Abstract
Landscape trees typically grow slowly for several years after transplanting. We investigated whether fertilization could speed tree growth during this establishment period, which fertilization regimes were most effective, and whether fertilization interacted with irrigation. Fifty-four each of landscape-sized, balled-and-burlapped red maple (Acer rubrum) and littleleaf linden (Tilia cordata) were planted into a relatively infertile silt loam soil and were fertilized (1.5 kg N/100 m2 [3 lb N/1000 ft2]) each spring (either including or not including at planting), each fall, or not fertilized. Each of these fertilizer regimes was either irrigated or not irrigated during 3 years. An additional treatment of an unirrigated, split (spring/fall) fertilizer application was included. There was no evidence that fertilization affected irrigated trees differently than unirrigated trees. Overall, fertilization did not speed establishment and did not affect trunk growth, shoot extension, or leaf nitrogen content. There was no evidence that fall fertilization might be more effective than spring fertilization. There was no indication that fertilized trees experienced increased drought stress. Nitrogen rates and factors affecting fertilizer uptake are discussed.
Newly planted trees, until they have established and put on significant growth, provide few of the economic, environmental, and aesthetic benefits expected from landscape trees. In this study, we considered whether fertilization could hasten the establishment process, thus accelerating a newly planted tree’s entry into the more environmentally productive phases of its life. Tree fertilization has been studied for many years but with ambiguous results (see Struve 2002 for a general review). Fertilization research to date does not provide the definitive answers needed to make fertilization recommendations for newly transplanted trees, especially balled-and-burlapped (B & B) trees and those planted in the less-than-ideal soils found in urban areas. Whitcomb and his colleagues studied tree response to fertilizer during establishment. Eleven species of bare-root trees showed no growth response to nitrogen during establishment, although leaves of fertilized trees were visibly darker green in the second year (Shoup et al. 1981). Silver maple, a highly vigorous species, grew more rapidly with increased nitrogen when planted in a good clay loam soil but had no response when planted on a poor site with nutrient-deficient silt loam soil (Schulte and Whitcomb 1975). After these mixed results, Whitcomb recommended fertilizing lightly at planting (Whitcomb 1984). Kelting et al. (1998a) found that fertilizing small bare-root trees at planting had little effect on growth in a relatively infertile silt loam soil, and plants were not harmed by quick-release fertilizers at 14.5 g N/m2 (3 lb N/1000 ft2). Neely (1980) found that established trees in fertile soil received only a small benefit from fertilization. In a recent study on two urban sites in Milan, Italy, fertilization increased photosynthesis rate of Japanese pagoda tree (Styphnolobium japonicum) and sweetgum (Liquidambar styraciflua) but not European ash (Fraxinus excelsior) during the first year (Ferrini and Baietto 2006). In subsequent years, this effect disappeared or was reversed and in no case was growth affected by fertilizer. The variable results of past studies have led many practitioners to recommend no fertilization at planting (although little is actually known about when during the first few years after planting that fertilization may speed establishment). As a result, there is much controversy about fertilization at planting and during establishment, and many arborists and urban foresters are hesitant to fertilize at these times. Harris et al. (2004) recommend that trees not be fertilized at planting except in nutrient-deficient soils and that quick-release fertilizers placed in the planting hole should be avoided. Others suggest very low rates of slow-release fertilizer at planting (0 to 4.9 g N/m2 [0 to 1 lb N/1000 ft2]) (Gill et al. 2001). The revised American National Standards Institute (ANSI) standard for tree care operations recommends slow-release fertilizers for new transplants (an earlier version recommended no fertilization) (ANSI 1998, 2004). A rate of between 2 and 4 lb N/1000 ft2 (9.8 to 19.5 g/m2) for slow-release (half that rate for quick-release) is recommended to address a defined objective in established trees. Even if not harmful to the plant, unnecessary fertilizer is obviously not cost-effective and raises concerns of degrading water resources through runoff or nitrogen leaching.
As extensive research in turfgrass and row crops has shown, the timing and rate of fertilizer application is critical to produce the healthiest and most productive plants as economically as possible and without harming the environment (Petrovic 1990; Cassman et al. 2002; Easton and Petrovic 2004). Arborists, however, have little information available to guide fertilization decisions. Prescription (also known as prescriptive) fertilization (adding nutrients to correct an identified deficiency) may be useful for mature trees (Chacalo 2000; Lilly 2001). Many mature trees presumably also derive nutrients from the fertilization of lawns with which they share rooting space, although this has not been documented. The needs of newly planted trees, however, cannot be addressed in these ways. Furthermore, the physiological disruption caused by transplanting (root severance and loss, desiccation, vessel embolism) may affect tree response to added nutrients.
