AOBPreview originally published online on August 26, 2005
Annals of Botany 2005 96(6):1009-1017; doi:10.1093/aob/mci253
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Change in Spatial Distribution Patterns of a Biennial Plant between Growth Stages and Generations in a Patchy Habitat
Department of Biological Sciences, Graduate School of Science, Tokyo Metropolitan University, Minami-Osawa 1-1, Hachioji, Tokyo 192-0397, Japan
* For correspondence at: KYOUSEI Science Center for Life and Nature, Nara Woman's University, Kitauoya-nishimachi, Nara 630-8506, Japan. E-mail rsuzuki{at}cc.nara-wu.ac.jp
Received: 4 February 2005 Returned for revision: 25 April 2005 Accepted: 8 July 2005 Published electronically: 26 August 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
• Background and Aims The aim of the study was to evaluate factors causing change in spatial distribution patterns of plants between growth stages and generations for a monocarpic biennial plant, Lysimachia rubida. It was assumed that habitat heterogeneity was a primary factor determining spatial patterns of plants, and a randomization procedure was developed for testing the null hypothesis that only spatial association with ground surface conditions determined spatial patterns of plants.
• Methods A 5-year demographic census was conducted on an open dry habitat that was heterogeneous with regard to the ground surface conditions.
• Key Results There was significant habitat association in that plants at vegetative and reproductive stages were denser in areas with smaller gravel than with larger gravel. Point process analyses rejected the null hypothesis of the spatial association with ground surface conditions.
• Conclusions The results suggest that other factors, such as patchy seed dispersal, secondary dispersal of the seeds and life-history variation at various spatial scales, also affected spatial patterns of individuals in a population of L. rubida. Spatial structures and dynamics of a local population in a patchy habitat represent various performances of plants within patches and seed dispersal within a patch and beyond the patch.
Key words: Monocarpic biennial, generations, environmental heterogeneity, ground surface conditions, growth stages, life history variation, Lysimachia rubida, patch dynamics, point process analysis, randomization test, seed dispersal, spatial patterns
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Plant survival and growth depend on local environments within a habitat, and they are heterogeneous (Jackson and Caldwell, 1993
; Farley and Fitter, 1999). Spatial configuration of suitable environments for plants is often patchily structured at various sizes within the habitat, like islands in a sea (Gibson, 1988; Kotliar and Wiens, 1990; Hiebeler, 2000). Performances of individual plants in response to the patchiness of environments are spatially non-random processes (Antonovics and Levin, 1980; Fowler and Antonovics, 1981; Stratton and Bennington, 1998), such as density-independent mortality (Casper and Cahill, 1996), patchy establishment of seedlings (Harms et al., 2001; Debski et al., 2002) and seed dispersal patterns within and between patches (Russell and Schupp, 1998). Structures and dynamics of a local population in a patchy habitat also depend on the structure of the habitat, which is called patch dynamics (Horvitz and Schemske, 1986; Watkinson et al., 2000), although it is usually applied to the large-scale dynamics of a regional population at landscape level (reviewed in Hanski, 1994). However, how the spatially non-random processes determine the patch dynamics of a local population has rarely been studied.
Aggregated patterns of plants are often observed as spatial structures in a local population especially in a patchy habitat, and these are scale-dependent. Aggregated patterns at the spatial scale corresponding to the size of patches would be the result of spatial variation in mortality and establishment of plants caused by the patchiness of the environment (Houle, 1992, 1998; Schupp, 1995; Hyatt, 1998). Many studies showed that patchy seed dispersal around parent plants is another important factor causing aggregated patterns of plants (Howe, 1989; Eriksson, 1994; Forget et al., 1999). The spatial scale of aggregation would depend on the distance seeds are dispersal. If seeds are frequently dispersed beyond a patch within a habitat, plants would occupy most of the suitable patches. On the other hands, if dispersal of seeds occurs only around reproductive plants within a patch, plants would aggregate within the ‘home’ patch and a number of suitable patches may remain unoccupied by plants. Therefore, the spatial scale and degree of aggregated patterns of a local population would depend on the sizes of suitable patches and the capability of plants to disperse their seeds and the capability of plants to persist within patches (Ehrlen and Eriksson, 2000; Rand, 2000). It is very difficult to detect which factors generate aggregated patterns of the local population in a patchy habitat, because those different factors can potentially generate the same aggregated patterns of plants (Schurr et al., 2004).
