium (0.67 g·L⁻¹) at 380% for 1 h. Nitrogen and P were determined by colorimetric methods, as described for leaf samples and Ca, Mg, K, and B were determined by the spectrophotometric methods described for leaves. Starting in 2002, LECO-N was determined on a 0.125-g subsample of freeze-dried sectors and Ca, Mg, and K on a 0.5-g freeze-dried subsample previously described for leaf samples. All fruit mineral analyses data were expressed on a fresh weight (FW) basis.

Soil sampling and microfaunal analyses.

Composite soil samples were taken from each plot in May and Oct. 2000 through 2004. Each sample was composed of eight 15-cm deep cores taken between 15 and 30 cm from the row middle in the vicinity of the four measurement trees in each plot. Soil samples were passed through a 6-mm sieve, and then Baermann pans (16 cm diameter) were used to extract nematodes from 60-mL subsamples of soil over 7 d (Forge and Kimpinski, 2007). Total nematodes in the sample were counted at 40× and then, at 200× and 400×, the first 100 nematodes observed along random transects through the sample were identified. Most specimens were identified to the genus level during these routine analyses at 400×, but Rhabditids and Diplogasterids were counted at the family level.

The abundance of nematodes in the major functional groups (bacterivorous, fungivorous, omnivorous, predacious) and the abundance of nematodes with “enrichment opportunist” ecological characteristics (as defined by Bongers, 1999) were calculated and data were expressed as nematodes per 100 g dry soil for each of the groups. The nematode community Structure Index (SI) was calculated for each sample after assigning c-p values to each genus as described by Ferris et al. (2001). Simpson’s index of diversity was calculated using genus-level abundance data (Washington, 1984). Protozoa were enumerated using the Most Probable Number technique (Rønn et al., 1995) in October of 2000 and 2001. The assays were performed in 24-well cell culture plates and the growth medium in each well was 0.8 mL autoclaved soil extract amended with 0.1% glucose. Four replicate 5-fold dilution series were prepared for each soil sample, starting with an initial 1:50 soil:sterile tap water suspension shaken at 240 rpm for 4 min. The plates were incubated at 20 °C. After 14 d, wells were scored for the presence of naked amoebae and flagellates using an inverted microscope. Tables of Woomer (1994) were used for estimating the most probable number of protozoa per gram of soil from the number of positive replicate wells at each dilution.

In 2004, root fragments were collected from each sample as it was sieved, and the root fragments were separated into greater than 2 mm and less than 2 mm diameter size fractions. The less than 2 mm diameter root fragments were chopped into 1- to 2-cm long pieces, washed over a 250-μm sieve with a stream of water, and subjected to root-lesion nematode extraction in a mist chamber for 7 d (Ingham, 1994). After nematode extraction, the root samples were air-dried and weighed. Soil samples collected Oct. 2004 were air-dried and subjected to a broad set of chemical analyses at a commercial laboratory (Norwest Laboratories, Lethbridge, Alberta, Canada).

All plant measurements and nematode community parameters were analyzed using a mixed model analysis of variance (ANOVA) for a randomized complete block design (PROC MIXED) in SAS (SAS Institute Inc., 2000). The ANOVA model had blocks as a random effects factor, mulch (present or absent) and amendment (present or absent) as fixed effects factors, and sample date or year factors were treated as repeated measures using the REPEATED statement in PROC MIXED. When main factor effects of mulch or amendment or their interaction with sample date were significant, the SLICE procedure was used to test the significance of differences between treatments at individual sample dates. Root-lesion nematode data were log-transformed before analysis to minimize heteroscedasticity. Soil chemical properties and cumulative yield were analyzed using PROC GLM for a two-factor randomized complete block experimental design.

Tissue nitrogen.

Increased leaf N concentration was associated with alfalfa mulch application (Tables 6 and 7). Increases were modest despite a significant pool of N contained within the mulch. Similarly, application of CDS amendment modestly affected leaf N (Table 6), increasing the main factor mean from 27.8 to 28.4 g·kg⁻¹ dry weight. A pattern of minimal but positive N effect of mulch was also apparent in fruit (Table 6). In general, N status was considered sufficient according to local leaf standards, regardless of treatment, with the lowest leaf N concentration measured in the non-mulched treatments at 25.5 g·kg⁻¹ in 2004. Furthermore, all plots were fertigated with 20 g N per tree annually over an 8-week period as calcium nitrate. The cumulative evidence therefore suggests that the modest augmentation of N nutrition associated with alfalfa mulch and CDS amendment was unlikely to have been the cause of improved vigor and yield associated with mulch and may also account for the lack of treatment effect on fruit color, which is usually associated with high N application rate. Similar minimal response of N-fertigated trees to N applied with biosolids has previously been reported (Neilsen et al., 2007). Our data suggest, however, that it may be possible to substantially reduce fertilizer inputs when alfalfa hay is used as mulch.

