基于网络K函数的西双版纳人工林空间格局及动态

区域植被格局的分布特征受诸多要素影响,但其空间格局和动态具有一定规律或自相关性,道路网络作为景观中显著的人工线性要素,在很大程度上影响着区域的植被格局特征,特别是人工植被的分布特征.运用网络K函数,分析了道路网络和人工林空间格局分布的相互关系,并且用二元网络K函数研究了人工林扩展对针叶林和阔叶林的影响.结果表明:人工林在1970-2000年间种群分布格局有非常明显的变化,特别是从1990到2000年,种群面积不断扩大,主要从北部地区扩展到西北和东南地区.1970-1990年人工林扩展主要集中在低海拔的道路网络附近,沿道路网络呈现明显的集聚分布,公路效应明显.但后期逐渐向距公路较远、海拔较高的地区扩展,到2000年在大尺度下人工林斑块呈显著随机分布.同时,人工林面积的增长对针叶林影响显著,对阔叶林有影响但是并不显著.二元网络K函数表明,在1970到1990年人工林与针叶林沿道路网络在小尺度为负关联,在局部地区存在着竞争,但在大尺度上对环境条件的要求具有一致性为正关联.到2000年,在大尺度上人工林与针叶林的种群分布格局呈显著负相关,人工林面积的不断扩展导致了针叶林面积的下降.     

Anisotropy analyses of population distribution patterns

Direction-dependence, or anisotropy, of spatial distribution patterns of vegetation is rarely explored due to neglect of this ecological phenomenon and the paucity of methods dealing with this issue. This paper proposes a new approach to anisotropy analysis of spatial distribution patterns of plant populations on the basis of the data resampling technique (DRT) combined with Ripley’s L index. Using the ArcView Geographic Information System (GIS) platform, a case study was carried out by selecting the population of Pinus massoniana from a needle-and broad-leaved mixed forest community in the Heishiding Nature Reserve, Guangdong Province. Results showed that the spatial pattern of the P. massoniana population was typically anisotropic with different patterns in different directions. The DRT was found to be an effective approach to the anisotropy analysis of spatial patterns of plant populations. By employing resampling sub-datasets from the original dataset in different directions, we could overcome the difficulty in the direct use of current non-angular methods of pattern analysis. Translated from Acta Scientialrum Naturalium Universitatis Sunyatseni, 2006, 45(2): 83–87 [译自: 中山大学学报 (自然科学版)]     

Spatial patterns on the sagebrush steppe/Western juniper ecotone

Analysis of the spatial patterns of woody plants is important to better understand the ecological processes that govern the worldwide expansion of woody plants across semi-arid ecosystems. Second-order characteristics of a marked spatial point pattern of western juniper (Juniperus occidentalis subsp. occidentalis) were analyzed using Ripley’s K-functions and the pair-correlation function g. The marked point process of crown diameters was produced via two-dimensional wavelet analysis of a fine scale aerial photograph at the woodland-steppe ecotone in the Reynolds Creek watershed in the Owyhee Mountains, southwestern Idaho. Colonization of J. occidentalis stems from mature juniper trees growing in rocky, fire resistant areas. Although these areas introduce components of natural heterogeneity within the landscape, the selected study area is situated within a single soil type, and we modeled the expansion of juniper plants into previously juniper-free sagebrush steppe as a homogeneous point process with constant intensity. Through this research we have identified two statistically significant spatial scales characteristic of J. occidentalis on the woodland/steppe ecotone: (1) We observed inhibition between J. occidentalis plants at distances <15 m, resulting in a regular pattern, rather than clumped or random. This short-distance inhibition can be attributed to competition for water and other resources. Recruitment of young J. occidentalis occurs significantly more often in a direction away from older plants, maximizing the utilization of water and light resources, and perpetuating the spread of the species into previously juniper-free shrub-steppe. (2) J. occidentalis on the ecotone exhibits significant clustering within a 30–60 m radius. Bivariate point pattern analyses provide evidence that, within a distance of 50–70 m, there is a spatial dependence in tree size such that medium trees are more likely than small trees to be close to large trees. We attribute these phenomena to the fact that juniper seeds are commonly dispersed by berry-eating birds with small territories (0.3–1 ha). Beyond a distance of 50–70 m, juniper plants are randomly distributed, suggesting that additional long-range seed dispersal processes are at work. We further acknowledge the importance of including a reference to spatial scale when formulating hypotheses in statistical analysis of spatio-temporal point patterns.     