The analysis of change in spatial distribution patterns of plants within and between generations is useful for understanding the processes determining structures and dynamics of a plant population, such as seed dispersal, intra- and inter-specific competition and environmental heterogeneity (Sterner et al., 1986; Kenkel et al., 1997; Barot et al., 1999; Dov
iak et al., 2001), which could not be directly addressed with the more usual snapshot approach to pattern analysis. The analysis of spatial scales of spatial patterns should also suggest differences in factors determining the spatial patterns (Cole and Syms, 1999; Dale, 1999). Moreover, it is necessary to develop statistical methods for detecting different factors separately (He and Duncan, 2000; Schurr et al., 2004).
In this study, the patchiness of local environments within a habitat is assumed to be a primary factor affecting spatial patterns of plants, and then a randomization procedure is developed for testing a null hypothesis that only spatial association with patches determined spatial patterns of plants. A local population of a monocarpic biennial plant Lysimachia rubida represents a model system for testing the hypothesis, because the micro-environments of habitats of these plants are remarkably heterogeneous with respect to ground surface conditions, which crucially affect spatial variation in mortality of L. rubida individuals (Suzuki et al., 2003). Change in spatial patterns of L. rubida between growth stages and generations was analysed to test a null hypothesis that only ground surface conditions determine spatial patterns of the plants, and underlying processes determining patch dynamics of a local population in the patchy habitat are discussed.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Plants and study sites
Lysimachia rubida Koidz. (Primulaceae) is endemic to the Bonin (Ogasawara) Islands, subtropical oceanic islands in the north-west Pacific Ocean. The plants are patchily distributed in open dry habitats on rocky coastal cliffs on the islands. These plants produce a large number of fine seeds (each approx. 18 µg) and do not reproduce asexually. Seedlings emerge mainly from January to March. Individual plants grow vegetatively for at least 1 year and each forms one to 20 rosettes. They flower from March to June and disperse mature seeds from May to July in the second year. All reproductive plants wither by August.
The study site was located on Minami-jima (27°05'N, 142 °12'E), an islet belonging to the Chichi-jima islands of the Bonin Islands. The average monthly temperature ranged from 27·7 °C in August to 17·8 °C in February, and the mean annual rainfall was 1333 mm (for the years 1986–1999; Japan Meteorological Agency). Precipitation was relatively high in May and October–November, while July–September and January–March were generally dry.
The study site was situated on a cliff, approx. 20 m above sea level, covered with limestone rocks of diameter <30 cm. Shrubs of Scaevola sericea Vahl or grasses of Sporobolus virginicus Kunth and Zoysia tenuifolia Willd. were scattered patchily at the site.
Demographic census and measurements
Demographic censuses were annually conducted every May from 1998 to 2002 in a 14 m x 8 m plot at the study site. All reproductive plants in the plot were mapped at each census. Vegetative plants were tagged and mapped in an 8 m x 8 m area at 0–8 m of the 14 m x 8 m plot in 1999, and those in all areas of the plot in 2000 and 2001. The year in which vegetative plants were established was identified based on their labels. The numbers of vegetative plants that emerged in the plot in 1998 and 2002 were recorded.
The heterogeneity of micro-environments within the study plot was characterized according to ground surface conditions. The study plot was divided into 11 200 10 m x 10 m subplots and each of them was categorized into five classes based on the maximum gravel size: 1 (<1 cm), 2 (1–3 cm), 3 (3–6 cm), 4 (6–10 cm) and 5 (>10 cm). Category 1 showed the largest water-holding capacity and category 5 the lowest water-holding capacity (for details, see Suzuki et al., 2003).