Trees, especially those in urban areas, often appear to languish for many years before their roots can acquire adequate water and nutrients to sustain significant growth. One area of research interest is whether fertilizer can be effectively used to improve the nutrient status of trees during this period and hasten their entry into the environmentally productive phase of their life. In this study, we evaluate the benefit of fertilization of newly transplanted landscape trees and the effect of fertilizer timing during the establishment period. Our objectives were:
To determine if increases in plant growth can be realized through early fertilization;
To determine if spring or fall fertilization is more effective;
To determine if early fertilization interacts with typical transplant stress caused by lack of irrigation; and
To compare the effectiveness of fertilizing at planting to fertilizing after one growing season.
METHODS
Treatments and Experimental Design
Fifty-four landscape-sized, field-grown, seedling-grown B & B trees each of red maple (Acer rubrum) and littleleaf linden (Tilia cordata) were obtained from Willow Springs Nursery (Christiansburg, VA, U.S.) and transplanted on 24 March 2000 at the Urban Horticulture Research Center in Blacksburg, Virginia. The soil at the research center is a Groseclose silt loam soil (clayey, mixed, mesic Typic Hapludults) with pH 6.2 and a native N content of approximately 1350 mg N/kg of soil (0.0216 oz N/lb of soil) total Kjeldahl nitrogen typical of the relatively infertile soils of the region. Average height and caliper (standard error mean in parentheses) at planting were 3.6 (0.09) m (11.9 [0.3] ft) and 4.19 (0.10) cm (1.7 [0.04] in), respectively, for red maples and 2.69 (0.04) m (8.9 [0.13] ft) and 4.66 (0.06) cm (1.9 [0.02] in) for lindens. Trees of each species were placed 2.7 m (9 ft) apart into one of nine treatments arranged in a completely random design. Treatments were structured as an augmented factorial (Lentner and Bishop 1993) with irrigation and fertilizer regime as factors and an unirrigated, split-application treatment as the augmented treatment (Table 1). Treatments included (1) fertilized each spring and at planting and irrigated (FSPI); (2) fertilized each spring except at planting and irrigated (FSI); (3) fertilized each fall and irrigated (FFI); (4) not fertilized, but irrigated (NFI); (5) fertilized each spring and at planting and not irrigated (FSP); (6) fertilized each spring except at planting and not irrigated (FS); (7) fertilized each fall and not irrigated (FF); (8) fertilized each spring and fall (except at planting) with a split application and not irrigated (FSp); and (9) not irrigated or fertilized (NF). All trees were mulched with shredded hardwood bark in a ring extending approximately 30 cm (1 ft) beyond the root ball. Rows were maintained weed-free.
Abbreviations for experimental treatments applied to littleleaf linden and red maple trees.
Fertilizer applications were broadcast over a 1 m2 (10.8 ft2) area centered around each tree at a rate of 1.5 kg N/100 m2 (3 lb N/1000 ft2) and half that rate for each application in the FSp treatment. The fertilizer was a slow-release, sulfurcoated product from Southern States Cooperative, Inc. (Richmond, VA) with a 27 to 3–12 analysis (5.0% ammoniacal N, 1.8% water-insoluble N, 17.6% urea N, and 2.6% other water soluble N) described by the manufacturer as having a 4 month release time. Fertilizer was applied according to the treatment as follows: 31 March 2000 (at planting), Sept. 2000 (first fall), April 2001 (second spring), Oct. 2001 (second fall), April 2002 (third spring), and Oct. 2002 (third fall).
All trees were hand watered for 2 weeks after planting. Subsequently, trees in unirrigated treatments received no supplemental water. In each irrigated treatment, soil moisture sensors (Watermark, Irrometer Co. Inc., Riverside, CA, U.S.) were installed 20 cm (8 in) deep inside and just outside of the root ball of three randomly selected replications. A given treatment was irrigated using drip emitters when soil tension reached 0.055 MPa (55 centibars) on two replications. A replication was considered to have reached this tension when one of the two sensors reached 0.055 MPa (55 centibars). This ensured that trees were irrigated before experiencing drought stress. Rainfall in Blacksburg during the 2000 and 2001 growing seasons was slightly below 30 year norms; for 2002, it was slightly above (Virginia State Climatology Office).