Sex ratio and spatial pattern of males and females in the dioecious sandhill shrub, Ceratiola ericoides ericoides (Empetraceae) Michx

Sex ratio, size, age, and spatial pattern were investigated for males and females of the dioecious shrub Ceratiola ericoides ericoides (Empetraceae) Michx. within seven mapped populations in Georgia and South Carolina, USA. Among the sites studied, two are regularly burned and one site long-unburned. Age was estimated from node counts of individual shrubs. Only one (fire-suppressed) population showed a female-biased sex ratio, while all others did not differ significantly from 1:1. Mean age estimates did not differ between sexes at any site nor did mean shrub canopy diameter. Bivariate Ripley’s K analysis with a null hypothesis of random labeling was used to investigate whether any of the mapped populations exhibited spatial segregation of the sexes (SSS). No population showed strong evidence of SSS. Rather all sites but one showed males and females to be associated (though not significantly) at a scale of 1–10 m. At a scale of 10–35 m male and female shrubs were located randomly with respect to one another at all sites.     

Spatial pattern analysis in ecology based on Ripley's K-function: Introduction and methods of edge correction

Abstract. Spatial pattern analysis based on Ripley's K-function is a second-order analysis of point patterns in a twodimensional space. The method is increasingly used in studies of spatial distribution patterns of plant communities, but the statistical methods involved are sometimes poorly understood or have been modified without evaluating the effects on results. The procedures of field data acquisition, statistical analysis, and the test for the null hypothesis of complete spatial randomness are described and the presentation of results is discussed. Different methods of edge correction were tested on a computer-generated random pattern and a mapped distribution of a Mediterranean shrubland. The inclusion of buffer zones around mapped plots describes the spatial pattern most accurately, but may not warrant the additional labour involved. Three variations of the weighted edge correction yielded comparable results for the distribution patterns tested. The toroidal edge correction may give biased results for non-random patterns. Recommendations for standardisation of the statistical procedures and data presentation are given.     

A new non‐parametric method for analyzing replicated point patterns in ecology

Most ecological studies that involve point pattern analyses are based on a single plot, which prevent the separation of the effects of various processes that could act simultaneously, as well as limiting the conclusions that can be extracted from these studies. However, considering the spatial distribution of individuals in several plots as replicates of the same process could help to differentiate its specific effects from those of other confounding processes. Thus, we introduce a new method for analyzing spatial point patterns that are replicated according to a two‐factorial design. By summarizing the spatial patterns as K‐functions, the proposed method computes the average K‐functions for each level of the two factors (i.e. predictors) and for each combination of levels, before estimating the sum of squared deviations from the overall mean K‐function. Inferences of the strength of the relationship between the predictors, their interaction, and the spatial structure are made based on a non‐parametric bootstrap procedure, which considers the dependency among spatial scales. We illustrate the proposed approach based on an analysis of the effects of altitude (with two levels: low and high) and slope (with two levels: flat and steep slopes) on the spatial pattern of Croton wagneri, a dominant shrub in an Andean dry scrubland. Our method detected a significant effect of the interaction between slope and altitude, which could not have been detected using current point pattern analysis methodology. The prevalence of single‐plot analysis in ecological studies may be due to a lack of familiarity with appropriate methods for replicated point patterns, as well as the greater complexity of these methods and the absence of appropriate software. Our approach can be applied to a significant number of ecological questions while maintaining a simple, understandable, and easily reportable methodological framework.     

植物空间分布格局中邻体距离的概率分布模型及参数估计

最近邻体法是一类有效的植物空间分布格局分析方法,邻体距离的概率分布模型用于描述邻体距离的统计特征,属于常用的最近邻体法之一。然而,聚集分布格局中邻体距离(个体到个体)的概率分布模型表达式复杂,参数估计的计算量大。根据该模型期望和方差的特性,提出了一种简化的参数估计方法,并利用遗传算法来实现参数优化,结果表明遗传算法可以有效地估计的该模型的两个参数。同时,利用该模型拟合了加拿大南温哥华岛3个寒温带树种的空间分布数据,结果显示:该概率分布模型可以很好地拟合美国花旗松(P.menziesii)和西部铁杉(T.heterophylla)的邻体距离分布,但由于西北红柏(T.plicata)存在高度聚集的团簇分布,拟合结果不理想;美国花旗松在样地中近似随机分布,空间聚集参数对空间尺度的依赖性不强,但西北红柏和西部铁杉空间聚集参数具有尺度依赖性,随邻体距离阶数增加而变大。最后,讨论了该模型以及参数估计方法的优势和限制。     