Analyses
Torus-translation tests of habitat association
To investigate the spatial association between distributions of plants and ground surface conditions, a torus-translation method based on that developed by Harms et al. (2001) was used. A null hypothesis assumes that association between the spatial distribution of plants and the arrangement of ground surface conditions is caused coincidentally. An association expected under the null hypothesis was generated by converting the study plot to a torus, and then randomly translating the plant location in a year in the x and y directions relative to the ground surface conditions. In the randomization method, critical properties of the spatial structure of both the plants and the ground surface conditions are thus maintained, while the relative locations of plants with respect to the ground surface conditions are altered for each translation. The method statistically tests whether the observed densities of reproductive or vegetative plants in a year in a ground surface condition category were significantly different from null distributions of the expected densities. The null distributions of expected densities were generated from 1000 translations.
Point process analyses of spatial pattern and association
Univariate spatial patterns (aggregation/regularity) of plants were analysed using Ripley's K(t) function (Ripley, 1977; Diggle, 1983). The K(t) function is defined as the expected number of plants within distance t from a randomly chosen plant; under complete spatial randomness K(t) = 
t2 (for detailed properties of the function K(t), see Diggle, 1983; Haase, 1995). When L(t) is defined as 
[K(t)/] – t, under complete spatial independence, the expected value of L(t) is zero. In the present analyses, L(t) was calculated at 10-cm intervals up to 400 cm. A null hypothesis assumed that spatial randomness with ground surface conditions only determined the spatial aggregation of plants. Tests were carried out to find out whether the distribution of the observed vegetative or reproductive plants in each year were more aggregated or regular than the random distributions expected from the null hypothesis. A random distribution expected from the null hypothesis was generated by randomly arranging the position of each individual in a year, while the density of plants on each category of ground surface conditions remain fixed to the observed density in the year.
The spatial associations between reproductive plants in a year and those in another year and between reproductive plants in a year and vegetative plants in the following year were analysed using a bivariate function L12(t). The L12(t) function is a transformation of Ripley's K12(t) function that is a generalization of the function K(t) to a bivariate point process. The K12(t) function is defined as the expected number of plants in 1 year (group 2) within distance t from a randomly chosen plant in the different year (group 1). For each calculation, two functions can be computed, with either group 1 or 2 as the first group. The empirical function K* can be computed as the mean of K12 and K21, weighted by the numbers of points in the two groups. For association between reproductive plants, therefore the spatial interaction analyses were based on the K* function. Because the relationship between reproductive plants in a year and vegetative plants in the following year is clearly asymmetrical, the calculation of K12 was based only on the reproductive plants as centres. K* was also calculated and the results obtained were almost identical to those based on K12. The results of the K12, therefore, are presented for the association between reproductive plants and vegetative plants. The L12(t) function was calculated at 10-cm intervals up to 400 cm. A null hypothesis assumed that the spatial association between the two groups is only caused by the spatial association with ground surface conditions. Tests were carried out to find out if the spatial associations between the two groups were more attracted or repulsed than the attractions expected under the null hypothesis. An association expected under the null hypothesis was generated by arranging randomly the position of each individual in the later year (group 2), while the density of plants in each category of ground surface conditions remain fixed to the observed density in the group.
The 99 % confidence envelopes of univariate functions and of bivariate functions were estimated from 1000 simulations by random arrangement of plant positions under a certain habitat association. When the observed L(t) or L12(t) values were larger or smaller than the envelopes of the expected L(t) or L12(t) under the null hypotheses, the spatial pattern (aggregation/regularity) or the spatial association (attraction/repulsion) between the two generations or between reproductive and vegetative plants was statistically significant at distance t, respectively.
In this study, the ‘scale’ of spatial patterns and associations was defined following Barot et al. (1999), i.e. as the distance with maximal deviation between the significant L(t) or L12(t) values and the mean L(t) or L12(t) of 1000 simulations under the null hypothesis.
