Soil and Plant Response to Applications of Municipal Solid Waste Compost and Fertilizer to Willamette Raspberries

Abstract

An experiment was initiated in 1997 to compare two application rates of municipal solid waste (MSW) compost with commercial fertilizer for the cultivation of “Willamette” raspberries. Mehlich 3 extractable soil elements, plant leaf tissue, and fruit elemental analysis were assessed from 2000 to 2002. The lower rate of compost was applied to provide the same amount of N used for the fertilizer treatment (135 kg N ha^-1); the higher rate of MSW was three times the low rate. Fruit yields increased from 2000 to 2002; the fertilizer produced the highest yields all three years. In the last year of the study, M-3 soil Ca and Zn were higher in the MSW plots than the fertilizer plots; leaf N and Mn were highest in the fertilizer plot while leaf P was lowest in this plot. Fruit N and P were significantly higher in municipal solid waste (MSW3) plots than the fertilizer treatment while the Mn content of the fruit followed the opposite trend. It was determined that neither plants nor fruit accumulated potentially toxic trace metals; therefore, the MSW applications would not pose a threat to human health.

KEYWORDS:

• fruit elements
• leaf nutrients
• MSW compost
• raspberries
• soil fertility

INTRODUCTION

Composting of source-separated municipal solid waste (MSW) is a legislated practice in Nova Scotia in order to divert refuse from landfills, water courses and incinerators. As more communities produce MSW compost there is increased interest in its application to agricultural land including high value horticultural crops. The positive aspects of using MSW compost may outweigh its negative aspects due to its ability to increase soil organic matter, water holding capacity, plant nutrients and soil fertility (He et al., 1995; Hargreaves et al., 2008).

The application of various organic mulches (bark, sawdust, shredded wastes) is a common practice for berry crop production. Few studies, however, have been published on the application of organic fertility amendments to raspberries. Cheng (1982), Dean et al. (2000), and Gercekcoglu (2008) used farmyard, poultry and sheep manure, respectively, to grow raspberries in Canada and Turkey. The use of composts as fertilizers has become more common; Atiyeh et al. (2000) used various composts to grow raspberry plants in greenhouse trials, whereas, Hargreaves et al. (2008a) used two types of composts and their water-extracted teas to grow raspberries in the field.

Several studies have shown that the use of MSW compost in horticulture has many benefits to soil and crops and no negative effect on the environment (Hargreaves et al., 2008a, 2008b; Hicklenton et al., 2001; Warman and Shanmugam, 2008; Warman et al., 2004; Zheljazkov and Warman, 2004). Therefore, the objectives of this experiment were to compare two different application rates of MSW compost with commercial fertilizer by examining the following parameters: yield, extractable soil elements, and leaf and fruit element concentrations.

MATERIALS AND METHODS

A field experiment was conducted at Lower Onslow, Nova Scotia, for five years beginning in 1997. Eleven “Willamette” raspberry canes were planted in rows 3 m apart and spaced 50 cm apart within the row to form nine plots 6.5 m long. The soil type was a Pugwash sandy loam (Humo-Ferric Podzol), considered a CLI class two agricultural soil with limitations due to adverse climate and low fertility. The municipal solid waste compost (MSW) was produced under aerobic conditions using the in-vessel system at the Lunenburg Regional Recycling and Composting Facility in Wynott Settlement, NS, Canada. Source-separated MSW is brought in, shredded and placed into the vessel. The composting process takes at least three weeks, after which, the organic material is taken outside for final maturing (3–4 months) in windrows. Throughout this phase, the temperature and moisture were monitored and the material was turned at least twice a week. The chemical analysis of the MSW compost is shown in Table 1; some chemical characteristics of the soil are as follows: pH 5.8, 16 g organic C kg⁻¹, 50 mg P kg⁻¹, 62 mg K kg⁻¹, 750 mg Ca kg⁻¹, 160 mg Mg kg⁻¹, and 15 mg S kg⁻¹.

