Comparison of Split or Single Application of Gypsum for Wild Lowbush Blueberry (Vaccinium angustifolium Ait.)

Abstract

Gypsum (CaSO₄·2H₂O) was evaluated as a single pre-emergent application (4,482 kg/ha) or a split application (2,242 kg/ha pre-emergent and 2,242 kg/ha 2 weeks later) for its effect on soil nutrient release, nutrient uptake, and plant growth at two commercial lowbush blueberry fields (Clary Hill and Marshville, Maine). Gypsum applications were compared to diammonium phosphate (448 kg/ha) and a control using 0.9 m × 15 m plots arranged in a randomized complete block design with six blocks. Composite leaf and soil samples were collected within each treatment plot in July 2009 and analyzed for nutrient concentrations. Within each treatment plot, stems from four randomly placed 0.023 m² quadrats were collected in October 2009 for stem length, branching, and flower bud measurements. Gypsum raised soil Ca and S concentrations at both fields compared to diammonium phosphate and the control. One of the fields (Marshville) was deficient in leaf N and P. Diammonium phosphate raised leaf N and P concentrations compared to the controls at both fields; but gypsum (single or split application) raised leaf N and P only at the deficient Marshville field. Diammonium phosphate increased stem branching, number of branches, length of branched stems, and average stem length at both fields. Gypsum treatments did not affect stem characteristics at either field except for stem branching, which was raised by the split application at the deficient Marshville field. Blueberry yield was increased by diammonium phosphate at Clary Hill and Marshville by 57 and 117%, respectively, compared to the control. Neither of the gypsum treatments increased yield. A split application of gypsum was not more effective than a single application.

Keywords:

• fertility
• mineral nutrition
• calcium sulfate
• production
• diammonium phosphate

INTRODUCTION

Wild lowbush blueberry (Vaccinium angustifolium Ait.) production follows a unique two-year production cycle due to the practice of biennial aerial shoot pruning to promote plant vigor. Wild blueberry growers in Maine have relied primarily on weed control, and prune-year fertilization to stimulate vigorous growth and flower bud formation to ultimately improve berry yield for the following crop year. Blueberry growers are encouraged to take leaf tissue samples and to fertilize the field in the next prune year with diammonium phosphate (DAP) if N and P deficiency is found (Smagula and Dunham, 1995). This has resulted in increased blueberry yield only when a deficiency was corrected. Application of DAP, when the field does not have a deficiency of both leaf N and P, can result in an accumulation of either soil N or P. There is a concern about the potential problem of surface runoff of dissolved reactive phosphorus (DRP) and ammonium nitrogen (NH₄-N) into nearby streams, and enhanced weed growth due to excessive N fertilization.

There is evidence that application of gypsum effectively reduces surface runoff of DRP and ammonium nitrogen (NH₄-N) by 80 and 59%, respectively (Favaretto et al., 2006) due to the conversion of readily desorbable soil P into a less soluble Calcium-P complex (Traynor, 1980). Gypsum is a mineral that is capable of releasing soil nutrients bound in the soil cation exchange sites. Calcium (Ca⁺⁺) ions from gypsum replaces the lower charged ions (NH₄ ⁺, K⁺) making them available for root uptake. Since gypsum is primarily made up of Ca and sulfate (SO₄ ²⁻), its use as a fertilizer will not add unnecessary phosphorus (P) and nitrogen (N) in the soil causing DRP and NH₄-N runoff. Sanderson et al. (1996) tested the efficacy of gypsum (CaSO₄; 4,000 kg·ha⁻¹) on wild blueberry in Prince Edward Island and found an average of 47% increase in lowbush blueberry yield in three out of six fields. In addition, gypsum is currently recognized as an organic fertilizer by the Maine Organic Farmers and Growers Association. The recommended rate of gypsum for wild lowbush blueberry fields in Canada is 4,000 kg·ha⁻¹ (Sanderson et al., 1996). The effect of applying this rate of gypsum as a split application has not been tested.

