Chapter IV
Effects of pattern and amount of nutrients on root placement and
Competition is inevitably intensified regardless of nutrient distribution when the amounts of nutrients are limited. Therefore, the total amount of nutrients has more profound effects on growth of a plant than patterns of nutrients, which can be enhanced by competition, especially, in low nutrients.
Key-words:
biomass; competition index; nutrient heterogeneity; patch size; soil resource availabilityIntroduction
Nutrients potentially used by plants are heterogeneously distributed in soil (Jackson and Caldwell 1989, 1993a,b). Plants under heterogeneous nutrient environments often grow larger than those grown in a homogeneous nutrient environment when total amount of nutrient is held constant (Birch and Hutchings 1994; Jackson and Caldwell 1996). The increase in biomass of a plant is the results of responses of the plant to a nutrient-rich patch (Caldwell and Pearcy 1994; Hutchings et al. 1999;
Robinson 1994; Hodge 2004, 2006) such as morphological specialization, selective root placement or root proliferation (Drew et al. 1973; Drew 1975; Drew and Saker 1975; Crick and Grime 1987), and nutrient uptake per unit of roots (Jackson et al. 1990; Schottelndreier and Falkengren-Grerup 1999).
Although there are many studies about the effect of spatial pattern of nutrients on an individual plant, our knowledge about its effect on a plant under competition for nutrients is limited (Hutchings et al. 2003; Hodge 2004; Schenk 2006). Recent research, however, suggests that aggregation of roots in a nutrient-rich patch and their acquisition of nutrients can intensify competition in a heterogeneous environment (Weiner 1990; Casper and Jackson 1997; Robinson et al.
1999; Day et al. 2003c; Hutchings et al. 2003). In addition, preemption of nutrients (Scwinning and Weiner 1998) and space occupation (Raynaud and Leadley 2004) that are caused by dominant use of nutrients by a few plants could occur through selective root placement, which would affect the
outcome of competition.
In addition to the spatial pattern of nutrients, the total amount of nutrients is also an important factor for plants under competition because competition in a nutrient-poor environment would be more intense than those in a nutrient-rich environment (Tilman 1987, 1988; Tilman and Pacala 1993; Cahill 1999; Pugnaire and Luque 2001). In a nutrient-poor environment, the reduction of nutrient availability limits growth of plants, and thus would affect outcomes of competition.
If the total amount of nutrients is the same among spatial patterns of nutrient supply, competition in a nutrient-poor environment would be intensified, independent of the spatial patterns.
However, our knowledge is very limited on relationships between effects of competition and patterns of nutrient supply on the growth of plants.
There are very few studies on direct effects of the total amount of nutrients on a plant with neighbours in a heterogeneous environment, but there is a study on a plant growing alone. A model proposed by Lamb et al. (2004) describes how a plant provided different amounts of nutrients could respond to changes in contrast of nutrient status between a nutrient-rich patch and elsewhere, and the model shows the root responses of an individual plant to a nutrient-rich patch would depend on the total amount of nutrients.
We do not fully understand how the amount of nutrients and their distribution affect individual plants under competitive conditions particularly in a nutrient-poor environment. In a nutrient-poor environment, distribution patterns of nutrients have been assumed to be homogeneous (Tilman and Pacala 1993) or heterogeneous (Goldberg and Novoplansky 1997; Rajaniemi 2003).
Several field studies found that distribution of nitrogen tends to be more heterogeneous as nutrient availability decreases (Kelly and Canham 1992; Gross et al. 1995; Cain et al. 1999). We, therefore, conducted a greenhouse experiment on the effects of the quantity and distribution of nutrients, and conspecific neighbours on root placement, biomass and intensity of competition. Ipomoea tricolor was grown in pots with or without neighbours in heterogeneous or homogeneous nutrient conditions at three nutrient levels. The following predictions were tested derived from the hypothesis; effects of
nutrient distribution would be larger and intensity of competition stronger in low nutrients conditions:
Prediction 1. A plant in a heterogeneous condition acquires more nutrients than that in a homogeneous condition by means of selective root placement in nutrient-rich patches, which causes larger biomass in a heterogeneous condition, especially, in low nutrients.
Prediction 2. Intensity of competition between plants is stronger in heterogeneous nutrient conditions than in homogeneous conditions providing the same total amounts of nutrients. Biomass of a plant with neighbours would be less than that without negihbours. The difference in plant biomass due to competition would be larger in heterogeneous conditions than in homogeneous conditions.
Prediction 3. Intensity of competition between plants under low nutrient conditions will be stronger than competition under high nutrient conditions regardless of the distribution of nutrients. The difference in plant biomass due to competition in low nutrient conditions would be larger than that in high nutrient conditions.
Materials and methods Material
An annual vine, Ipomoea tricolor var. “Heavenly Blue” (Sakata Seed, Kanagawa, JAPAN) was used in this study, because distribution patterns of roots of I.tricolor are significantly influenced by the spatial patterns of nutrients according to the results of our pilot experiments (data is not shown). The magnitude of root responses to a nutrient-rich patch was relatively large comparing with other species (see a review by Robinson 1994).
Seeds of I. tricolor were sown on wet vermiculite in a growth chamber, KOITOTRON (PC-02, Koito Industries, Kanagawa, JAPAN) at 22ºC and 70% of humidity after hard seed coats were removed by 50 minutes treatment of 98.0% (w/w) H2SO4. Similar sized-seedlings after a few days of the germination were transplanted into pots.
