Environmental drivers of anuran calling phenology in a seasonal Neotropical ecosystem

Temporal variation represents an important component in understanding the structure of ecological communities and species coexistence. We examined calling phenology of an assemblage of anurans in the Gran Chaco ecoregion of Bolivia by deploying automated recording devices to document nocturnally vocalizing amphibians nightly at seven ponds from 20 January 2011 until 31 October 2011. Using logistic regression, we modelled the relationships between temperature, rainfall and photoperiod with calling activity. There was a distinct seasonal effect with calling activity concentrated in the rainy season with no species detected during the dry season from June until the end of October. Calling activity was positively and significantly correlated with photoperiod in 9 of the 10 species analyzed, but there were distinct species‐specific relationships associated with rainfall and temperature. All of these species utilize ephemeral ponds as breeding sites, which can account for their reliance on rainfall as an important driver in calling activity. Two prolonged breeders exhibited similar seasonal breeding patterns across the rainy season, but differed in their response to daily abiotic factors, which might be attributed to the constraints imposed by their reproductive mode. Explosive breeders needed several days of rain to elicit calling. Two pairs of congeners had distinct species‐specific relationships between their calling activity and abiotic factors, even though the congeners shared the same reproductive mode, suggesting that the reproductive modes vary in the constraints imposed on calling activity. The patterns observed suggest that calling phenology of tropical anurans is determined by the interaction of exogenous factors (i.e. climatic variables) and endogenous factors (i.e. reproductive modes).     

Multivariate regression trees for analysis of abundance data

Multivariate regression tree methodology is developed and illustrated in a study predicting the abundance of several cooccurring plant species in Missouri Ozark forests. The technique is a variation of the approach of Segal (1992) for longitudinal data. It has the potential to be applied to many different types of problems in which analysts want to predict the simultaneous cooccurrence of several dependent variables. Multivariate regression trees can also be used as an alternative to cluster analysis in situations where clusters are defined by a set of independent variables and the researcher wants clusters as homogeneous as possible with respect to a group of dependent variables.     

A theoretical analysis of the energetic costs and consequences of parental care decisions

Should a parent care for its young or abandon them before they reach independence? We consider parental care behaviour as an adaptive decision, involving trade-offs between current and future reproduction. The condition of the parent is expected to influence these trade-offs. Using a dynamic programming model we explore how changes in the levels of energetic reserves, and time in the season, determine changes in parental care decisions. The novel feature of our model is that we have included the possibility of remating within the current breeding season in a consistent manner by explicitly modelling the behaviour of unmated animals. We show that there may be several fluctuations in the average duration of care during the breeding season. We also show that, because of the dependence of parental care behaviour on both the condition of the parent and time during the breeding season, changing some of the costs of care may increase the duration of care during one part of the season and decrease it at another. The model also shows that the conditions prevailing for animals with dependent offspring can affect the way in which an unmated animal behaves. For example, the behaviour of unmated animals may change to compensate (partly) for increases in the costs of raising offspring, which are produced at a later date (for example, by increasing the duration of foraging between breeding attempts). Overall, the model provides a good framework for understanding how various ecological and life-history variables should influence parental care behaviour during a breeding season.     

Resources driving landscape‐scale distribution patterns of grazers in an African savanna

In order to effectively manage and conserve indigenous herbivores, a good understanding is needed of how resources drive their distribution patterns. This study employed a unique dataset to test a range of ecological theories and hypotheses on free‐ranging grazers. Using aerial census data collected over 14 yr across the 2 million ha Kruger National Park (South Africa), this study employs spatial autologistic regression models to explore the spatial relationships that exist between the distribution of eight indigenous grazer species and a set of resource variables. It was found that ecological theories relating to feeding guild, water‐dependence, allometric scaling, gut‐morphology and vulnerability to predation could explain most of the grazer distribution patterns observed in relation to surface‐water, forage quality, forage quantity and habitat openness. All the grazers studied were water‐dependent and occurred close to a permanent source of water in the dry season. This was ascribed to the lack of moisture in the diet of grazers during the dry season. Most ruminants’ distribution patterns were significantly associated with areas of high forage quality whereas hind‐gut fermentors were neutral towards forage quality. Average forage quantity was not a significant predictor of long‐term, landscape‐scale distribution patterns for any of the grazer species studied. Most small‐ and medium‐bodied grazers preferred open habitats above closed habitats, probably due to higher visibility and a lower predation risk. Large‐bodied grazers did not bias their distribution patterns towards open habitats. The way in which grazers distribute themselves with respect to different resources can potentially inform management actions on appropriate scales.     