The objectives of this study are (1) to evaluate the efficacy of gypsum in correcting leaf nutrient deficiency under Maine soil conditions compared to conventional DAP fertilizer, and (2) to determine if split application of gypsum is more effective than a single application in releasing soil nutrients and improving plant growth, nutrition, and yield of wild lowbush blueberry.

MATERIALS AND METHODS

Two commercial lowbush blueberry fields (Clary Hill and Marshville, Maine) in the prune year cycle were used in this study. At each field, treatment plots measuring 0.91 m × 15.2 m with 1.8 m alleyways between plots were established. Treatments included a control plot, diammonium phosphate (DAP; 18–46–0) applied before the aerial shoots emerge from the soil (preemergent) at the standard rate of 448 kg/ha and Gypsum applied pre-emergent at 4,484 kg/ha or as a split application of 2,242 kg/ha (pre-emergent) + 2,242 kg/ha (post-emergent, about 3 weeks later). Treatments were applied using a Gandy Model 42 drop spreader (Gandy Company, Owatonna, MN, USA).

Clary Hill Field

Pre-emergent applications of the single application of DAP and gypsum were made on May 15, 2009, while the second half of the split application (2,242 kg/ha post-emergent) was applied on June 3, 2009. The second half of the split application was applied 2 weeks after the first rain (>1.1 cm). Composite leaf samples were taken from 50 randomly selected stems in each treatment plot on July 6, 2009 after the stems had stopped elongating, the tip dieback stage, when changes in leaf nutrient concentrations are minimal (Trevett et al., 1968). Leaf samples were analyzed for leaf nutrients at the Maine Agriculture and Forestry Experiment Station Analytical Laboratory, Orono, Maine. Leaf tissue samples were dried at 70°C and ground to pass through a 40-mm mesh sieve. All leaf nutrient concentrations except nitrogen were measured at the Maine Agriculture and Forestry Experiment Station Analytical Laboratory, Orono, Maine using inductively coupled plasma emission spectroscopy (ICP-OES) following the procedure of Kalra and Maynard (1991).Ten soil sample cores within each treatment plot were collected on July 6, 2009 using a standard soil sample tube (2 cm in diameter) inserted to a depth of 7.6 cm. Soil sampling was done following the leaf samples to determine soil pH, organic matter, cation exchange capacity, and soil nutrient concentrations. Soil organic matter content was measured by loss on ignition (LOI) at 375°C. Soil nutrients were extracted in pH 4.8 ammonium acetate (Modified Morgan method) and measured using an ICP-OES. Stem density, height, branching, and flower bud formation were measured from stems collected on October 26, 2009 from within four 0.023 m² quadrats per treatment plot. Berry yield was estimated on August 6, 2010 by harvesting a 0.61 m strip in the 0.91 m wide plots using a mechanical harvester.

Marshville Field

Gypsum and DAP were applied pre-emergent on May 22, 2009. Composite leaf samples from 50 randomly selected stems in each treatment plot were collected on July 15, 2009 for leaf nutrient concentrations; on the same day, 10 soil sample cores were collected within each treatment plot and analyzed as described above. Stems within four 0.023 m² quadrats per treatment plot were collected on October 30, 2009 for the measurement of stem characteristics. Blueberry yield was collected on July 28, 2010 using a mechanical harvester.

Statistical Analysis

Each treatment was assigned in a randomized complete block with six blocks. Data were subjected to Analysis of Variance (ANOVA) using SAS General Linear Model (SAS Institute Inc., 2006). Data were tested for assumptions of normality and homoscedasticity and transformed when appropriate. Percent branching was analyzed as its arcsine, but is presented as original percentage. Significant treatment effects were determined by the Ryan-Einot-Gabriel-Welsch (REGW) Multiple Range Test at the 5% significance level.

RESULTS AND DISCUSSION

Clary Hill Field

Soil analysis

Soil pH was measured in calcium chloride (CaCl₂) because gypsum-treated soils have a high electrical conductivity due to the high rate of gypsum applied and this affects the pH electrodes. In another gypsum study, pH measurement in water and CaCl₂ extractant were compared and revealed that the CaCl₂ was more appropriate (Santiago and Smagula, 2011). The pH of soil sampled from gypsum-treated plots was not meaningfully raised (Table 1). The slight increase of soil pH will not likely affect blueberry growth since the soil pH in gypsum-treated plots were within the optimum soil pH range for wild lowbush blueberry. Soil organic matter (LOI) in this field ranged from 11.9 to 12.8%.

