Training and Mobility of Researchers Network

• acquire an improved understanding of the variation in the biology and dynamics of populations living under distinctly dissimilar environmental conditions

• substantially advance understanding of demographic and genetic causes of population and metapopulation extinction in fragmented landscapes

• construct parameterized models of metapopulation dynamics to predict the consequences of habitat fragmentation

• study the evolutionary and genetical processes taking place in populations living in fragmented landscapes

General mission

General research objectives

Current international state-of-the-art

Conceptual foundations. The metapopulation concept has become well-established in population biology and conservation, but there continues to be disagreement about the types of spatially structured populations that it can be legitimately applied to. There is a widely recognized need to forge a synthesis between landscape ecology and metapopulation ecology, but no breakthrough has yet been made in this direction. The metapopulation concept is widely applied in conservation; it has served the very useful purpose of drawing attention to regional processes, but doubts remain about the appropriate domain of applications.

Metapopulation theory. Deterministic metapopulation theory is well-developed for single species and for two-species predator-prey and competitive interactions. Some stochastic models demonstrate substantial promise as tools for modelling real metapopulations. Significant progress has been made in the construction of more complex theory, based on the theory of structured population models. Genetic consequences of metapopulation dynamics have been modelled both analytically and using simulation, often aimed at understanding the level of genetic variability that a particular metapopulation structure can retain.

Metapopulation processes. Key metapopulation processes include population extinction, population establishment, and generally the role of migration in population dynamics and in adaptive evolution. The role of genetic processes in population extinction remains poorly understood; the role of ecological processes is relatively well understood. Measurement of migration rate remains a great challenge for field studies, and the role that migration and spatial processes in general play in ecology and evolution remains a topical research objective.

Field studies. The empirical literature on metapopulation biology has increased rapidly during the past five years. Ecological consequences of habitat fragmentation have been documented for a large range of taxa. The increasing use of high-resolution DNA markers has facilitated genetic studies. There remains a great potential for more comprehensive metapopulation-level studies of population biology.

Contributions expected from this project to the advancement of the study of the dynamics and evolution of species in fragmented landscapes

Exceptional data base for comparative analyses and modelling. We plan to conduct well co-ordinated field work on metapopulations of four butterfly species living in more or less fragmented landscapes in the centre and towards the margins of their geographical ranges in Europe. These data will be superior in terms of quantity and quality, and will form a solid basis for data analysis and modelling. The data will be unique in that we will establish protocols to ensure exact comparability of data from different regions and species, providing an unrivalled opportunity to contrast the structures of populations, both within and between species. This should provide a major advance in this field, which is bedevilled by a lack of conformity in data collection, making direct comparisons among studies difficult or impossible.

Modelling closely integrated with field work. We design the field work in the context of on-going theoretical work on metapopulation models, and vice versa. This research network has been assembled of teams that are used to cross the theory- empirical work boundary, and we will build upon the previous experiences and successes. We will use models paramaterised from one region to make specific predictions about the distribution of populations in other regions, based solely on the distributions of habitats. We will also run the models by using parameters of one species on the habitat patterns of the other species to evaluate quatitatively the extent to which differences in metapopulation distributions of different species can be attributed to dispersal and reproductive patterns of each species, as opposed to the habitat networks available to each. This rigorous testing of the ability of models to predict distributions accurately from region to region, and from species to species, will form a valuable contribution in the drive to make ecology an increasingly precise and predictive science.

Meaningful integration of ecological, genetic, and evolutionary studies. The network includes teams with varied but sufficiently overlapping interests to allow us to expect successful integration of ecological, genetic, and evolutionary studies. A particular focus of the planned research is on migration, which involves significant ecological, genetic, and evolutionary problems. For example, the rate of migration among habitat patches within a metapopulation may evolve depending on the level of habitat fragmentation in a particular region, because the probability of finding a new patch successfully, and the probability that the discovered patch will be vacant, will have substantial effects on the costs and benefits of migration. If migration rate evolves in response to fragmentation, this will affect the proportion of habitat patches occupied at equilibrium, which in turn may affect the further evolution of migration rate. In some cases, individual selection within metapopulations may even contribute to the risk of extinction by reducing migration rate.

Sound biological basis for conservation of species living in fragmented landscapes . As population biologists, we are anxious to see the results of our science being appropriately applied to conservation and landscape management. The comprehensive approach taken in this research network will include an explicit consideration for the implications of our results for conservation. Development of realistic models that can be applied to real systems will provide quick and cheap means of assessing the potential impacts of alternative landscape management options. By the end of the programme, we hope that the models will be sufficiently refined and tested that it will be possible to start applying them routinely to conservation and landscape management problems throughout Europe and beyond.