Hierarchical partitioning as an interpretative tool in multivariate inference

Abstract Much of biogeography, conservation and evolutionary biology, and ecology involves very large spatial and temporal extents. Direct manipulation to test hypotheses is usually almost impossible at appropriate scales so that multivariate modelling and especially regression are used to draw causal inferences about which ‘independent’ variables influence the distribution and abundances of species. Such inferences clearly are crucial for the successful management of biological resources and for conserving threatened species. A succession of regression approaches has arisen, many of which yield inconsistent implications. The main problem has been the quest for one (the ‘best’ or the ‘optimal’) regression model from which the impacts of independent variables are inferred. This note is to draw the attention of ecologists to a relatively recent method, hierarchical partitioning, that does not aim to identify a best regression model as such but rather uses all models in a regression hierarchy to distinguish those variables that have high independent correlations with the dependent variable. Such variables are likely to be most influential in controlling variation in the dependent variable. Hierarchical partitioning is not to be regarded as a substitute for experimental manipulation when that is appropriate, but it is likely to produce better deductions than common regression approaches in the many ecological situations in which manipulation is impossible or of doubtful value.     

Rainfall and birthweight distribution in rural Tanzania

This paper explores the possibility of using different techniques of time-series analysis lagged regression analysis for the study of the complicated interaction between ecological, socioeconomic and nutritional variables. The data discussed originate from a rural area in Tanzania, where life is subject to a marked seasonal variation, dependent on a bi-modal rainfall pattern. The unreliability of rainfall is characteristic: both drought and excessive rainfall are frequent occurrences. Seasonal variationns in birthweights over a 10-year period (1972-81) are here compared with the local rainfall pattern. Data on birthweight for this period show that birthweight was highest at the end of the dry season, fell rapidly during the 1st 1/2 of the rainy season, rose again during the later part of the rainy season, and fell to its lowest value just after the rans. It is argued that there are good reasons to attribute these variations to changes in labor output, food consumption, and health status, all of which are related to the seasonal cycle. Mean birthweight fell by about 60g. in the course of the rainy season. Lagged regression analysis reveals a negative impact of rainfall on birthweight after 3 months. Deviations from average rainfall show a positive correlation with birthweight after 4 months, but only in the months when rainfall is critical for food crop production. The interactions of these contrary effects are held responsible for the observed sudden fluctuations in mean birthweight.     

Partial least squares regression as an alternative to current regression methods used in ecology

This paper briefly presents the aims, requirements and results of partial least squares regression analysis (PLSR), and its potential utility in ecological studies. This statistical technique is particularly well suited to analyzing a large array of related predictor variables (i.e. not truly independent), with a sample size not large enough compared to the number of independent variables, and in cases in which an attempt is made to approach complex phenomena or syndromes that must be defined as a combination of several variables obtained independently. A simulation experiment is carried out to compare this technique with multiple regression (MR) and with a combination of principal component analysis and multiple regression (PCA+MR), varying the number of predictor variables and sample sizes. PLSR models explained a similar amount of variance to those results obtained by MR and PCA+MR. However, PLSR was more reliable than other techniques when identifying relevant variables and their magnitudes of influence, especially in cases of small sample size and low tolerance. Finally, we present one example of PLSR to illustrate its application and interpretation in ecology.     

宁夏生态足迹影响因子的偏最小二乘回归分析

生态足迹分析方法是一种度量区域生态可持续程度的有效方法,偏最小二乘回归法(PLS)能有效解决多元回归分析中变量的多重相关性问题,具有容易操作,相关分析精度高等特点。以宁夏为研究区域,在计算了宁夏2001—2010年人均生态足迹的基础上,应用偏最小二乘回归分析法,对影响宁夏生态足迹的各因子的重要程度进行了分析。通过变量投影重要性分析、特异点分析和预测分析,证明所得偏最小二乘回归模型具有较好的精度。研究结果为:2001—2010年,宁夏人均生态足迹由1.818103793 hm2上升至2.894958909 hm2,生态赤字由1.28352051 hm2上升至2.42316627 hm2,生态承载力由0.53458328 hm2下降至0.47179264 hm2;全区GDP、城镇居民人均生活消费支出、第二产业产值和第一产业产值是影响宁夏生态足迹的显著因子。     