TABLE 1 Effect of Single or Split Application of Gypsum and DAP on Soil Nutrient Concentrations at Two Maine Wild Blueberry Fields

Treatments	Rates (kg·ha^−1)	pH (CaCl2)	OM (%)	Ca (ppm)	S (ppm)	Mg (ppm)	K (ppm)	P (ppm)	Mn (ppm)	Zn (ppm)
CLARY HILL
Control	0	3.39 ^z b	12.83 a	206.68 ^y b	163.83 ^x b	33.41 a	95.38 a	15.74 ab	44.75 a	2.38 a
DAP	448	3.88 b	11.90 a	209.47 b	204.51 b	35.89 a	87.18 b	16.54 a	38.17 ab	2.03 ab
Gypsum (single)	4484	4.10 a	12.02 a	616.48 a	414.05 a	22.97 b	86.21 b	13.16 c	21.32 c	1.80 b
Gypsum (split)	2,242 × 2	4.08 a	12.20 a	729.98 a	488.48 a	24.18 b	95.85 a	13.56 bc	28.12 bc	2.03 ab
MARSHVILLE
Control	0	3.73 b	14.33 a	413.94 b	25.87 b	78.63 a	66.75 a	14.11 b	27.03 a	2.95 ab
DAP	448	3.69 b	15.52 a	458.63 b	27.93 b	85.81 a	78.46 a	20.24 a	26.38 a	3.42 a
Gypsum (single)	4484	3.87 a	14.43 a	880.20 a	278.07 a	36.32 b	67.02 a	13.77 b	17.36 b	2.61 b
Gypsum (split)	2,242 × 2	3.89 a	16.30 a	1010.92 a	266.81 a	42.98 b	82.03 a	15.14 b	25.81 a	3.20 ab
^zMeans within a column in each location not followed by the same letter are significantly different at P < 0.05 level using Ryan-Einot-Gabriel-Welsch (REGWQ) Test.
^yCa data at the Clary Hill field was log transformed for analysis but are presented as original means in the table.
^xS data at both fields were log transformed for analysis but are presented as original means in the table.

Both soil Ca and Sulfur (S) were raised by the gypsum treatments compared to the control and DAP. In contrast, soil magnesium (Mg) was decreased by both gypsum treatments compared to the control and DAP. One of the concerns regarding heavy use of gypsum is the leaching of soil Mg. Several studies (Syed-Omar and Sumner, 1991; Alva et al., 1998; Caires et al., 2002) reported induced Mg deficiency due to high application rates of gypsum. In our study, the split application had the same effect as the single application on reducing soil Mg concentrations. Split application of gypsum had no effect on soil potassium (K) concentration compared to the control, but both the single application of gypsum and DAP treatments decreased soil K. Application of DAP did not increase the soil phosphorus (P) concentration compared to the control. However, the single application of gypsum reduced the soil P concentration while the split application had no effect. Traynor (1980) reported that heavy gypsum application ties up soil phosphorus by creating an insoluble Ca-P complex. In a recent study, Favaretto et al. (2006) reported a decrease in dissolved reactive phosphorus (DRP) with application of gypsum in a Miami silt loam soil and postulated that this was due to the formation of a less readily soluble Ca-P complex. These studies confirm our assumption that the free Ca from gypsum forms an association with the displaced P from the exchange sites. Both gypsum treatments decreased soil manganese (Mn) concentrations compared to the control but DAP did not. Only the single application of gypsum reduced soil Zn concentration compared to the control.