Exceptional training opportunity for young European researchers and contribution to other EU science programmes. The research programme will help to train post docs in and fulfill European Commission research goals under the Terrestrial Ecosystem Research Initiative, Science Plan (TERI). Training and research will contribute specifically to areas (1) Effects of land-use change, (4) Biodiversity, population biology and ecosystem functioning, and (5) Integration, upscaling and scenario studies. It will make a major contribution towards overall TERI investigations of large-scale temporal and spatial dynamics. Our TMR network should also contribute to the establishment of a scientific basis for national and EU environmental policies, and train personnel in this area, with direct potential benefits for the EU Climate and Environment and other programmes. Post-docs will benfit directly from contacts throughout the consortium, and also through the much larger numbers of professional contacts between members of the network and researchers throughout Europe and elsewhere in the world.

International leadership in metapopulation biology. In brief, we envision that this network will comprise the most significant research consortium on metapopulation biology world-wide in the years 1998-2000. We aim at a significant scientific impact in this field.

Research strategy: combination of skills

We have selected a model group of species to conduct the bulk of empirical research (task #1). Three other specific research tasks are supported by the results of these studies. The implications of the results for conservation and management are analysed in task #5, which comprises the ultimate aim of the project.

Our research strategy involves combining the skills of the research groups with complementary interests. Four teams have first-rate expertise in field population biology (HKI, LDS, UCL and COR). One team has specialised in evolutionary studies of metapopulations (MNP) and another team has done pioneering population genetic work in the area (LDN). Two teams have made first-rate contributions to modelling of metapopulations (HKI, LPZ). All teams share a common interest in the biology and conservation of species living in fragmented landscapes.

The model group of species

We have chosen as the core of the network (task #1) a comparative study of four species of butterflies, which will be used as convenient model organisms to investigate general questions. The four species are Melitaea cinxia, Proclostiana eunomia, Plebejus argus and Maniola jurtina (see task #1, section 5, Work plan). Five of the seven research teams have conducted and continue to conduct research on butterflies. We have chosen butterflies for the following reasons:

• Butterflies are exceptionally well known in Europe, including their taxonomy, geographical distributions, life histories, and history of occurrence in many countries

• Butterflies provide important practical advantages for laboratory and field work

• Laboratory rearing of butterflies is easy; their complete life cycles can be studied in laboratory conditions; large numbers of individuals can be reared simultaneously under various experimental conditions

• Field work: adults and larvae are closely related to their host-plants; suitable habitats are rather homogenous and occur in discrete patches which can be located a priori; adults are conspicuous, patches are exhaustively sampled; adults, eggs and late caterpillars may be monitored and handled in the field, allowing field experiments; individual characteristics related to reproductive condition (age, sex, virginity), can be evaluated in the field; population parameters, like density and sex-ratio, can be easily manipulated.

• Butterflies have suffered more than perhaps any other group of animals of the recent environmental changes

Documented rates of butterfly extinction now equal or exceed vertebrate extinctions in the few parts of Europe where butterfly recording effort is anywhere near as high as for birds and mammals. In Britain, which we use as an example because of the detailed records available rather than because it is unusual, 8% of butterfly species have become extinct, and 20% are listed as threatened, vulnerable, or endangered in red data books. Widespread declines and extinctions are predominantly due to the agricultural and silvicultural revolutions of the 20th century. These changes are continuing, and increasingly affecting Europe (not just the EU). Unless adequate, large-scale metapopulation conservation measures are put in place very soon, we can expect to see large numbers of species contracting to small areas or becoming extinct entirely, thus reducing the overall diversity within any region. Since each butterfly or group of butterflies is characteristic of a particular type of habitat, large numbers of other invertebrates, vertebrates and plants, can be expected to benefit from successful measures taken to maintain butterfly biodiversity. Advances in butterfly conservation research have repeatedly provided a stimulus for subsequent developments in the conservation of other invertebrate taxa, and we expect this to continue through the proposed programme.

Methodology

The seven research teams have developed and successfully applied the following relevant metholodogies:

The main field method is mark-release-recapture (MRR). Adults are caught, individually marked and released. Subsequent recaptures will (1) indicate movements from their previous release point and (2) yield a capture history for each individual, which allows the accurate estimation of demographic parameters like population size and survival estimates. Other parameters, such as phenology of reproduction, can be computed from such data. Estimation of these parameters for several generations provides valuable information about the dynamics of local populations. Besides MRR, other field methods are commonly used: (1) adult and larval tracking, which allows the recording of their movements and associated behavioural patterns, (2) egg, larval or adult transect counts and (3) egg, larval or adult translocations.