Using multi-target clustering trees as a tool to predict biological water quality indices based on benthic macroinvertebrates and environmental parameters in the Chaguana watershed (Ecuador)

Macroinvertebrates were sampled in the Chaguana river basin in SW Ecuador in the wet season (March) and the dry season (September) of 2005 and 2006. To assess the robustness of several biological indicators, correlations were calculated between both years and between the wet and the dry season. In addition, it was tested if the indices gave significantly different results for sites with a bad, poor, moderate and good ecological water quality. Composition measures performed poorly in most cases, however, abundance, diversity and richness measures often performed better and tolerance measures, the so-called biotic indices, performed very well, even indices developed for temperate regions. By using pruned multitarget clustering trees, it was possible to predict several well-performing ecological water quality indices simultaneously on the basis of the occurring key macroinvertebrate taxa or, alternatively, on the basis of key environmental variables. In contrast to unpruned trees, which resulted in complex trees that were difficult to interpret and performed inferiorly, pruning resulted in transparent trees. Water quality indices scored high when Hydropsychidae were present and even higher when in addition also Megapodagrionidae were present. When no Hydropsychidae nor Libellulidae were present, the indices reached the lowest scores. However, this model based on key taxa occurrences did not perform well during validation. Water quality indices scored higher with increasing dissolved oxygen concentrations and a strong current velocity. The latter model based on environmental variables also performed well during validation. In the presented study, the ecological water quality could thus be accurately predicted solely on the basis of dissolved oxygen concentration and current velocity. It can therefore be concluded that multitarget clustering trees can be easily used as a practical tool for cost-effective decision support by water quality managers.     

Weekly carbon dioxide exchange trend predictions in deciduous broadleaf forests from site-specific influencing variables

Weekly net-ecosystem-exchange (NEE) data recorded and verified over multiple years, together with site-specific influential ecological variables, for distinct deciduous-broadleaf-forest (DBF) sites in North America can reveal useful relationships regarding their functions as long-term carbon dioxide (CO2) sources and sinks. Machine learning (ML) and regression models have greater success at predicting weekly NEE from some DBF sites than others, from the available site variables. In particular, support vector regression (SVR) and extreme gradient boosting (XGG) ML methods perform better than multi-linear regression in the weekly NEE predictions they generate using up to 24 influential variables. The DBF sites studied are distinguished into those that have followed distinctive, albeit fluctuating seasonal NEE trends, and those that are characterized by abrupt fluctuations in NEE across the leaf-on season. ML models predict weekly NEE for the former sites more reliably than for the latter sites. Consideration of the relative influence of the variables on the XGB and regression model NEE predictions identifies which variables are most influential at specific sites. Short wave radiation (in and out) and air temperature are found to be variables exerting substantial influence on the prediction models for the sites studied. From the prediction results and the relative influences of the available environmental variables, it is concluded that complex processes are involved at those sites showing rapid NEE fluctuations in the leaf-on seasons that are not readily detectable from the environmental variables currently being continuously recorded at those sites.     

Relative importance of demographics, locale, and seasonality underlying louse and flea parasitism of raccoons (Procyon lotor)

A variety of demographic, seasonal, and site-specific variables may influence parasitism, but the relative importance of these variables is generally unclear. We measured the relative ability of host characteristics, season, and site to explain louse (Trichodectes octomaculatus) and flea (Orchopeas howardi) infestation across 10 populations of raccoons (Procyon lotor). Lice are highly dependent on specific hosts and are predicted to display a relatively strong relationship with factors intrinsic to the host, when compared to fleas, which can infest multiple species and survive off-host for weeks without feeding. We developed a priori models that represented explicit hypotheses and contrasted their ability to predict infestation patterns. While the abundance of lice was seasonal, models that included solely host age and sex best predicted prevalence and abundance, in part because males were infested with 3 times the number of lice than were females. Conversely, flea prevalence and abundance, which peaks sharply in the spring, was best predicted by season; factors intrinsic to the host were relatively unimportant for predicting abundance. These, and other, recent findings emphasize the need to simultaneously assess the relative importance of multiple ecological variables between parasite species when attempting to describe general trends and constraints of host-parasite associations.     