Spatial analyses were also conducted using only the data of distributions of current-year seedlings and results obtained similar to all vegetative plants including current-year seedlings and plants established in previous years. Here only the results of the vegetative plants are shown.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
The numbers of vegetative and reproductive plants varied between years (Table 1). The number of vegetative plants tended to decrease over the years. Their number in 1998 was large as was the number of reproductive plants in 1999. Most of the vegetative plants were current-year seedlings, but a few were plants that emerged as seedlings in previous years. The percentages of those vegetative plants established in previous years were 3·5, 14·0 and 2·2 in 1999, 2000 and 2001, respectively (Table 1).
|
Torus-translation tests of habitat association
The spatial distribution of ground surface conditions in the 14 m x 8 m plot is shown in Fig. 1. Spatial distributions of vegetative plants and of reproductive plants varied between the five years (Fig. 1). Plant density monotonically decreased from gravel category 1 to gravel category 5 for vegetative plants in all years and reproductive plants in most years (Tables 2 and 3). The torus-translation tests revealed that the association of density of vegetative and reproductive plants with gravel categories was statistically significant in most cases. Positive association with smaller gravel of categories 1 and 2 was observed for vegetative and reproductive plants for several years (Tables 2 and 3). Negative association with largest gravel of category 5 was observed for vegetative and reproductive plants in all years (Tables 2 and 3).
|
|
|
Spatial aggregation of plants
Spatial distributions of vegetative and reproductive plants in all years showed more significant aggregation (P < 0·01) than the distributions expected from the null hypothesis that plants are spatially associated only with ground surface conditions (Fig. 2). Shapes of L(t) curves were broadly similar between vegetative plants in different years and also between reproductive plants in different years, but different between vegetative and reproductive plants. In general, reproductive plants were aggregated at larger scales than vegetative plants. For vegetative plants, all L(t) curves peaked around 30–60 cm of distance classes (Fig. 2A). On the other hand, for reproductive plants, all L(t) curves peaked around 100 cm of distance class, and also second peaks of L(t) curves were observed around 200 cm of distance class in 1998, 1999 and 2000 (Fig. 2B).
|
Spatial association between vegetative and reproductive plants
In general, the vegetative plants were significantly attracted (P < 0·01) to the reproductive plants of the previous years (Fig. 3). Peaks of attraction with reproductive plants were observed at the spatial scale of 40 cm for vegetative plants in 1999, 130 cm in 2000 and 90 cm in 2001. For vegetative plants in 2001, a second peak of the L(t) curve was also observed around 240 cm. For vegetative plants in 1999 and 2000, significant repulsion patterns were also observed at large spatial scales with reproductive plants in the previous years (Fig. 3A and B, P < 0·01).
|
Spatial association between plants reproducing in different years
All reproductive plants showed a positive association with reproductive plants in the later years, and the positive associations were statistically significant in most cases. Peaks of the positive associations were generally around 100–200 cm. Reproductive plants in 1998 were significantly attracted (P < 0·01) to reproductive plants in 1999, 2001 and 2002 at various spatial scales (Fig. 4). For reproductive plants in 1999, a significant attraction was observed with all reproductive plants in the later years at the spatial scales around 40–130 cm (Fig. 4; P < 0·01). Reproductive plants in 2000 and 2001 were significantly attracted (P < 0·01) to reproductive plants in 2002 at broad spatial scales (Fig. 4). A significant repulsion at large scales around 350–400 cm between the reproductive plants in 1998 and those in 2000 was observed. Reproductive plants in 2001 also showed significant repulsion around 350–400 cm against those in 1999, and against those in 2000 (Fig. 4; P < 0·01).
|
In bivariate point process analyses, statistical tests according to the random toroidal shift or random re-labelling are often used for detecting patterns of spatial association between two groups. The toroidal shift method randomizes the relative locations between two groups, while within-group spatial structures remain fixed (Diggle, 1983
). However, the random toroidal shift destroys the habitat association if the association between plant density and habitat type is strong (Roxburgh and Matsuki, 1999). The re-labelling method randomly replaces the properties between two groups, while spatial distributions of two groups remain fixed (Dale, 1999). The results obtained from the re-labelling, however, would often show segregated association between two groups if there were aggregated patterns within the groups. Therefore, the results obtained from both the toroidal shift and the re-labelling would make it difficult to infer processes determining the spatial structures of groups in this study. Statistical tests of both the methods were also conducted. The results of the toroidal shift method were generally less significant and the results of the re-labelling method showed segregated patterns (data not shown). |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Both vegetative and reproductive plants of L. rubida showed aggregation within small-gravel areas, which became intensified with growth stages (Tables 2 and 3). These results suggest that patches of smaller gravel are suitable for survival of the vegetative and reproductive plants. A previous study showed that density-dependent mortality due to competition among individuals was higher within the patches of the smaller gravel with high local densities (Suzuki et al., 2003
). Although those two opposing processes affected the spatial patterns of L. rubida individuals, it was concluded that the effect of ground surface conditions was relatively larger than that of intra-specific competition, because habitat association was intensified with progressive life stages.