TABLE 1 Dry Weight Analysis of MSW Compost Macronutrients, Micronutrients and Trace Elements

Nutrients	Units	1999	2000	2001	2002
N	g kg^−1	30.0	23.2	25.5	27.1
P		9.9	6.0	6.3	8.3
K		5.6	7.8	8.1	9.1
Ca		50	25	26	78
Mg		4.5	2.6	2.9	6.6
S		3.6	3.1	6.4	4.6
Fe		1.2	6.8	7.9	9.5
Cu	mg kg^−1	270	99	104	99
Mn		1910	870	884	1135
Zn		598	206	213	218
B		27	12	18	29
Cd		0.7	0.4	0.2	1.4
Cr		33	15	14	12
Ni		21	9	12	7
Pb		205	68	70	63

Treatments consisted of one rate of chemical fertilizer (N-P₂0₅-K₂0), based on soil samples taken in1997 which indicated the need for 135-230-230 kg/ha⁻¹ according to the Nova Scotia Soil Test Recommendations, and two rates of MSW compost (MSW1 and MSW3). Beginning in 1997, all amendments were surface applied to the plots twice a year; the fertilizer was 65-90-90 and 70-140-140 kg N ha⁻¹, spring and fall, respectively. The MSW1 rate was also based upon 65 and 70 kg N ha⁻¹; the MSW3 rate of compost was three times the MSW1 rate, chosen on the assumption that 33% of the total N in the compost would be plant available. Compost was dried and analyzed for C and N using the Leco CNS-1000 Analyzer (Leco Corporation, St. Joseph, MI). Moist compost was applied based on its dry weight total N content.

Fruit harvest began at the end of July 1998 and continued until the end of August in all years. Fruit yield and elemental analysis data for soil, tissue, and fruit are presented in this paper for the years 2000 to 2002. Leaf and soil sampling was performed during the harvest period. Three to five soil cores (10–15 cm deep) were taken from each raspberry plot, bulked, and analyzed as a single composite. Approximately fifteen raspberry floricane (fruiting canes) or primocane (vegetative canes) leaves per plot were sampled along the top 30 cm of the canes in August and September, respectively, following the suggestion of John et al. (1976) that leaf composition was relatively stable during the early fruit harvesting period.

Soil, Tissue, and Fruit Analysis

Soil mineral elements (P, K, Ca, Mg, S, Na, Fe, Cu, Mn, Zn, B, Cd, Cr, Ni, and Pb) were extracted using the Mechlich-3 extractant (M-3) and determined using Inductively Coupled Argon Plasma Emission Spectroscopy (Thermo Jarrell Ash ICAP 1100, Thermo Jarrell Corp., Waltham, MA). All plant tissue was rinsed with distilled water and dried at 65°C for 48 h. The dried leaves and fruit were ground and digested with nitric acid according to the method of Zheljazkov and Warman (2002). The digests were analyzed for the same elements as the soil using ICAP. The total weight of all berries harvested from the plots every two days was recorded for calculation of total fruit yield (g m⁻²). A random sample of fruit from each plot at each picking was frozen and then a sample of the composite was taken, dried and analyzed.

Statistical analysis was completed using the SAS System software package (SAS Institute, 1999). Following verification of assumptions, a one-way ANOVA, at a probability level of 0.05, was used to determine if the treatments produced a significant effect for that year; if significant, treatment means were compared using a Tukey's Means Comparison Test (p = 0.05).

RESULTS AND DISCUSSION

In 2000, the highest levels of extractable P, Ca, Cu and Zn were found in soil from plots treated with MSW3 (Tables 2 and 3). However, the P level in MSW3-treated plots was statistically similar to plots treated with fertilizer. The level of Zn in soil treated with MSW1 was intermediate between the other two treatments. In 2001, there were higher levels of extractable Ca and S in soil from plots treated with MSW3; Ca level in plots treated with MSW1 was also significantly higher than plots treated with fertilizer. Mean levels of M-3 P, K, Fe and Cu in soils of all three treatments were lower in 2001. In 2002, only Ca and Zn levels were statistically higher in both of the MSW-treated soils (these were similar to each other) than the fertilizer-treated soil. Although none were statistically tested, the following trends over time can be seen in extractable levels of nutrients: P displayed an increasing trend; Mg and S showed a decreasing trend.

TABLE 2 Mehlich 3 Soil Macronutrient Contents (mg kg⁻¹)

Year	Nutrient	MSW1	Fertilizer	MSW3	Mean	Std. Dev.
2000	P	57a	99ab	114b
2001		61	72	83	73	19
2002		147	92	138	126	71
2000	K	66	101	109	92	23
2001		77	82	48	68	32
2002		76	92	78	82	30
2000	Ca	1834b	1225a	2254c
2001		1941b	1014a	2699b
2002		1639b	629a	1587b
2000	Mg	314	222	292	276	48
2001		206	152	254	211	55
2002		197	111	193	167	59
2000	S	56	52	68	59	8
2001		23a	20a	42b
2002		14	14	15	14	2
Means with different letters in a row are significantly different at p < 0.05.