Leaf analysis

Plants in the control plots at Clary Hill were not deficient in N (>1.60%) or P (>0.125%) (Table 2). Application of DAP raised leaf N and P concentrations but gypsum as a single or split application had no effect on these leaf nutrient concentrations. In contrast, application of DAP resulted in lower leaf Ca concentration compared to the control while both gypsum treatments had no effect. Reduction in leaf Ca concentration with application of DAP could be due to the presence of NH₄ ⁺ ions in the soil, which has an antagonistic effect on the uptake of Ca (Mills and Jones, 1996); or it could be due to a dilution effect resulting in more or larger leaves stimulated by the N in DAP. None of the treatments affected leaf K, Mg, Fe, B, Al, or Mn concentrations. With the exception of Fe and B, Clary Hill was not deficient in any of these nutrients. Induced Mg deficiency by gypsum could be a concern with gypsum use in this field; both gypsum treatments reduced the soil Mg concentrations and consequently the leaf Mg concentrations (Table 2) below the Trevett (1972) standard for Mg (0.13%).

TABLE 2 Effect of Single or Split Application of Gypsum and DAP on Leaf Nutrient Levels and Yield of Two Wild Blueberry Fields in Maine

Treatments	Rates (kg·ha^−1)	N (%)	P (%)	Ca (%)	K (%)	Mg (%)	Fe (ppm)	B (ppm)	Al (ppm)	Mn (ppm)	Yield (kg·ha^−1)
CLARY HILL
Control	0	1.94 ^z b	0.154 b	0.436 a	0.532 a	0.131 a	40.06 a	21.78 a	120.97 a	2,275.0 a	5,240.0 b
DAP	448	2.27 a	0.180 a	0.374 b	0.552 a	0.119 a	40.42 a	23.87 a	77.32 a	2,111.7 a	8,248.0 a
Gypsum (single)	4,484	1.92 b	0.159 b	0.468 a	0.546 a	0.118 a	40.45 a	21.42 a	91.90 a	2,246.7 a	5,492.0 b
Gypsum (split)	2,242 × 2	2.00 b	0.163 b	0.467 a	0.558 a	0.123 a	42.63 a	21.60 a	105.03 a	2,153.3 a	5,313.0 b
MARSHVILLE
Control	0	1.47 b	0.116 b	0.459 ab	0.393 b	0.176 a	24.62 a	23.93 b	72.87 ab	1,116.2 bc	3,594.1 b
DAP	448	1.67 a	0.140 a	0.402 b	0.440 a	0.162 ab	26.63 a	27.02 ab	64.03 b	958.7 c	7,810.2 a
Gypsum (single)	4,484	1.56 a	0.130 a	0.524 a	0.432 a	0.163 ab	26.23 a	27.55 ab	78.52 a	1,420.2 ab	4,257.3 b
Gypsum (split)	2,242 × 2	1.60 a	0.134 a	0.528 a	0.432 a	0.153 b	27.65 a	28.97 a	71.37 ab	1,581.7 a	4,990.5 b
^zMeans within a column in each location not followed by the same letter are significantly different at P < 0.05 level using Ryan-Einot-Gabriel-Welsch (REGWQ) Test.

Stem characteristics

None of the treatments had any effect on stem density (Table 3). The number of shoots emerging from the rhizome is not usually affected by fertilizer the first year of application. However, it is possible that application of DAP every cycle might increase the stem density in the long term. Application of gypsum did not have any treatment effect on the average stem length, branched or unbranched stems length, branching, or number of branches, compared to the control. Similar results were reported by Sanderson et al. (1996) on the total number of stems, branched and unbranched stems, and average stem length. Neither gypsum nor DAP treatments had any effect on flower bud density, number of flower buds in branched and unbranched stems, or average flower bud per stem (Table 4).

TABLE 3 Effect of Single or Split Application of Gypsum and DAP on Wild Blueberry Stem Characteristics at Two Maine Wild Blueberry Fields