The spatial distribution of vegetative and reproductive plants of L. rubida in all years was more highly aggregated than expected from the spatial association with ground surface conditions. Vegetative plants in a particular year were predominantly aggregated around the reproductive plants in the year before (Fig. 3). Patchy seed dispersal near the mother plants explains the aggregation of seedlings around reproductive plants (Eriksson, 1994; Ehrlen and Eriksson, 2000). Spatial scales of the aggregations of the vegetative plants of L. rubida were 30–60 cm, which is likely to correspond to the mean distance of seed dispersal. On the other hand, the reproductive plants were aggregated at spatial scales around 100 cm and, secondarily around 200 cm in 1998, 1999 and 2000. The two scales of patterns indicate that clumps of plants were also clumped at larger scales (Dale, 1999). The spatial aggregation of the reproductive plants at two spatial scales would have been induced by patchy seed dispersal and by heterogeneity in ground surface conditions, respectively. The second scale of the aggregation of reproductive plants may indicate the mean size of suitable patches in which plants survive and reproduce. From the results, it is suggested that most of the seeds are likely to be dispersed only within the ‘home’ patch in which they were produced.
The secondary dispersal of seeds, i.e. seeds drifted to the ground by wind, rainfall and spindrifts (Silvertown and Lovett-Doust, 1993; Russell and Schupp, 1998), can be one of other factors determining the spatial aggregation of plants. When seeds of L. rubida were secondarily dispersed, they could gather in the areas with small gravel and/or established plants where seeds are more likely to be trapped. The distribution of dispersed seeds of the grass species Bromus pictus is heterogeneous because vegetation traps seeds during secondary dispersal (Aguiar and Sala, 1997). The spatial heterogeneity in seed capture would enhance the degree of spatial aggregation of the recruited seedlings around reproductive plants and within areas of small gravel.
Most L. rubida plants showed the life cycle of a typical biennial that completes its life within 2 years of germination. In a population of a typical biennial plant, generations flowering in an odd year are expected to be spatiotemporally separated from those in an even year (Kelly, 1985), which may also produce differences in genetic structure between the chronological subpopulations. However, all reproductive plants showed positive association with reproductive plants in the later years at large scales around 100–200 cm, which suggests that different generations constantly occupied identical patches in the habitat. On the other hand, there were patches of small gravel in the habitat that were unoccupied by individuals of L. rubida. Existence of empty patches available for plants suggests limitation of seed recruitment to within the patches (Eriksson and Ehrlen, 1992). Moreover, a negative association between generations at large scales was observed in three cases, suggesting that plants were locally extinct in the occupied patches and/or plants were newly established in unoccupied patches through their regeneration processes. Seed dispersal beyond the ‘home’ patch may have caused establishment in unoccupied patches. Even if the number of seeds is small, seed dispersal beyond the ‘home’ patch would be advantageous for maintaining a local population in a patchy habitat (Green, 1983; Hiebeler, 2000), because seeds can escape from severe competition among siblings (Eriksson and Kiviniemi, 1999) and from risks of local extinction of the ‘home’ patch (Donohue, 1997).
Variations in life history traits among individual plants that break the life cycle of a typical biennial would also have overridden the spatiotemporal separation between the chronological subpopulations. In the study population, a few individuals produced seeds in the third or even later years after their germination. In addition, a seed bank of L. rubida would exist (Y. Ichikawa, Tokyo Metropolitan University, Japan, pers. comm.). Delays in reproduction and a seed bank commonly observed in many biennial plants (Klemow and Raynal, 1981, 1985; Lacey, 1986; de Jong and Klinkhamer, 1988; Kelly, 1989; Klinkhamer et al., 1996) enhance long-term geometric growth rates of populations in spatiotemporally variable environments (Klinkhamer and de Jong, 1983; Kachi and Hirose, 1985; de Jong et al., 1987; Kalisz and McPeek, 1993). The variations in seed dispersal and life-history traits surely play important roles in the dynamics and persistence of a local population of L. rubida.