Neighborhood interactions influencing tree population dynamics in nonpyrogenous boreal forest in northern Finland

Open-canopy moss-rich woodlands dominated by Picea abies and Betula pubescens in northern Finland may undergo cyclic development with reciprocal replacement of the tree species due to the positive feedbacks on soil conditions. Although relations to the abiotic environment are well understood, intra- and interspecific interactions during development of sparse boreal forests have received less attention. We studied tree regeneration, growth and survival with respect to size and density of neighboring trees in four stands representing roughly four stages of the Picea–Betula forest cycle. We conducted spatial analysis (Ripley’s K-function) of mapped locations of live and dead stems to reconstruct the distribution of stems prior to mortality, and to infer possible causes of tree population decline. The prevalence of standing dead stems enabled us to test if mortality was associated with density and size of neighboring trees. Logistic regression was used to test for relationships between tree survival and local crowding indices. We also examined spatial autocorrelations of individual size characteristics to determine the mode and spatial extent of tree interactions. Picea abies had reduced recruitment in open areas occupied by mosses and dwarf-shrubs, and preferentially regenerated near B. pubescens, whereas B. pubescens formed small clumps (and occasionally these consisted entirely of stems from a single tree) that showed local repulsion from large P. abies trees. Size of neighboring trees was the primary determinant of individual growth and survival, whereas neighborhood density per se had only a weak effect. Picea abies had negatively correlated sizes among close neighbors (0–4 m radius) indicating that dominant trees suppress their smaller neighbors. Negative autocorrelations prevailed at the transition stages where the patches of smaller trees were concentrated around evenly spaced large trees. Tree sizes became spatially independent at the mature phase. We hypothesize that both low light and soil nutrient availability causes the P. abies population decline. Dominant trees in this high latitude forest have large light depletion zones and shallow root system to promote strong above- and below-ground competition with younger trees. Higher mortality rates within canopy patches were not compensated for by recruitment in gaps, causing P. abies population to decline steadily.     

庐山森林景观空间分布格局及多尺度特征

森林景观是区域整体景观的重要组成部分,研究其空间分布格局对于优化区域景观整体结构和发挥生态经济效益具有重要意义。基于庐山2010年植被斑块数据,将森林景观划分为常绿阔叶林、落叶阔叶林、松类、杉类和竹类等5类景观,从森林景观的不同发育阶段和林分类型角度出发,运用点格局分析法分析5类森林景观空间分布格局特征。最邻近距离分析表明:5类森林景观空间分布类型均服从集聚分布但聚集强弱有变化;不同发育阶段的森林景观空间分布类型以集聚分布为主,随机分布为辅,尤其幼龄林比较显著,中龄林和老龄林次之;5类森林景观的天然林均服从集聚分布,人工林大多趋于随机分布,只有松类和杉类呈显著集聚分布。Ripley's K函数揭示了不同发育阶段和林分类型的森林景观的多尺度集聚特征,即在小尺度范围内服从随机分布,随着空间距离的增大,以空间特征尺度为分界线,空间聚集强度先逐渐增强,随后不断减弱。总体来看,庐山森林景观的发育阶段主要处于幼年时期,原始植被遭到人类大肆破坏,幼龄林大片分布,属于典型的恢复性植被,未来要重点保护好天然林,减少人为干扰,实现森林景观适度集聚。研究庐山森林景观的空间分布和多尺度特征可以为生态环境保护和实现森林可持续经营提供理论指导。     

Detecting spatial aggregation from distance sampling: a probability distribution model of nearest neighbor distance

Spatial point pattern is an important tool for describing the spatial distribution of species in ecology. Negative binomial distribution (NBD) is widely used to model spatial aggregation. In this paper, we derive the probability distribution model of event-to-event nearest neighbor distance (distance from a focal individual to its n-th nearest individual). Compared with the probability distribution model of point-to-event nearest neighbor distance (distance from a randomly distributed sampling point to the n-th nearest individual), the new probability distribution model is more flexible. We propose that spatial aggregation can be detected by fitting this probability distribution model to event-to-event nearest neighbor distances. The performance is evaluated using both simulated and empirical spatial point patterns.     