Table 6.

Summary of results of mixed model analysis of variance of N, P, K, Mg, and Ca concentrations in leaf and fruit tissues.

View Table

Table 7.

Effects of alfalfa hay mulch (N, K, Mg, Ca) and composted dairy manure solids amendment (P) on leaf N, P, K, Mg, and Ca of ‘Braeburn’ apple on M.9, 2000–04.

View Table

Tissue phosphorus.

Tree P uptake was most affected by CDS amendment (Table 6), which increased leaf P concentration in 4 of 5 years (Table 7) and fruit P concentration in all 4 years (Table 8). Mulch caused a modest additional increase in leaf P (Table 6). The differential uptake of P among treatments reflected the higher P inputs associated with the CDS amendment (Table 1) and the highest available soil P values for CDS-amended soils at the end of the experiment (Table 5). Application of monoammonium phosphate in the planting hole has been shown to stimulate apple tree growth in old orchard soils (Neilsen and Yorston, 1991) and was the rationale for fertigation of ammonium polyphosphate to all treatments in this experiment in Year 1. Highest leaf P concentrations were measured in the first year, but midsummer leaf P concentrations, regardless of treatment, generally exceeded 1.5 g P/kg values normally considered adequate for apples (Shear and Faust, 1980). Much less is understood about desirable fruit P concentrations, but in 2004, fruit P concentrations from unmulched and unamended trees dropped below values of 8.0 mg P/100 g FW. Recent research on several apple cultivars other than ‘Braeburn’ recommended harvest fruit P concentrations of 10 to 12 mg/100 g FW to optimize fruit quality (Neilsen et al., 2008). The lack of response of ‘Braeburn’ fruit quality to higher P regimes associated with the CDS amendment (Table 4) implies the P concentrations measured in our study were generally sufficient for all treatments. Nevertheless, CDS would be an effective source of P in apple orchards where P nutrition is a concern.

Table 8.

Effects of alfalfa hay mulch (N, K, Ca) and composted dairy manure solids amendment (P, Mg) on fruit N, P, K, Mg, and Ca of ‘Braeburn’ apple on M.9, 2001–04.

View Table

Tissue potassium, magnesium, and calcium.

Potassium nutrition was strongly influenced by application of both alfalfa mulch and CDS amendment. There was a significant mulch × amendment interaction for leaf K (Table 6) with leaf K increased in trees grown with mulch but with increases less pronounced in trees also receiving amendment (Table 7). Both mulch and CDS amendment increased fruit K to similar degrees (Tables 6 and 8). The increases in plant K status associated with both alfalfa mulch and CDS amendment are consistent with the increased extractable K content of soil measured for both treatments by the end of the experiment in 2004 (Table 5). Improvements in plant K status were sometimes associated with a decline in leaf and fruit Mg and Ca concentration. Decreased leaf Mg concentration was particularly associated with mulch (Tables 6 and 7). Fruit Mg concentration was unaffected by mulch but was higher for trees to which CDS was applied (Tables 6 and 8). Collectively the Mg results are consistent with increased soil-extractable Mg status measured for CDS-amended but not mulched soils by 2004 (Table 5). Leaf Ca concentration was decreased by both mulch and amendment (Tables 6 and 7). There was a significant mulch × amendment interaction for fruit Ca (Table 6) with mulch only causing a significant reduction in trees that had not been treated with CDS amendment (Table 8). These plant Ca results were independent of soil Ca status, which was unaffected by mulch and had a modest increase as a result of CDS application (Table 5).

Changes in leaf K, Mg, and Ca status observed during the study did not result in deficient levels of these nutrients (Shear and Faust, 1980) and hence were unlikely explanations for the improved yield and vigor associated with mulching. Improvements in leaf K as a result of alfalfa mulch are consistent with previous reports (Hogue and Neilsen, 1987; Neilsen et al., 2003a, 2003b) and imply that both alfalfa and CDS would be valuable additions for sites with low soil K. Magnesium nutrition was a concern of the cooperating grower at this site as indicated by the application of several foliar Mg applications in the block during the study. This may have overridden potential negative consequences to decreased leaf Mg associated with mulch because mulched trees grew better. High application of soil K and Mg to soil have been cited as a potential concern for fruit Ca accumulation and hence fruit quality (Faust, 1989). Both alfalfa mulch and CDS amendment were effective sources of either K (both) or Mg (CDS) and were sometimes associated with decreased fruit Ca concentration. However, in years (2001, 2003, and 2004) when fruit Ca concentration was below the critical 4.0 mg Ca/100 g FW threshold (Neilsen and Neilsen, 2003), all treatments in general had low Ca. Thus, effective foliar Ca applications would have been recommended for the whole block regardless of mulch or amendment treatment (Peryea et al., 2007). This was also consistent with the lack of fruit quality effects associated with treatments.