The patch dynamics of a local population is analogous to the metapopulation dynamics resulting from local population dynamics and regional processes of migration, extinction and colonization (Husband and Barrett, 1996; Hanski and Gilpin, 1997). Freckleton and Watkinson (2002, 2003) argued that patch dynamics within a local population are different from the metapopulation dynamics because of lack of processes that operate at local and regional scales. Hanski (1999) stated that, in a patchy population, frequent dispersal between patches effectively prevents local extinctions and therefore individuals belong to the same local breeding population, although there is no sharp distinction between the patchy population scenario and the classical metapopulation scenario. The patch dynamics of a local population presented in this paper should be non-random processes within and between patches in terms of time and space, which does not mean it is merely an assemblage of local processes restricted within a patch or dynamics of one large population with random mixture of patches. Further studies are necessary to evaluate the significance in order to understand the patch dynamics of a local population in the context of the metapopulation dynamics.
In conclusion, spatial structures and dynamics of a local population of plants in a patchy habitat represent various performances of plants within suitable patches and seed dispersal within a patch and beyond the patch. A careful assessment on the spatiotemporal variation in conditions, the spatial pattern and the scale of patchiness would be necessary before the mechanism regulating the patch dynamics of a local population of plants could be understood.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
We thank the Ministry of Environment and the Department of National Forests in the Ogasawara Islands for permission to study the plants on Minami-jima. We are grateful to Yuriko Ichikawa, Toshio Katsukawa, Asako Takano and Kenji Hata for their help in the field, and also to Fumihiko Sato, Makoto Inaba and other staff members at the Ogasawara Marine Center for providing transport to Minami-jima. Naomi Hosaka and Dr Tomas Herben made constructive comments on an earlier draft of the manuscript. Professor Judy Noguchi edited the English of the manuscript. We are also thankful for helpful comments made by anonymous reviewers. The study was financially supported by a Research Project on Conservation Methods of Subtropical Island Ecosystems, headed by Dr Seiichi Nohara, National Institute for Environmental Studies, Ministry of Environment.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
-
Aguiar MR, Sala OE. 1997. Seed distribution constrains the dynamics of the Patagonian steppe. Ecology 78: 93–100.[CrossRef]
Antonovics J, Levin DA. 1980. The ecological and genetic consequences of density-dependent regulation in plants. Annual Review of Ecology and Systematics 11: 411–452.
Barot S, Gignoux J, Menaut J-C. 1999. Demography of a savanna palm tree: predictions from comprehensive spatial pattern analyses. Ecology 80: 1987–2005.
Casper BB, Cahill Jr JF. 1996. Limited effects of soil nutrient heterogeneity on populations of Abutilon theophrasti (Malvaceae). American Journal of Botany 83: 333–341.[CrossRef]
Cole RG, Syms C. 1999. Using spatial pattern analysis to distinguish causes of mortality: an example from kelp in north-eastern New Zealand. Journal of Ecology 87: 963–972.[CrossRef]
Dale MRT. 1999. Spatial pattern analysis in plant ecology. Cambridge: Cambridge University Press.
Debski I, Burslem DFRP, Palmiotto PA, Lafrankie JV, Lee HS, Manokaran N. 2002. Habitat preferences of Aporosa in two Malaysian forests: implications for abundance and coexistence. Ecology 83: 2005–2018.[Web of Science]
Diggle PJ. 1983. Statistical analysis of spatial point patterns. London: Academic Press.
Donohue K. 1997. Seed dispersal in Cakile edentula var. lacustris: decoupling the fitness effects of density and distance from the home site. Oecologia 110: 520–527.[CrossRef]
Doviak M, Frelich LE, Reich P. 2001. Discordance in spatial patterns of white pine (Pinus strobus) size-classes in a patchy near-boreal forest. Journal of Ecology 89: 280–291.[CrossRef]
Ehrlen J, Eriksson O. 2000. Dispersal limitation and patch occupancy in forest herbs. Ecology 81: 1667–1674.[CrossRef]
Eriksson O. 1994. Seedling recruitment in the perennial herb Actaea spicata L. Flora 189: 187–191.