TABLE 3 Mehlich 3 Soil Micronutrient Contents (mg kg⁻¹)

Year	Nutrient	MSW1	Fertilizer	MSW3	Mean	Std. Dev.
2000	Fe	238	267	243	249	16
2001		70	74	58	67	9
2002		111	142	124	126	24
2000	Cu	2.7a	1.4a	4.3b
2001		1.8	0.8	0.4	1.0	0.8
2002		2.0	1.0	1.0	1.3	1.0
2000	Mn	45	43	47	45	2
2001		28a	20a	72b
2002		63	40	62	55	16
2000	Zn	9b	4a	15c
2001		16b	4a	13b
2002		13b	3a	8b
2000	B	0.7	0.4	1.1	0.7	0.4
2001		1.0	0.6	0.6	0.7	0.4
2002		0.7	0.1	0.5	0.4	0.3
Means with different letters in a row are significantly different at p < 0.05.

Floricane leaf tissue nutrient levels were only affected in a few cases. In 2000, only Ca was found to be significantly higher in MSW1-treated plots than the MSW3 treatment, with fertilizer at an intermediate level (Table 4). In both 2001 and 2002, higher P levels were found in both MSW treatments than in plots treated with fertilizer (Table 4), while the opposite was true for Mn, in which levels were twice as high in fertilizer plots as the MSW-treated plots (Table 5). Over the three years, there was an increasing trend in P and a decreasing trend in S content, although neither of these was tested for significance (Table 4). Primocane leaf macro- and micronutrient tissue levels were analyzed only in 2000 and only Mn was found significantly higher in fertilizer plots than either of the MSW plots (Table 6). A comparison of primocane leaf tissue macronutrient levels from 2000 with those of the same year's floricanes revealed similar numbers with the exception of Ca, which appeared to be much lower in the fruiting canes (Tables 4 and 6), and Mg, which was a little higher in primocane (5.5) than floricane (4.5) leaf tissue. The same comparison for micronutrients showed little difference for Cu, Mn and Zn, but appreciable differences for Fe and B, which were both at lower levels in fruiting canes (Tables 5 and 6).

TABLE 4 Floricane Leaf Macronutrient Contents (g kg⁻¹)

Year	Nutrient	MSW1	Fertilizer	MSW3	Mean	Std. Dev.
2001	N	23	23	22	23	2
2002		17a	19b	18a
2000	P	2.4	1.9	2.3	2.2	0.3
2001		2.6b	1.9a	2.6b
2002		2.8b	2.0a	2.9b
2000	K	8	10	9	9	1
2001		6	9	9	8	2
2002		5	9	9	8	3
2000	Ca	11.5a	9.5ab	9.3b
2001		13.0	11.0	12.0	12.0	2
2002		13.0	12.0	12.0	12.3	0.7
2000	Mg	5.4	4.2	4.0	4.6	0.8
2001		4.4	3.7	3.9	4.0	0.6
2002		5.0	4.0	4.0	4.3	1
2000	S	3.0	3.0	2.7	2.9	0.2
2001		1.0	1.0	1.2	1.1	0.2
2002		0.8	1.0	0.8	0.9	0.07
Means with different letters in a row are significantly different at p < 0.05.

TABLE 5 Floricane Leaf Micronutrient Contents (mg kg⁻¹)

Year	Nutrient	MSW1	Fertilizer	MSW3	Mean	Std. Dev.
2000	Fe	62	53	54	56	5.1
2001		97	108	92	99	16
2002		92	90	87	90	15
2000	Cu	7.0	7.0	6.5	6.8	0.3
2001		4.0a	4.0a	5.8b
2002		5.5	5.8	5.3	5.6	0.7
2000	Mn	80	164	85	110	47
2001		74a	182b	71a
2002		80a	259b	64a
2000	Zn	42	16	16	25	15.3
2001		18	15	18	17	4
2002		20	20	21	20	2
2000	B	43	32	38	38	5.6
2001		54	47	51	51	14
2002		56	40	50	49	12
Means with different letters in a row are significantly different at p < 0.05.