水曲柳落叶松人工混交林中树木个体生长的竞争效应模型

本研究从水曲柳落叶松混交林中树木的种内种间竞争机制出发,对与距离有关的竞争模型和邻体干扰模型进行了改进,提出了以竞争指数分量和竞争效应系数来量化树木间的空间竞争。根据野外的实测数据,以年材积生长量为因变量各竞争指数分量为自变量,通过多元回归分析建立了树木个体生长的竞争效应模型,回归结果有很好的拟合优度。本研究为空间竞争模型的进一步改进和生态学研究植物竞争生长关系提供了新的思路。     

Co‐declining mammal–dung beetle faunas throughout the Atlantic Forest biome of South America

The millennial–scale evolutionary relationships between mammals and dung beetles have been eroded due to several drivers of contemporary biodiversity loss. Although some evidence of co‐decline has been shown for mammals and dung beetles at some Neotropical sites, a biome‐scale analysis for the entire Atlantic Forest of South America would strengthen our understanding of how relictual sets of mammal species can affect dung beetle co‐occurrences and co‐declines. We therefore collated hundreds of assemblages of both dung beetles and medium‐ to large‐bodied mammals throughout the world's longest tropical forest latitudinal gradient to examine to what extent mammal assemblages may exert a positive influence on dung beetle species composition and functional assembly, and whether this relationship is scale dependent. We also collated several climatic and other environmental variables to examine the degree to which they shape mammal–dung beetle relationships. The relationships between local mammal and dung beetle faunas were examined using regression models, variation partitioning, dissimilarity indices and ecological networks. We found a clear positive relationship between mammal and dung beetle species richness across this forest biome, indicating an ongoing process of mammal–dung beetle niche‐mediated co‐decline. We found a strong relationship between the species composition of both taxa, in which dung beetle species dissimilarity apparently track changes in mammalian dissimilarity, typically in 80% of all cases. Co‐variables such as phytomass and climatic variables also influenced mammal–dung beetle patterns of co‐decline along the Atlantic Forest. We conclude that dung beetle diversity and community assembly are shaped by the remaining co‐occurring mammal assemblages and their functional traits, and both groups were governed by environmental features. We emphasize that ecosystem‐wide effects of mammal population declines remain poorly understood both quantitatively and qualitatively, and curbing large vertebrate defaunation will ensure the persistence of co‐dependent species.     

Dung beetle assemblages organization in two contrasted areas of the Mediterranean region: affinities and divergences

Quantification of the species richness and diversity is important when comparing sites; these variables are influenced by local and regional factors. The structure of assemblages of dung beetles as well as their species composition and diversity were compared between two sites on both sides of the Mediterranean Sea (France FR and Tunisia TN). From a total of 132 traps, 15,778 specimens were collected belonging to 86 species (46 in FR vs. 40 in TN). Species composition varied highly between FR and TN (only 10 species in common). The temporal occurrence of Aphodiidae and Scarabaeidae showed different patterns due to several factors: phenology of species, geographical distribution and local ecological requirements. When quantifying the beta diversity, the turnover of species was for almost cases higher within the same season than between seasons. Moreover, seasonal turnover was different at both sites. Evenness and rank-abundance analyses demonstrated that autumn-winter was the most constraining season in FR, with one dominant species (Agrilinus constans) in winter assemblages (96.8% of individuals). Conversely summer was the most constraining season in TN, with Onthophagus taurus as dominant species in assemblages (87.7%). The influence of local and historical factors is discussed.     

Breeding ecology of Kori Bustard Ardeotis kori strunthiunculus in the Serengeti National Park

The breeding ecology of the Kori Bustard Ardeotis kori strunthiunculus was studied in the plains of the Serengeti National Park, Tanzania in 2014 and 2015. Random transects were used to search for male courtship displays, nests, chicks and subadults. GPS satellite collars were used to locate nesting females. Linear regression analyses and post hoc tests were used to determine the predictors that contributed most to the variation of the dependent variables (courtship display, nest, chicks and subadults). The results indicate that courtship behaviour peaks during the short dry and short rainy season before the peaks in nests and chicks. The highest nest frequency was found in short grass habitats. Female Kori Bustard may undergo repeated nestings within a single breeding season. The adult sex ratio was female skewed during the breeding season. The Kori Bustard breeding season in the Serengeti plains is relatively long, lasting for almost nine months, and taking place during all seasons except for the long dry season. We recommend that management authorities conduct assessments of Kori Bustard recruitment as well as habitat suitability in the Serengeti ecosystem to develop future conservation strategies.     