Measurements
Tree growth and nutrient status were monitored through three growing seasons to ensure that the entire establishment period was studied (Watson 1985). At the end of each growing season, tree caliper readings were taken in two directions, along the row and across the row, and averaged for each tree. Caliper was used to calculate cross-sectional trunk area increase to give an integrated assessment of overall tree growth. Shoot extension during the 2000 and 2001 growing seasons was obtained by measuring new growth in February and March the next year before budbreak. The seven shoots nearest the top that showed the most growth (suppressed shoots, if any, were not included) were measured on each tree. Shoot extension for each replication was taken to be the mean of the remaining five shoots after the two longest shoots were excluded. Foliar nitrogen levels were determined in early summer of the second growing season using four, randomly selected replications of each treatment. We selected five recently matured leaves from separate branches throughout the canopy of each tree selected and pooled them to make one sample. Percent nitrogen content was analyzed using the micro-Kjeldahl method (Peterson and Chesters 1964). To maximize statistical power, only a priori (preplanned) questions of interest were formulated as linear contrasts. These a priori contrasts were analyzed by multivariate repeated measures within the GLM procedure of SAS (version 8; SAS Institute, Cary, NC, U.S.).
RESULTS
For the most part, fertilization did not affect cross-sectional trunk increase (Figure 1; P values for preplanned contrasts of repeated measures analysis shown in Tables 2 and 3), shoot extension (Figure 2, Tables 2 and 3), or leaf nitrogen content (Tables 2 and 3) of either maples or lindens. Among irrigated treatments, there is no evidence that fertilizer applications affected growth (Tables 2 and 3). Among fertilized treatments, there is little evidence that irrigation affected growth or other variables. Cross-sectional trunk area growth rates and shoot extension were low the first year. This transplant shock effect was most dramatic for lindens (Figures 1 and 2). Repeated measures analysis looks at the growth rate over time (i.e., how size moves through time) and therefore reduces error that might be introduced by differences in initial tree size. Consequently, although spring-fertilized, irrigated red maples appear to have larger cross-sectional areas than other trees, the rate of growth (indicated by the slope of the lines in Figure 1) is similar to other trees (as indicated by the P values in Table 2). In lindens, leaf nitrogen content was higher for FSp trees than in those with other nonirrigated fertilizer regimes (FSP = 2.39%, FF = 2.07%, FSP = 2.15%, FS = 2.08%) as indicated by the contrasts in Table 3. The line slope (Figure 1) for FFI red maples between 2000 and 2002 is different from that of other treatments, indicating they grew at a slower rate than other treatments during that period. This is also confirmed by the contrast shown in Table 2 (P = 0.014) between fall and spring fertilization for irrigated red maples. Because FSI red maples show a similar decline in growth rate between 2002 and 2003, however, spring fertilization does not appear to be consistently more effective. Both red maples and lindens with FFI treatment had lower average shoot extension in 2002 than their spring fertilized counterparts; however, these differences are slight and may be the result of factors other than treatment.
Cross-sectional area of trunk for red maple (Acer rubrum) and littleleaf linden (Tilia cordata) for three growing seasons after transplant under different fertilization and irrigation regimes. Measurements were taken at the end of the growing season except where indicated. Bars indicate standard error of the means of six replications. See Tables 2 and 3 for statistical comparisons.
Shoot extension (mean of five shoots per replication) for red maple (Acer rubrum) and littleleaf linden (Tilia cordata) for two growing seasons after transplant under different fertilization and irrigation regimes. Bars indicate standard error of the means for six replications. See Tables 2 and 3 for statistical comparisons.
Red maple (Acer rubrum) P values of a priori contrasts for cross-sectional trunk area growth for 3 years after transplanting, shoot extension (mean of five shoots per replication) for 2 years after transplanting, and leaf nitrogen content (pooled sample of five leaves per replication) the third year after transplanting.
Littleleaf linden (Tilia cordata) P values of a priori contrasts for cross-sectional trunk area growth for 3 years after transplanting, mean shoot extension (mean of five shoots per replication) for 2 years after transplanting, and leaf nitrogen content (pooled sample of five leaves per replication) the third year after transplanting.