TABLE 6 Primocane Leaf Macro- and Micronutrient Contents in 2000

Nutrient	Units	MSW1	Fertilizer	MSW3	Mean	Std. Dev.
P	gkg^−1	2.1	1.7	2.2	2.0	0.3
K		6.8	8.4	9.0	8.1	1.1
Ca		15.0	13.7	14.3	14.3	0.7
Mg		6.2	5.1	5.3	5.5	0.6
S		3.1	3.5	3.2	3.3	0.2
Fe	mg kg^−1	90.0	96.1	83.2	89.8	6.5
Cu		8.4	7.4	7.6	7.8	0.5
Mn		68a	219b	51a
Zn		22.0	21.0	25.7	22.9	2.5
B		58.4	45.1	65.5	56.3	10.4
Co		0.11	0.18	0.25	0.18	0.07
Means with different letters in a row are significantly different at p < 0.05.

John et al. (1976) wrote that leaf concentrations of N, P, K, Mg, S, and Cu were higher in primocane leaves, whereas leaves from fruiting canes contained higher levels of Mn, Ca, Fe, Zn, Al and B. Kowalenko published two papers (1981 and 2006) from his work in British Columbia where he examined the effect of N and B fertilizer applications to cultivar. Willamette leaf N, B, and Mn content. Using 134 kg ha⁻¹, he found higher leaf N (both floricane and primocane) levels than the present study, and considerably lower B and Mn than reported here (Kowalenko, 2006). In the present study, the relationships were reversed; lower leaf N levels were associated with higher B and Mn levels (Tables 5 and 6). Normally, one associates higher leaf B or Mn with lower pH soils; however, the soil at Lower Onslow was at a higher pH throughout the study (about 5.8 for the fertilizer plots) than what is evident in the Kowalenko studies.

Fruit nutrient content was not assessed in 2000. In 2001, the Ca level in the berries was significantly higher in MSW3-treated plots than the other two treatments (Table 7) and Fe was significantly higher in fertilizer–treated fruit than either of the two MSW-treated plots (Table 8). However, in 2002, the Fe level was numerically much higher in the MSW3 treatment, but not statistically so, which can be explained by high field variability. Fruit Mn content was found to be significantly higher in fertilizer-treated plots than either of the MSW-treated plots in both years (Table 8). Also in 2002, fruit N was significantly higher in MSW3 plots than the other two treatments and fruit P was higher in both MSW treatments than the fertilizer treatment (Table 7).

TABLE 7 Fruit Macronutrient Content (g kg⁻¹) in 2001 and 2002

Year	Treatment	MSW-1	Fertilizer	MSW-3	Mean	S.D.
01	N	9.1	9.9	8.6	9.2	0.1
02	N	7.0a	7.0a	9.0b
01	P	1.6	1.6	1.8	1.7	0.1
02	P	1.9b	1.5a	1.8b
01	K	7	8	8	7	1
02	K	7	8	8	7	1
01	Ca	1.0a	1.0a	1.2b
02	Ca	1.2	1.1	1.2	1.2	0.1
01	Mg	0.8	0.8	0.9	0.8	0.1
02	Mg	0.9	0.8	0.9	0.9	0.1
01	S	0.5	0.6	0.5	0.5	0.1
02	S	0.8	0.5	1.2	0.8	0.4
01	Na	6	11	9	8	2
02	Na	11	15	25	17	10
Means with different letters in a row are significantly different at p < 0.05.

TABLE 8 Fruit Micronutrient Content (mg kg⁻¹) in 2001 and 2002

Year	Treatment	MSW-1	Fertilizer	MSW-3	Mean	S.D.
01	Fe	26a	38b	16a
02	Fe	46	44	70	53	31
01	Cu	4	4	5	5	1
02	Cu	5	4	9	6	3
01	Mn	14a	24b	12a
02	Mn	12a	21b	11a
01	Zn	15	16	15	15	2
02	Zn	16	15	15	15	4
01	B	8	11	9	9	2
02	B	7	9	8	8	2
Means with different letters in a row are significantly different at p < 0.05.

In the last year of the study (2002), trace metal contents were determined and only MSW3 soil Cd was found to be highest in MSW1 plots (Table 9), although the Cd content in leaf tissue did not reflect this result. Based upon the results shown in Table 9, there is little indication that the plants or the fruit have accumulated Cd, Cr, Ni, or Pb.