Using harvest data to predict habitat-population relationship of the wild boarSus scrofa in Northern Italy

The relationships between the wild boarSus scrofa, Linnaeus 1758 density and the habitat characteristics were investigated by collecting bag records in 18 hunting areas during the 1995 and 1996 hunting seasons. The study areas were located in the Northern Apennines (N-Italy) from 470 to 1160 m a.s.l. In each study area a Density Index (DI) was calculated by weighting the number of wild boar culled in each hunting season on the number of hunting days and on the surface of the study areas. Moreover 16 habitat variables were measured in each study area. Correlation and regression analyses were performed considering the DI as a dependent variable. Mixed deciduous woods were positively correlated with wild boar density together with chestnuts woods and conifer reafforestations. No significant negative correlations were found. Twelve multiple regression models each including three habitat variables were ranked by the information-theoretic approach (modified Akaike’s information criterion). The study stressed the importance of mixed woods for wild boar and the usefulness of hunting data as a mean to predict the habitat suitability for the species.     

Structure of populations and ecological nishes of ectoparasites in the parasite communities of small forest mammals

The paper reports the results of eight-year investigations on the ectoparasites of rodents and insectivores carried out in southern taiga of the Ilmen-Volkhov lowland (Novgorod Region) and Kurgolovsky reserve (Leningrad Region). Twelve species of small mammals were captured including three dominate species--bank vole Clethrionomys glareolus (2722 specimens), common shrew Sorex araneus (1658 specimens), and wood mouse Apodemus uralensis (367 specimens). Parasite community of the bank vole comprises 34 species of mites, ticks, and insects, the community of common shrew comprises 25 species, and the community of A. uralensis includes 28 species. Taxonomic diversity of the ectoparasite communities was shown to be based on the diversity of types of parasitism and ecological nishes of the host body. Permanent ectoparasites are found to be represented by 2 species of lie and 14 species of acariform mites. The group of temporary parasites includes 13 species of fleas, 10 species of gamasid mites. 3 ixodid species and 1 Trombiculidae. There is a common pool of temporary parasites of small mammals in the ecological system of taiga. Significance of different shrew and rodent species as hosts were found to be dependent on the population density in possible hosts and many other factors. Species diversity in the parasite communities of different small mammal species is dependent on the number of possible ecological nishes in the host body. Actual infill of these nishes by ectoparasites is usually lesser than potential one. Species composition of temporary parasites, their occurrence and abundance changes according to season. Interspecific competition in the temporary parasite species can decrease because of the seasonal disjunction of their population peaks. Diversification of the ecological niches of ectoparasites allow simultaneous feeding of more parasite individuals on one host, than in the case of parasitising of single species or several species with similar ecological nishes. The distribution of parasites on their hosts was also studied. The aggregative distribution has been found in ixodid larvae only, and the distribution of fleas was close to the Poisson distribution. Deviations from the aggregative distribution can be an effect of several independent factors, including limited ability of small mammals for providing numerous parasites with food. On the most part of hosts simultaneous parasitizing of no more than 1-3 individuals of each tick, mite, and flea species was registered. Excessive infestation by ectoparasites may probably be limited by effective reactions of self-purification in the mammal hosts.     

The TRM Model of Potential Natural Vegetation in Mountain Forests

Due to advances in spatial modeling and improved availability of digital geodata, traditional mapping of potential natural vegetation (PNV) can be replaced by ecological modeling approaches. We developed a new model to map forest types representing the potential natural forest vegetation in the Bavarian Alps. The TRM model is founded on a three-dimensional system of the ecological gradients temperature (T), soil reaction (R), and soil moisture (M). Within such a “site cube” forest types are defined as homogenous site units that give rise to forest communities with comparable species composition, structure, production and protective functions. The three gradients were modeled using regression algorithms with area-wide, high resolution geodata on climate, relief and soil as predictors and average Ellenberg indicator values for temperature, acidity and moisture of vegetation plots as dependent variables summarizing plant responses to ecological gradients. The resulting predictor-response relationships allowed us to predict gradient positions of each raster cell in the region from geodata layers. The three-dimensional system of gradients was partitioned into 26 forest types, which can be mapped for the whole region. TRM-based units are supplemented by 22 forest types of special sites defined by other ecological factors such as geomorphology, for which individual GIS rules were developed. The application of our model results in an intermediate-scale map of potential natural forest vegetation, which is based on an explicit function of temperature, reaction and moisture and is therefore consistent and repeatable in contrast to traditional PNV maps.     