DISCUSSION
The A300 fertilization standard emphasizes that trees should only be fertilized when there is a defined objective. The fertilization objective in this study was to speed the establishment of shade trees. Posttransplant growth of shade trees typically increases annually during the establishment period (Harris and Gilman 1991), so rapid establishment can accelerate investment return in terms of shade, cooling, carbon sequestration, runoff reduction, and so on. When trees are established, they may no longer need supplemental irrigation. Although there is a defined need in speeding establishment of trees on stressful sites, there is not, at present, a reliable means of achieving this through fertilization.
The total nutrients taken up by the tree from fertilizer is a function of the amount of fertilizer applied and the ability of the plant to absorb these nutrients. Typical fertilizer rates are described on a N per unit of soil area basis. Ideally, a given rate would provide the plant with sufficient nutrients to achieve a stated management objective while taking into account the plant’s ability to absorb these nutrients. When a field-grown tree is dug at the nursery and transplanted, a large part of its root system is lost. This places the tree in the situation of having an aboveground mass that is the same as it was before digging but a root system that is considerably smaller. It may then be virtually impossible to achieve an ideal rate: a “typical” rate may provide far too little fertilizer for the tree (see example subsequently), whereas raising the rate to compensate may overwhelm the ability of the remaining root system to take up nutrients. Although the rate used in this experiment conforms with current recommendations of between 2 and 4 lb N/1000 ft2 (9.8 to 19.5 g/m2) (ANSI 2004), this translates into approximately 15 g (0.53 oz) of actual nitrogen per tree when spread over the rootball area and just beyond (1 m2 [10.8 ft2] in the current study). For example, in a 7.5 cm (3 in) caliper B & B, 2.4 m (8 ft) tall tree, the transplanted root ball is generally 0.71 m (2.3 ft) in diameter, yielding a rooting ground surface area of 0.4 m2 (4.3 ft2) (ANSI 2004). The same size, fully established tree planted in open ground (using the rule of thumb that root spread is 1.5 to 2 times tree height) would have a root system ground surface area between 10.4 and 18.6 m2 (112 to 201 ft2). Thus, when fertilization is based on ground surface area of the root system, a tree that has not been transplanted would receive roughly 20 to 40 times as much fertilizer as the same tree after transplanting. One possible reason for the lack of tree response to fertilization in the current study is that the recommended rates simply do not provide a sufficient quantity of fertilizer in this scenario to produce a measurable response.
There is little evidence in the literature of high N rates damaging transplanted trees. Bare root liners receiving as much as 32.8 g N/m2 (6.7 lb N/1000 ft2) did not appear damaged by these high rates; although one of the three species studied did show reduced growth in the first year, this effect disappeared in the second year (Wright and Hale 1983). There is evidence that in some situations, optimum N application rates for growth of established shade trees may be anywhere from 29.3 g N/m2 (6 lb N/1000 ft2) to 58.6 g N/m2 (12 lb N/1000 ft2) (see Struve’s discussion [2002] of research by Neely, Himelick, and Crowley [Neely et al. 1965, 1970]). Gilman et al. (2000) found increases in tree height for container-grown magnolias fertilized in the landscape at a rate of 40 g N/m2 (8.3 lb N/1000 ft2) over those fertilized at only 20 g N/m2 (4.2 lb N/1000 ft2). All of these rates are considerably higher than current recommendations. Nonetheless, there are other impediments to transplanted B & B trees taking up nitrogen.
Nitrogen uptake by the tree may be suppressed by the root severance that occurs during harvesting and not recover until root regeneration later in the season (Dong et al. 2003). For red maple (Acer rubrum) in Blacksburg, VA, new root growth into the backfill soil may not be visible until 38 days after spring transplant, but other species can take much longer (Kelting et al. 1998b). For example, fringe tree (Chionanthus virginicus) roots may not be visible until 110 days after spring transplant (Harris et al. 1996). Thus, delivery of nutrients to the newly transplanted tree may be hindered by both the reduced physical size of the root system and its physiological state. In the current study, evidence suggests that littleleaf linden and red maple were neither aided nor damaged by fertilizer application. As discussed previously, possible explanations include what is, effectively (on a per tree basis), an extremely low fertilizer rate and of nitrogen uptake reduction caused by root severance that occurred during the transplanting process.