Temporal Analogues to Spatial K Functions

In this article, the spatial statistic known as the K function is adapted for temporal processes and patterns. The (optimal) K-function estimator is used in a testing procedure to determine whether behavior patterns of exposed rats versus control rats are different. Specifically, the temporal analogue to the K function is given and an approximately optimal estimator is developed. Next, a testing procedure, to determine whether a group of point patterns is generated from complete temporal randomness, is given. Finally, a testing procedure, to compare pairwise two groups of point patterns to each other, is given. The testing procedures are illustrated with rat-behavior data from both a control-control experiment as well as an exposed-control experiment, where in the latter case a difference in behavior is known to exist.     

Subpopulation genetic structure of a plant panmictic population of Castanea sequinii as revealed by microsatellite markers

Castanea squinii Dode, an endemic tree widely distributed in China, plays an important role both in chestnut breeding and forest ecosystem function. The spatial genetic structure within and among populations is an important part of the evolutionary and ecological genetic dynamics of natural populations, and can provide insights into effective conservation of genetic resources. In the present study, the spatial genetic structure of a panmictic natural population of C. sequinii in the Dabie Mountain region was investigated using microsatellite markers. Nine prescreened microsatellite loci generated 29–33 alleles each, and were used for spatial autocorrelation analysis. Based on Moran’s I coefficient, a panmictic population of C. sequinii in the Dabie Mountain region was found to be lacking a spatial genetic structure. These results suggest that a high pollen-mediated gene flow among subpopulations counteract genetic drift and/or genetic differentiation and plays an important role in maintaining a random and panmictic population structure in C. sequinii populations. Further, a spatial genetic structure was detected in each subpopulation’s scale (0.228 km), with all three subpopulations showing significant fine-scale structure. The genetic variation was found to be nonrandomly distributed within 61 m in each subpopulation (Moran’s I positive values). Although Moran’s I values varied among the different subpopulations, Moran’s I in all the three subpopulations reached the expected values with an increase in distances, suggesting a generally patchy distribution in the subpopulations. The fine-scale structure seems to reflect restricted seed dispersal and microenvironment selection in C. sequinii. These results have important implications for understanding the evolutionary history and ecological process of the natural population of C. sequinii and provide baseline data for formulating a conservation strategy of Castanea species. __________ Translated from Acta Phytoecologica Sinica, 2006, 30(1): 147–156 [译自: 植物生态学报]     

Spatial Characteristics of the Invasion of Acer platanoides on a Temperate Forested Island

We examined the spatial pattern of an introduced population of Norway maple (Acer platanoides L.) on a temperate forested island in order to quantify the influence of landscape context on invasion pattern. The spatial location of every Norway maple tree and sapling (≥0.5 m tall) that had invaded the island forest (n = 4496) was mapped using a global positioning system. The influence of landscape context was examined with the aid of a geographic information system and indices of spatial association. We found that the coniferous forest type was the most heavily invaded (71.9% of all Norway maple stems) when compared to either the hardwood or mixed conifer–hardwood forest types (5.4% and 19.3%, respectively). Across all forest types (excluding urban trees), the population was highly aggregated around roads and other Norway maple trees. For example, 90% of the population was within 40.8 m of a road with an average distance from road of 21.02 ± 0.40 m. This association around roads was significantly greater than would be predicted by chance alone (P < 0.001). Similarly, nearest neighbor distances averaged 4.5 ± 0.2 m with 90% of individuals within 8.3 m of another Norway maple. Measures of spatial association indicated that the invasion was significantly aggregated at both the stand and island scale. Nevertheless, a comparatively small but potentially influential set of individuals were observed at relatively long distances from the main invasion front. Ramifications of these disjunct establishments and other observed patterns are discussed in the context of current spread pattern theory, invasive species monitoring, and control efforts.     

Spatial genetic structure in populations of the terrestrial orchid Orchis cyclochila (Orchidaceae)

Orchid seeds are minute, dust-like, wind-borne and, thus, would seem to have the potential for long-distance dispersal. Based on this perception, one may predict near-random spatial genetic structure within orchid populations. In reality we do not know much about seed dispersal in orchids and the few empirical studies of fine-scale genetic structure have revealed significant genetic structure at short distances (< 5m), suggesting that most seeds of orchids fall close to the maternal plant. To obtain more empirical data on dispersal, Ripley’s L(d)-statistics, spatial autocorrelation analyses (coancestry, f_ij analyses) and Wright’s F statistics were used to examine the distribution of individuals and the genetic structure within two populations of the terrestrial orchid Orchis cyclochila in southern Korea. High levels of genetic diversity (H_e = 0.210) and low between-population variation were found (F_ST = 0.030). Ripley’s L(d)-statistics indicated significant aggregation of individuals, and patterns varied depending on populations. Spatial autocorrelation analysis revealed significant positive genetic correlations among individuals located <1 m, with mean f_ij values expected for half sibs. This genetic structure suggests that many seeds fall in the immediate vicinity of the maternal plant. The finding of significant fine-scale genetic structure, however, does not have to preclude the potential for the long distance dispersal of seeds. Both the existence of fine-scale genetic structure and low F_ST are consistent with a leptokurtic distribution of seed dispersal distances with a very flat tail.     