Limnological and ecological characteristics of tropical highland reservoirs in Tigray, Northern Ethiopia

The semi-arid highlands of Northern Ethiopia (Tigray) have numerous small reservoirs that have been created by microdams in an attempt to provide water supply for irrigation and livestock drinking. Although the reservoirs have substantial added value to residents, their use as water resource is jeopardized by eutrophication and a high occurrence of blooms of toxic cyanobacteria. So far, there is no systematic information available on the limnological and aquatic ecological characteristics of these dams. We carried out a standardized survey of 32 reservoirs and assessed a wide set of morphometric, abiotic and biotic variables. The sampling was performed during two seasons, September–October 2004 (end of the wet season/start of the dry season) and April–May 2005 (towards the end of the dry season). Using multivariate analysis, we revealed dominating patterns of variable associations and compared the variability in these patterns among seasons. According to standardized PCA and RDA analyses, the most important axis of variation was mainly represented by a gradient in nutrients and altitude that was also positively associated with phytoplankton biomass, suspended matter and oxygen concentration, and negatively with water transparency. For most variables, correlations between the wet and dry season were weak, which suggests that individual reservoirs behaved rather differently in their response to seasonal changes. Nevertheless, a Mantel correlation (r = 0.32; P = 0.035) showed a weak but significant overall concordance in the variable association patterns among seasons. A number of reservoirs became very shallow or fell dry in the dry season, a process that was associated with an increase in suspended matter and conductivity and a decrease in transparency. These reservoirs contained lower amounts of fish and tended to be less eutrophic than the deeper, permanent systems, as they had lower levels of phosphorus and chlorophyll-a. With multiple regression analysis, we constructed most parsimonious models in an attempt to explain the variation in key biotic variables: phytoplankton and cyanobacteria biomass, cladoceran biomass, fish biomass and the abundance of submerged vegetation. Phytoplankton and fish biomass tended to be positively related with the concentration of total phosphorus, whereas cladoceran biomass was not associated with nutrient concentrations. The positive association of fish and phytoplankton with nutrient concentrations suggests a bottom-up control, whereas the absence of an association between zooplankton and nutrient concentrations may be indicative for top-down control. The biomass of cyanobacteria was negatively related to the biomass of cladocerans (Daphnia), which likely reflects a top-down effect. Most reservoirs were turbid. The occurrence and abundance of macrophytes tended to be positively related to water transparency and was negatively associated to TP and the amount of livestock frequenting the reservoirs. However, macrophytes were not limited to clear-water reservoirs. Handling editor: L. M. Bini     

Red-shifts and red herrings in geographical ecology

I draw attention to the need for ecologists to take spatial structure into account more seriously in hypothesis testing. If spatial autocorrelation is ignored, as it usually is, then analyses of ecological patterns in terms of environmental factors can produce very misleading results. This is demonstrated using synthetic but realistic spatial patterns with known spatial properties which are subjected to classical correlation and multiple regression analyses. Correlation between an autocorrelated response variable and each of a set of explanatory variables is strongly biased in favour of those explanatory variables that are highly autocorrelated - the expected magnitude of the correlation coefficient increases with autocorrelation even if the spatial patterns are completely independent. Similarly, multiple regression analysis finds highly autocorrelated explanatory variables "significant" much more frequently than it should. The chances of mistakenly identifying a "significant" slope across an autocorrelated pattern is very high if classical regression is used. Consequently, under these circumstances strongly autocorrelated environmental factors reported in the literature as associated with ecological patterns may not actually be significant. It is likely that these factors wrongly described as important constitute a red-shifted subset of the set of potential explanations, and that more spatially discontinuous factors (those with bluer spectra) are actually relatively more important than their present status suggests. There is much that ecologists can do to improve on this situation. I discuss various approaches to the problem of spatial autocorrelation from the literature and present a randomisation test for the association of two spatial patterns which has advantages over currently available methods.