TABLE 9 Trace Element Content (mg kg⁻¹) in Soil, Floricane Leaf Tissue, and Fruit in 2002

	Treatment	Cd	Cr	Ni	Pb
Soil	MSW-1	0.06b	0.18	0.25	0.8
	Fertilizer	0.02a	0.18	0.27	0.3
	MSW-3	0.03a	0.20	0.25	0.7
	Mean		0.19	0.26	0.6
	S.D.		0.03	0.11	0.3
Leaf tissue	MSW-1	0.70	0.60	0.38	1.6
	Fertilizer	0.69	0.39	0.30	1.3
	MSW-3	0.34	0.31	0.33	1.1
	Mean	0.58	0.43	0.34	1.3
	S.D.	0.33	0.39	0.45	0.5
Fruit	MSW-1	0.3b	0.59	0.65	0.51
	Fertilizer	0a	1.13	1.37	0.86
	MSW-3	0.21b	0.88	0.71	0.44
	Mean		0.87	0.91	0.61
	S.D.		0.45	0.47	0.42
Means with different letters in a row are significantly different at p < 0.05.

Berry yields (per meter²) in each treatment increased from year to year, as seen in Fig. 1. When data were analyzed for treatment effects within each year, a significant effect was found only in 2001, when the MSW1 treatment resulted in a significantly lower yield than either of the other two. Even so, there was clearly a similar yield trend among the three treatments in each of the years: fertilizer>MSW3>MSW1. The results strongly suggest that the MSW3 rate or higher is required to provide the equivalent amount of available nutrients as the fertilizer rate. Based solely on N and the mineralization of the organic N in compost, it would appear that raspberries take up somewhat less than 33% of the surface applied organic N.

FIGURE 1 Mean fruit yields (grams meter²) 2000–2002.

Hargreaves et al. (2008a) study lends itself to a number of comparisons with results from Lower Onslow since “Willamette” raspberry and MSW compost from the same facility were used. In that study, the application of compost was weighed to provide 135 kg N ha^-1, the same application rate used in the present study's MSW1 treatment. In both studies, floricane N decreased appreciably from year to year by a similar amount (5 g kg^-1) and a similar decrease over time in primocane leaf tissue N was also seen in the Boutiliers Point study. Year to year increases in P were also seen in both studies; however, contrasting patterns were seen for K, with both increases and decreases found at both sites over the course of the studies. Values fell within a similar range with one exception, that being the third year in Boutiliers Point, when primocane K reached 13.8 g kg⁻¹. A strongly contrasting pattern was shown for Ca in which floricanes had lower Ca at Lower Onslow while primocanes had lower Ca at Boutiliers Point. A somewhat similar contrast was seen with Mg in which primocane levels in Lower Onslow were higher than floricane levels (in 2000) and floricane levels in Boutiliers Point were higher than primocane levels. Floricane S decreased from year to year at Lower Onslow while it remained the same at Boutiliers Point.

Fruit yields in the Hargreaves et al. (2008a) study at Boutiliers Point using MSW, for the best year of the three reported, were similar to the MSW1 plot yields in this study for 2002. Unlike this study, where yields increased every year from 2000 to 2002, yields declined at Boutiliers Point as that study progressed.

CONCLUSIONS

Fruit yields increased from 2000 to 2001 to 2002, and the recommended rate of commercial fertilizer produced the highest yields all three years. Based upon the yield response, one could conclude that N mineralization and uptake from the MSW is somewhat less than 33% of the total applied N. In the last year of the study, M-3 soil Ca and Zn were higher in the MSW plots than the Fertilizer plots; leaf N and Mn was highest in the Fertilizer plot while leaf P was lowest in this plot. Nonetheless, fruit N and P was significantly higher in MSW3 plots than the fertilizer treatment while the Mn content of the fruit followed the opposite trend. Thus, the only consistent response was the association of higher leaf and fruit Mn with the use of fertilizer, probably associated with the fact that MSW increases soil pH and thus reduces Mn availability to plants (Hargreaves et al., 2008; Warman and Shanmugam, 2008). Lastly, neither the plants nor the fruit accumulated potentially toxic trace metals; therefore, the MSW applications would not pose a threat to human health.

Notes

¹Funding for this study was provided by the Natural Science and Engineering Research Council (NSERC) of Canada. The author is indebted to many people who helped maintain and harvest the plots, perform soil, fruit and plant analyses, evaluate the data and edit the manuscript.