A Comparative Study on the Power of van Lieshout and Baddeley's J-Function

Summary functions such as the empty space function F and the nearest neighbour distance distribution function G are often used as test statistics for point patterns. Van Lieshout and Baddeley recently proposed an alternative statistic, the J-function, which is defined as J = (1 — G)/(1 — F). Theoretical advantages of the J-function over the F- and G-statistics are that it measures the type, strength and range of interaction, and that it can be evaluated explicitly for a larger class of models. In this simulation study we investigate empirically how the power of tests based on J compares to that of tests based on F and G.     

Spatio-temporal patterns of tree establishment are indicative of biotic interactions during early invasion of a montane meadow

Tree invasions of grasslands are occurring globally, with profound consequences for ecosystem structure and function. We explore the spatio-temporal dynamics of tree invasion of a montane meadow in the Cascade Mountains of Oregon, where meadow loss is a conservation concern. We examine the early stages of invasion, where extrinsic and intrinsic processes can be clearly delineated. In a 0.21-ha plot, we mapped and aged 929 trees ≥0.3-m tall, yielding a detailed record of the spatio-temporal dynamics of invasion. For the primary species, Abies grandis and Pinus contorta, we correlated age structures (unimodal in both species) with climate (precipitation, temperature, and snowpack) and cone production, but found weak or non-significant relationships. Evidence of biotic interactions within and between species was obtained by examining the spatial associations of trees to a distance of 5 m and how these changed over time. We used multiple methods including uni- and bivariate forms of the Ripley’s K and pair-correlation function (pcf) (corrected for inhomogeneity), the J-function, an evolving nearest-neighbor metric, and a test for directional bias in establishment. Pinus and Abies contributed in contrasting ways to the pace and spatial structure of invasion. Shade-intolerant Pinus tended to establish in the open, initiating clusters. In contrast, shade-tolerant Abies established in association with Pinus or in conspecific clusters. Preferential establishment of Abies to the north of older Pinus suggests that facilitation occurs by shading. The factors responsible for initial establishment remain unresolved, but positive interactions are pivotal in accelerating invasions, once initiated. Similar processes are likely to occur in other grasslands undergoing rapid conversion to woodland or forest. In combination, analyses of spatial and temporal patterns of establishment provide insight into the processes that structure invasions.     

Complex spatial structure in a population of <Emphasis Type="Italic">Didymopanax pittieri</Emphasis>, a tree of wind-exposed lower montane rain forest

Didymopanax pittieri is a common shade-intolerant tree colonizing treefall gaps in the elfin forests on windswept ridgecrests in the lower montane rain forests of the Cordillera de Tilarán, Costa Rica. All D. pittieri taller than >0.5 m in a 5.2-ha elfin forested portion of a gridded study watershed in the Monteverde Cloud Forest Preserve were located, mapped, and measured. This local population of D. pittieri is spatially inhomogeneous, in that density increases with increasing wind exposure; D. pittieri are more abundant near ridge crests than lower on windward slopes. The important and ubiquitous phenomenon of spatial inhomogeneity in population density is addressed and corrected for in spatial analyses by the application of the inhomogeneous version of Ripley’s K. The spatial patterns of four size classes of D. pittieri (<5 cm dbh, 5–10 cm dbh, 10–20 cm dbh, and >20 cm dbh) were investigated. Within the large-scale trend in density driven by wind exposure, D. pittieri saplings are clumped at the scale of treefall gaps and at the scale of patches of aggregated gaps. D. pittieri 5–10 cm dbh are randomly distributed, apparently due to competitive thinning of sapling clumps during the early stages of gap-phase regeneration. D. pittieri larger than 10 cm dbh are overdispersed at a scale larger than that of patches of gaps. Natural disturbance can influence the distribution of shade intolerant tree populations at several different spatial scales, and can have discordant effects at different life history stages.     