Eriksson O, Ehrlen J. 1992. Seed and microsite limitation of recruitment in plant populations. Oecologia 91: 360–364.[CrossRef]
Eriksson O, Kiviniemi K. 1999. Evolution of plant dispersal. In: Vuorisalo TO Mutikainen PK, eds. Life history evolution in plants. Dordrecht: Kluwer Academic Publishers, 215–238.
Farley RA, Fitter AH. 1999. Temporal and spatial variation in soil resources in a deciduous woodland. Journal of Ecology 87: 688–696.[CrossRef]
Forget P-M, Mercier F, Collinet F. 1999. Spatial patterns of two rodent-dispersed rain forest trees Carapa procera (Meliaceae) and Vouacapoua americana (Caesalpiniaceae) at Paracou, French Guiana. Journal of Tropical Ecology 15: 301–313.[CrossRef]
Fowler NL, Antnovics J. 1981. Small-scale variability in the demography of transplants of two herbaceous species. Ecology 62: 1450–1457.
Freckleton RP, Watkinson AR. 2002. Large-scale dynamics of plants: metapopulations, regional ensembles and patchy populations. Journal of Ecology 90: 419–434.[CrossRef]
Freckleton RP, Watkinson AR. 2003. Are all plant populations metapopulations? Journal of Ecology 91: 321–324.[CrossRef]
Gibson DJ. 1988. The relationship of sheep grazing and soil heterogeneity to plant spatial patterns in dune grassland. Journal of Ecology 76: 233–252.
Green DS. 1983. The efficacy of dispersal in relation to safe site density. Oecologia 56: 356–358.[CrossRef]
Haase P. 1995. Spatial pattern analysis in ecology based on Ripley's K-function: introduction and methods of edge correction. Journal of Vegetation Science 6: 575–582.
Hanski I, Gilpin ME. 1997. Metapopulation biology: ecology, genetics, and evolution. San Diego: Academic Press.
Hanski I. 1994. Patch-occupancy dynamics in fragmented landscapes. Trends in Ecology and Evolution 9: 131–135.[CrossRef]
Hanski I. 1999. Metapopulation ecology. Oxford: Oxford University Press.
Harms KE, Condit R, Hubbell SP, Foster RB. 2001. Habitat associations of trees and shrubs in a 50-ha neotropical forest plot. Journal of Ecology 89: 947–959.[CrossRef]
He F, Duncan RP. 2000. Density-dependent effects on tree survival in an old-growth Douglas fir forest. Journal of Ecology 88: 676–688.[CrossRef]
Hiebeler D. 2000. Populations on fragmented landscapes with spatially structured heterogeneities: landscape generation and local dispersal. Ecology 81: 1629–1641.[CrossRef]
Horvitz CC, Schemske DW. 1986. Seed dispersal and environmental heterogeneity in a neotropical herb: a model of population and patch dynamics. In: Estrada A, Fleming TH, eds. Frugivores and seed dispersal. Dordrecht: Dr W. Junk Publishers, 169–186.
Houle G. 1992. Spatial relationship between seed and seedling abundance and mortality in a deciduous forest of north-eastern North America. Journal of Ecology 80: 99–108.
Houle G. 1998. Seed dispersal and seedling recruitment of Betula alleghaniensis: spatial inconsistency in time. Ecology 79: 807–818.
Howe HF. 1989. Scatter- and clump-dispersal and seedling demography: hypothesis and implications. Oecologia 79: 417–426.[CrossRef]
Husband BC, Barrett SCH. 1996. A metapopulation perspective in plant population biology. Journal of Ecology 84: 461–469.[Web of Science]
Hyatt LA. 1998. Spatial patterns and causes of overwinter seed mortality in Phytolaccea americana. Canadian Journal of Botany 76: 197–203.
Jackson RB, Caldwell MM. 1993. The scale of nutrient heterogeneity around individual plants and its quantification with geostatistics. Ecology 74: 612–614.[CrossRef]
de Jong TJ, Klinkhamer PGL. 1988. Population ecology of the biennials Cirsium vulgare and Cynoglossum officinale in a coastal sand-dune area. Journal of Ecology 76: 366–382.
de Jong TJ, Klinkhamer PGL, Metz JAJ. 1987. Selection for biennial life histories in plants. Vegetatio 70: 149–156.