In our study, there was very slight evidence that spring applications of fertilizer might be more effective than fall applications, but this could be the result of the overall lack of effect of any fertilizer regime. In any case, there was certainly no evidence that fall fertilization was more effective than spring. Other work suggests that nitrogen uptake in woody plants is likely better in spring (Aguirre et al. 2001) but may be inhibited temporarily by root severance (Dong et al. 2003). Studies conducted in the same geographic location as this experiment examined root growth of several species, including red maple. Active root growth was typically present slightly before or after budbreak until leaves begin to color for the fall (i.e., the senescence process begins), excluding both hot, dry periods in the summer when root growth is negligible and 1 to 3 months into the growing season after transplant (depending on species) (Harris et al. 1995, 1996).
It is unclear whether practitioners can reasonably expect to provide sufficient fertilizer to transplanted B & B trees through traditional methods and rates. We suggest that further study is merited using higher application rates or other methods (such as foliar application or nutrient loading during nursery production). There is evidence that nutrient loading of container-grown trees at the nursery before harvest can affect growth rates of transplanted trees during the first year (Lloyd et al. 2006). The importance of nitrogen reserves to growth is also apparent in several other studies. In apple trees, reserve nitrogen was found to be a more important determinant of spring shoot flushes than reserve carbohydrates (Cheng and Fuchigami 2002). Dyckmans and Flessa (2001) studied nitrogen mobilization in young beech trees growing in washed sand (labeled N was provided through nutrient solutions) and concluded that internal nitrogen stores, rather than current supply, had the strongest influence on growth. In the case of B & B trees, the degree to which nutrient loading might be offset by the loss of much of the root system would also need study.
CONCLUSIONS
Currently, standards recommend slow-release fertilizer for new transplants and rates comparable to those used in this study. It may be that using nutrients to enhance rapid establishment of B & B trees can be achieved by other means (different rates, timing of application, or nutrient loading). Current recommendations, however, did not speed establishment for the species studied here. Soil nutrient supply may also be a factor. Although the study site has relatively poor native nitrogen availability as is typical of many urban sites in the mid-Atlantic region of the United States, it is not severely compacted or scraped. There are many sites with more extreme soil conditions—such as disturbed sites where the A (and sometimes B) soil horizons have been removed. Two experiments in progress will examine fertilization at higher rates in soils disturbed by building construction to determine if fertilization can achieve the goal of speeding establishment under these more extreme conditions. At present, however, it appears that other site treatments such as improving overall soil conditions by adding organic matter and reducing compaction (Day et al. 1995) and increasing soil volume (Grabosky and Gilman 2004) may be the most worthwhile investments to speed establishment and long-term growth. Unfortunately, these treatments tend to be more expensive and require planning and coordination, which are often difficult to obtain in urban environments. Our findings from the current study indicate:
Current fertilizer rate recommendations do not speed establishment of B & B red maple and littleleaf linden in moderately poor soil conditions.
Fall applications of fertilizer are not more effective than spring applications for red maple and littleleaf linden in these conditions. No fertilization regime improved growth, however, so differences may not be apparent.
There was no evidence that fertilizer treatments had negative effects on the growth of unirrigated littleleaf lindens and red maples when compared with irrigated trees, even when rainfall was below average.
IMPLICATIONS FOR ARBORICULTURE
Research concerning fertilization at planting of balled-and-burlapped trees still has not provided definitive answers, and the arboriculture profession should expect recommendations to change as more information becomes available. Fertilizing B & B trees does not speed establishment under average to moderately poor landscape conditions for littleleaf linden and red maple, whether irrigated or not, in the soil conditions described in this experiment. There may be situations in which fertilization at this time is beneficial, but these have not been fully defined. The results of this study do not provide evidence that fall transplanting is more effective than spring transplanting. Our research further did not suggest that fertilization at transplanting or fertilization of unirrigated trees during establishment will result in tree stress.
Acknowledgments.
This research was funded in part by the TREE Fund John Z. Duling Grant Program. We are grateful for this support and for the invaluable technical assistance of Jody Fanelli and John James at the Urban Horticulture Center of Virginia Tech. We are also grateful to Dr. Dan Ward (former Statistician for the Virginia–Maryland Regional College of Veterinary Medicine, currently Assistant Professor of Horticulture at Rutgers University) for statistical consultation.
- © 2007, International Society of Arboriculture. All rights reserved.
LITERATURE CITED
Jump to section
Related Articles
Cited By...
- No citing articles found.