A systematic comparison of summary characteristics for quantifying point patterns in ecology

Many functional summary characteristics such as Ripley's K function have been used in ecology to describe the spatial structure of point patterns to aid understanding of the underlying processes. However, their use is poorly guided in ecology because little is understood how well single summary characteristics, or a combination of them, capture the spatial structure of real world patterns. Here, we systematically tested the performance of combinations of eight summary characteristics [i.e. pair correlation function g(r), K‐function K(r), the proportion E(r) of points with no neighbor at distance r, the nearest neighbor distribution function D(r), the spherical contact distribution H_s(r), the kth nearest‐neighbor distribution functions D_k(r), the mean distance nn(k) to the kth neighbor, and the intensity function λ( x )]. To this end we used point pattern data covering a wide range of spatial structures including simulated (stationary) as well as real, possibly non‐stationary, patterns on tree species in a tropical forest in Panamá. To measure the information contained in a given combination of summary characteristics we used simulated annealing to reconstruct the observed patterns based only on the limited information provided by this combination and assessed how well other characteristics of the observed pattern were recovered. We found that the number of summary characteristics required to capture the spatial structure of stationary patterns varied between one (for patterns with near random structures) and three (for patterns with complex cluster and superposition structures), but with a robust ranking g(r), D_k(r), and H_s(r) that was largely independent on pattern idiosyncrasies. Stationary summary characteristics [with ranking g(r), D_k(r), H_s(r), E(r)] captured small‐ to intermediate scale properties of non‐stationary patterns, but for describing large‐scale spatial structures the intensity function was required. Our finding revealed that the current practice in ecology of using only one or two summary characteristics bears danger that essential characteristics of more complex patterns would not be detected. The technique of pattern reconstruction presented here has wide applications in ecology.     

生态学中的点格局研究概况及其在国内的应用

点格局分析是研究生态学格局的工具之一,近年来在生态学中的应用越来越多。为深入了解点格局分析方法在国内的研究与应用情况,以所总结的研究进展、一般步骤和基本要点为背景,分析评述了1996—2015年期间以点格局为主题的国内中文核心期刊文献。结果表明,在国内生态学格局研究中,应用研究占据主导地位,研究对象广泛,包括以树木为主的乔、灌、草等不同生活型的植物,甚至包括景观;基础研究,包括概括性统计量、零模型与点过程模型等方面,以及专用软件工具包的开发等研究薄弱。在应用中存在一定问题,主要表现为:概括性统计量使用单一,且以Ripley的K-函数及其变形为主;零模型(或点过程模型)是科学问题的统计表达,但是有一半以上的研究未明确给出零模型。建议在未来应用研究中重视多种统计量的组合使用和原假设的建立,在探讨热带、亚热带森林等具有复杂空间结构系统的多样性格局时,考虑对象的不同世代和系统的不同垂直层次,并加强多变量或三维概括性统计量的开发、点格局分析方法与动态过程模型的结合研究等工作。     

Multi-proxy evidence for competition between savanna woody species

Coexistence of trees and grasses in savannas should be possible if competition between the woody and the grassy components is less intense than the competition within each component. Although several studies have investigated competition between trees and grasses, little is known about tree–tree interactions. We used a multi-proxy approach to examine the spatial pattern of Acacia mellifera and other savanna woody species in a semi-arid savanna in South Africa. Spatial analysis of the point patterns of young and reproductively mature shrubs detected decreasing aggregation with size/age over all spatial scales. This indicated the prevalence of competition although the overall spatial shrub pattern was aggregated. In contrast to point pattern statistics that detect changes only when competition has led to the death of the inferior competitor, we also applied methods identifying the competitive effect on sizes of individual trees. Competition should lead to a negative spatial autocorrelation in size, which we observed in half of the studied cases. Quantile regressions show that nearest-neighbour distance increased steeply with combined size of the target shrub and its neighbours indicating strong competitive effects. The medians of the distributions of maximum root lengths of A. mellifera, of the scale of regular patterns, and of negative autocorrelations were not significantly different, suggesting that overlapping root systems mediate competitive interactions. A competitor removal experiment did not lead to increased shrub sizes, which may be due to the limited duration of the experiment. From the nearest neighbour and autocorrelation analyses, we conclude that competition had a strong impact on growth rates of savanna woody species. Competition-induced mortality only becomes obvious when analysing the shift towards less aggregated spatial patterns when shrubs become reproductively mature. As the overall clustered spatial pattern masks the perceptible effect of competition, a time component should always be included in spatial pattern-based inference of competition.