Kachi N, Hirose T. 1985. Population dynamics of Oenothera glazioviana in a sand-dune system with special reference to the adaptive significance of size-dependent reproduction. Journal of Ecology 73: 887–901.
Kalisz S, McPeek MA. 1993. Extinction dynamics, population growth and seed banks: an example using an age-structured annual. Oecologia 95: 314–320.[CrossRef]
Kelly D. 1985. On strict and facultative biennials. Oecologia 67: 292–294.[CrossRef]
Kelly D. 1989. Demography of short-lived plants in chalk grassland. I. Life cycle variation in annuals and strict biennials. Journal of Ecology 77: 747–769.
Kenkel NC. 1988. Pattern of self-thinning in Jack pine: testing the random mortality hypothesis. Ecology 69: 1017–1024.[CrossRef]
Kenkel NC, Hendrie ML, Bella IE. 1997. A long-term study of Pinus banksiana population dynamics. Journal of Vegetation Science 8: 241–254.[CrossRef]
Klemow KM, Raynal DJ. 1981. Population ecology of Melilotus alba in a limestone quarry. Journal of Ecology 69: 33–44.
Klemow KM, Raynal DJ. 1985. Demography of two facultative biennial plant species in an unproductive habitat. Journal of Ecology 73: 147–167.
Klinkhamer PGL, de Jong TJ. 1983. Is it profitable for biennials to live longer than two years. Ecological Modelling 20: 223–232.
Klinkhamer PGL, de Jong TJ, de Heiden JLH. 1996. An eight-year study of population dynamics and life-history variation of the ‘biennial’ Carlina vulgaris. Oikos 75: 259–268.
Kotliar NB, Wiens JA. 1990. Multiple scales of patchiness and patch structure: a hierarchical framework for the study of heterogeneity. Oikos 59: 253–260.[CrossRef][Web of Science]
Lacey EP. 1986. The genetic and environmental control of reproductive timing in a short-lived monocarpic species Daucus carota (Umbelliferae). Journal of Ecology 74: 73–86.
Rand TA. 2000. Seed dispersal, habitat suitability and the distribution of halophytes across a salt marsh tidal gradient. Journal of Ecology 88: 608–621.[CrossRef]
Ripley BD. 1977. Modelling spatial patterns. Journal of the Royal Statistical Society, London: Series B 39: 172–212.
Roxburgh SH, Matsuki M. 1999. The statistical validation of null models used in spatial association analyses. Oikos 85: 68–78.
Russell SK, Schupp EW. 1998. Effects of microhabitat patchiness on patterns of seed dispersal and seed predation of Cercocarpus ledifolius (Rosaceae). Oikos 81: 434–443.
Schupp EW. 1995. Seed-seedling conflicts, habitat choice, and patterns of plant recruitment. American Journal of Botany 82: 399–409.[CrossRef]
Schurr F, Bossdorf O, Milton SJ, Schumacher J. 2004. Spatial pattern formation in semi-arid shrubland: a priori predicted versus observed pattern characteristics. Plant Ecology 173: 271–282.[CrossRef]
Silvertown J, Lovett-Doust J. 1993. Introduction to plant population biology, 3rd edn. Oxford: Blackwell Science.
Sterner RW, Ribic CA, Schatz GE. 1986. Testing for life historical changes in spatial patterns of four tropical tree species. Journal of Ecology 74: 621–633.
Stratton DA, Bennington CC. 1998. Fine-grained spatial and temporal variation in selection does not maintain genetic variation in Erigeron annuus. Evolution 52: 678–691.[CrossRef]
Suzuki RO, Kudoh H, Kachi N. 2003. Spatial and temporal variations in mortality of the biennial plant, Lysimachia rubida: effects of intraspecific competition and environmental heterogeneity. Journal of Ecology 91: 114–125.[CrossRef]
Watkinson AR, Feckleton RP, Forrester L. 2000. Population dynamics of Vulpia ciliata: regional, patch and local dynamics. Journal of Ecology 88: 1012–1029.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||