Treatments	Rates (kg·ha^−1)	Stem density (number per 0.023 m^2)	Average stem length (cm)	Branched stem length (cm)	Unbranched stem length (cm)	Percent branching	Number of branches
CLARY HILL
Control	0	41.33 a ^z	10.16 ^y b	11.82 ^y b	8.84 ab	35.82 ^x b	0.764 ^y b
DAP	448	38.67 a	11.83 a	13.90 a	9.76 a	54.54 a	1.469 a
Gypsum (single)	4,484	40.58 a	10.05 b	11.79 b	8.51 b	41.82 b	0.774 b
Gypsum (split)	2,242 × 2	41.13 a	9.79 b	11.37 b	8.11 b	44.43 b	0.919 b
MARSHVILLE
Control	0	33.29 a	8.89 ^w b	8.88 b	8.89 b	32.23 c	0.691 ^w b
DAP	448	32.92 a	11.30 a	11.68 a	10.88 a	50.43 a	1.207 a
Gypsum (single)	4,484	32.83 a	9.27 b	9.07 b	9.86 ab	38.66 bc	0.765 b
Gypsum (split)	2242 × 2	33.38 a	9.79 b	9.61 b	10.03 ab	44.66 ab	0.960 ab
^zMeans within a column in each location not followed by the same letter are significantly different at P < 0.05 level using Ryan-Einot-Gabriel-Welsch (REGWQ) Test.
^yStem length and branched stem length data at the Clary Hill field was transformed (square root transformation) for analysis but are presented as original means on the table.
^xPercentage data was transformed using arcsin transformation and analyzed but are presented as the original percentage in the table.
^wBranch data at the Marshville field was log transformed for analysis but are presented as original means on the table.

TABLE 4 Effect of Single or Split Application of Gypsum and DAP on Wild Blueberry Flower Bud Formation at Two Maine Wild Blueberry Fields

Treatments	Rates (kg·ha^−1)	Flower bud density (number per 0.023 m^2)	Number of unbranched stem flower buds	Number of branched stem flower buds	Flower bud per stem
CLARY HILL
Control	0	51.79 a ^z	0.692 a	1.99 a	1.35 a
DAP	448	57.33 a	0.452 a	2.63 a	1.67 a
Gypsum (single)	4,484	51.54 a	0.672 a	2.03 a	1.27 a
Gypsum (split)	2,242 × 2	43.54 a	0.380 a	1.97 a	1.21 a
MARSHVILLE
Control	0	49.92 ab	1.18 a	2.18 ab	1.52 a
DAP	448	54.21 a	1.12 a	2.43 a	1.72 a
Gypsum (single)	4,484	37.13 b	1.00 a	1.81 b	1.27 a
Gypsum (split)	2,242 × 2	44.33 ab	1.10 a	1.89 ab	1.51 a
^zMeans within a column within each location not followed by the same letter are significantly different at P < 0.05 level using Ryan-Einot-Gabriel-Welsch (REGWQ) Test.

Berry yield

Neither gypsum treatment increased blueberry yield at the Clary Hill field. This contrasts to the report of Sanderson et al. (1996) where application of 4,000 kg·ha⁻¹ gypsum resulted in an average of 47% increase in yield in three out of six fields. In a later study, Sanderson and Eaton (2004) applied 4,000 kg·ha⁻¹ gypsum alone or in combination with complete fertilizer to plots in five fields and did not report an increase in blueberry yield compared to the control. In our study, application of the recommended rate of DAP (448 kg·ha⁻¹) raised blueberry yield by 57%, compared to the control.

Marshville Field

Soil analysis

When soil pH was measured in CaCl₂, pH was slightly raised by both gypsum treatments compared to the control and DAP; but the increase was very small and unlikely to affect plant growth. Earlier studies on gypsum reported varying results; either soil pH was reduced (Carter et al., 1978; Sanderson et al., 1996), increased (Toma et al., 1999) or no effect at all (Sanderson and Eaton, 2004). According to Elrashidi et al. (2010), the ion exchange between gypsum and the cation exchange sites will determine the resultant soil pH; Ca will release H and Al ions while SO₄ ²⁻ will replace OH via ligand exchange. The resultant pH will, therefore, depend on the extent of the reactions between Ca or SO₄ ²⁻ and the exchange sites. In this study, it appears that the reaction releasing OH was stronger than the release of H and Al. Soil organic matter content was not affected by any of the treatments and ranged from 14.3 to 16.3%. As expected, both gypsum treatments increased soil Ca and S compared to the control and DAP. In contrast, soil Mg was reduced by both gypsum treatments compared to the control and DAP, as was the case in the Clary Hill field. None of the treatments affected soil K concentrations. Soil P concentration was raised by the DAP treatment, while both gypsum treatments were ineffective in raising soil P compared to the control. Only the single application of gypsum reduced the soil Mn concentration compared to the control, while the split application and the DAP treatments did not have any effect. None of the treatments affected soil Zn concentration compared to the control. When compared to DAP, only the single application of gypsum lowered soil Zn concentration.