Experimental design
A 3 × 3 × 2 factorial design (pattern of nutrients × amount of nutrients × competition) with nine replicates per treatment was carried out in a randomized block arrangement in a vinyl tunnel at the experimental garden in Tokyo Metropolitan University (35o37’ N, 139o23’ E). Plants were grown in 16 cm × 13 cm (diameter × depth) terracotta pots. All pots were filled with 2000 ml of a granular red clay and vermiculite mixture (ratio 1:1 v/v) with a 1-year slow release fertilizer (Magamp K, 6-40-6 NPK, Hyponex JAPAN, Osaka, JAPAN). Soil nutrient were provided only as the fertilizer. Each pot was divided into four quadrants with an equal volume, and a pair of quadrants in opposing sites was called as ‘a patch’ (Fig. 4-1).
We used three patterns of nutrients; fertilizer was supplied evenly between patches in a pot (homogeneous condition: 50:50 treatment), or unevenly between a rich patch and a poor patch in a pot (heterogeneous conditions: 75:25 treatment and 100:0 treatment). In the heterogeneous conditions, a rich patch included 75% of the total amounts of the fertilizer and a poor patch 25% of them, which is called as ‘the 75:25 treatment’. As for the case of the 100:0 treatments, a rich patch included 100% of the total amounts of the fertilizer. These three patterns contained the same amounts of fertilizer. There were three levels of the total amount of nutrients: the 4 g treatment includes 4 g of the total amount of fertilizer per pot, the 8 g treatment 8 g, and the 16 g treatment 16 g. These amounts were decided according to the results of our pilot experiments on growth of I.tricolor (data is not shown). There were two treatments of competition: a single seedling per pot was planted, which is called as ‘without neighbours’ or four seedlings per pot, ‘with neighbours’. If there were four seedlings in a pot, the seedlings had to divide the total amount of nutrients between them. In single-plant pots, the seedling was planted at the center of the pot and occupied a single stake (Fig. 4-1a). Four seedlings were located on the boarder between a rich patch and a poor patch on the circle in a diameter of 8 cm and shared a single stake (Fig. 4-1b).
The experiment was carried out for five weeks from 18 June until 27 July in 2005. Above-
and belowground parts of plants were separately harvested for all nine replicates of each treatment.
Aboveground parts of each plant were dried at 75ºC to a constant weight. Belowground parts of a plant without neighbours were harvested separately between a rich patch and a poor patch, washed, and dried at 75ºC to a constant weight. A summation of belowground parts of a rich patch and a poor patch was regarded as the root biomass of the plant. Belowground parts of plants with neighbours were harvested separately between a rich patch and a poor patch in the same way without distinction of individual plants in four of the nine replicated pots or individually without distinction between a rich and poor patch in the rest five-replicated pots.
Datal analysis
Data were analyzed using the statistical software R version 2.1.0 for Macintosh. Root biomass, shoot biomass and the total biomass that was the summation of shoot and root biomass were log-transformed because of the non-normality or heteroscedasticity (Zar 1999).
Three-way multivariate analysis of variance (MANOVA) and three-way analysis of variance (ANOVA) were conducted to evaluate effects of patterns of nutrients, amounts of nutrients and competition. If the results were significant, they were tested with the Holm multiple-means comparisons in order to determine significant differences between means (Dalgaard 2002).
Intensity of competition was determined by the mean competitive intensity (CI) that is the mean reduction in biomass between the biomass without neighbours and the biomass with neighbours (Kadmon 1995; Day et al. 2003c). The CIs were determined by the difference in the total biomass between the pair of pots chosen within a block with the same pattern of nutrients and the same amount of nutrients. The CIs were calculated according to the formula of Kadmon (1995):
CI = [4BBi – BcB]/4 (eqn 1)
where BBi denotes biomass of a plant without neighbours and BcB the sum of biomass of four plants in a pot. If plants with neighbours in a pot are not affected by competition, its CI is zero. Positive CIs indicate that plants are competing, and negative CIs that they benefit from neighbours. One-sample
t-tests determined whether the CIs significantly differed from zero. The effects of treatments (nutrient heterogeneity and the amount of nutrients) on CIs were investigated using two-way ANOVA with the treatment as the independent variable and the CI as the dependent variable.
Results
Root placement in patches
Multivariate analysis of variance revealed that root biomass per plant in a rich patch and a poor patch was significantly affected by spatial patterns of nutrients and marginally affected by amounts of nutrients (Table 4-1a and see also Appendix 4-1). The univariate analyses revealed that the patterns of nutrients and the amounts of nutrients also affected root biomass in a poor patch, but not root biomass in a rich patch (Table 4-1b).
Proportion of root biomass in a rich patch to the total of root biomass per pot (hereafter root proportion) was significantly affected by the patterns of nutrients (F2, 98 = 23.76, P < 0.001) and by the amounts of nutrients (F2, 98 = 3.13, P < 0.05). In the 100:0 treatments, mean of the root proportion in the 16 g treatment was the largest (0.64). In the 50:50 treatments, means of the root proportion were approximately 0.5 irrespective of the amounts of nutrients (Fig. 4-2).