Leaf analysis

Plants in the control plots at Marshville were identified as deficient in both N (<1.60%) and P (<0.125%) (Table 2). Both gypsum treatments and DAP raised leaf N and P concentrations above the standard (Trevett, 1972) concentrations; a split application of gypsum was not, however, more effective than the single application. Even though gypsum raised soil Ca concentrations, both gypsum treatments did not have any effect on leaf Ca concentrations; the DAP treatment, however, reduced leaf Ca concentration, likely due to a dilution effect caused by growth stimulation by N from DAP. Leaf K concentrations were raised by both gypsum treatments and DAP, compared to the control. Leaf Mg concentration was reduced by the split application treatment, while the single application and DAP had no effect compared to the control. This decline corresponded to the reduction in soil Mg concentration resulting from gypsum application. The decline in leaf Mg concentration was not enough to induce Mg deficiency in this field since the initial leaf Mg concentration was much higher (0.18%) than the standard (0.13%). None of the treatments had any effect on leaf Fe concentrations. Only the split application of gypsum raised leaf B concentration. None of the treatments reduced leaf Al concentrations compare to the control. However, only the single application of gypsum had higher leaf Al concentrations compared to DAP. Leaf Mn concentration was raised by the split application of gypsum only compared to the control.

Stem characteristics

Application of DAP or gypsum did not have any effect on stem density (Table 3). Average stem length and branched and unbranched stem lengths were raised by application of DAP but not by gypsum. Of all the treatments, only the DAP and split application of gypsum increased branching compared to the control. An increase in branching usually results in an increase in potential yield (i.e., more flower buds per unit area). However, none of the treatments increased the number of flower buds per stem (Table 4). Gypsum treatments did not increase the number of branches per stem compared to the control. Treatments had no effect on flower bud density, number of branched and unbranched stem flower buds and number of flower buds per stem compared to the control (Table 4). The split application was not better than the single application in affecting stem density, length, branching, and flower bud formation.

Berry yield

Application of DAP (448 kg·ha⁻¹) resulted in a 117% increase in yield, compared to the control. Neither of the gypsum treatments, however, increased blueberry yield.

CONCLUSIONS

Care should be taken when measuring the soil pH after application of large amounts of gypsum due to the resultant high electrical conductivity in the soil solution. Gypsum slightly raised the soil pH (CaCl₂,) at both locations but the increase was not enough to affect the growth of wild blueberry since the resultant pH in gypsum treated plots was still within the optimum range. Application of gypsum could potentially induce leaf Mg deficiency by releasing the soil Mg from the exchange sites where leaching through the soil profile makes it inaccessible to the plant. At Marshville, however, the reduction in leaf Mg did not lead to deficiency since the field had initially high leaf Mg concentration. Similar to the result of Sanderson et al. (1996), only one of the fields in this study responded to gypsum application. Only at the deficient Marshville field were leaf N and P raised up to the Trevett (1972) standards. It is, therefore, important to analyze leaf samples for N, P, and Mg concentrations before using gypsum as a fertilizer alternative. Neither of the gypsum treatments increased blueberry yield at either location. Application of DAP, however, increased blueberry yield at both locations. The split application of gypsum was not better than the single application in releasing soil nutrients, improving plant growth, foliar nutrient concentrations, or yield of Maine wild blueberry.

ACKNOWLEDGMENTS

We wish to acknowledge the support of the Maine Wild Blueberry Commission and Hatch Act for funding this research. The technical help of Kristen McGovern and the cooperation of growers Mike Bailey and Brian Powers are greatly appreciated. This manuscript is publication number 3122 of the Maine Agriculture and Forest Experiment Station.