Laboratory stocs of species of butterfly, especially Bicyclus anynana, which has a short generation time and no viral pathogens, have provided critical insight about evolutionary processes which are difficult to probe equally pecisely with field studies. Genetic variances for different types of traits and for fittness components can be followed buring the evolution of captive populations enabling the relevance of population genetics theory to be explered and tested. Replicated captive (meta)populations can be manipulated to examine the importance of population size and turnover, and of gene flow,to population persistance.

Our groups (especially MNP and HKI) have expertise in studying the genetic structuring of fragmented populations of plants and animals using allozymes, RAPD, microsatellites and other DNA markers. We use spatially explicit metapopulation models to study the consequence of various life-history attributes (overlapping generations, mating systems) on effective population size. Research methology involves mathematics of identity coefficients and computer simulations. Theoretical work on the evolution of migration rate, reproductive effort and mating systems in fragmented habitats involves ESS theory and computer Simulations.

A standard method of evaluating stochastic simulation models for extinction has been developed. By this method the transient state of a population can be separated from the established state and a general quantification of the risk of extinction applicable for conservation problems is achieved. Based on this method for local population dynamics, a standard approach for stochastic, spatially explicit modelling of finite metapopulations has been developed. This is applicable for metapopulations with constant properties of the patches. For metapopulations in dynamic landscapes with resource dynamics or rare catastrophic events a useful simulation technique for metapopulations has been developed. An alternative simpler approach (incidence function modell) for modelling metapopulation dynamics is based on the first-order effects of fragment size on extinction and isolation on colonisation. The modell can be parameterized with simple presence-absence data, and the model makes predictions about transient and steady-state metapopulation dynamics.

Introductions of butterflies to formerly empty networks of habitat patches provide a powerful means of assessing metapopulation structure and dynamics. These experiments are rarely attempted because of the effort required, and the time-scale of experiments may be long. However, three groups (LDS, LON, HKI) have taken part in large-scale releases of butterflies into empty habitat networks. Such experiments are extremely revealing of dispersal distances achieved immediately following release. In the longer term, they reveal long-distance colonizations which are crucial to colonization and gene flow. The proposed programme will be able to take advantage of continued monitoring of on-going metapopulation establishment experiments in Finland (HKI), UK (LDS) and France (monitored by LON), established in the past: age varies from a few years to over 50 years from introduction. These experiments provide crucial tests of the ability of metapopulation models to predict spatial changes in the distribution of populations accurately.

The research conducted in this network has been structured into five distinct but interrelated research tasks. (see graph below). Task #1 provides the new empirical data basis for the network. Tasks #2 and 4 take advantage of the knowledge base, and focuses on three particular problems: evolution of life histories in fragmented landscapes, inbreeding depression in metapopulations, and development of predictive metapopulation models. All the work feeds to task #5, the main aim of the network, development of a sound framework for the conservation of species and biodiversity in the fragmented European landscapes.

The five research tasks

TASK 1 Comparative study of central and marginal metapopulations living under different degrees of habitat fragmentation

Objectives and approach. To characterize the biological differences of conspecific central and marginal metapopulations living under different levels of habitat fragmentation. A particular focus and a novel aspect of these studies is a concerted effort to reveal the causes of population extinction under a range of environmental conditions, including demographic, environmental, and genetic causes. For this task, we have selected four butterfly species: Melitaea cinxia, Proclossiana eunomia, Plebejus argus and Maniola jurtina. These species represent some of the best studied butterfly species in Europe, largely due to previous work by the partners of this network.

Description of task. Identify and describe patch networks supporting central and marginal metapopulations in more or less fragmented landscapes (2x2 networks for each species). Describe patterns of patch occupancy and local density. Observational and experimental studies on the causes of population extinction. The field sites will be selected from areas close to the home institutions of the network partners.

Objectives and approach. To develop spatially realistic metapopulation models that include demographic and environmental stochasticity and habitat dynamics. The models will be parameterized with empirical data obtained in task #1. The models will be used to make quantitative predictions about transient and equilibrium metapopulation dynamics, and to elucidate critical factors in the risk of population and metapopulation extinction under different environmental conditions (predictions to be used in task #5).

Description of task. The partners have extensive experience about modelling of metapopulation dynamics (simulation models, incidence function models, structured population models, stochastic models, individual-based models). The novel aspect of this project is a forceful effort to unite the skills of a team experienced in field ecology and deterministic metapopulation models (HKI) with the skills of a team experienced in theoretical ecology and experienced in stochastic extinction models (LPZ).

Objectives and approach. The partners involved in this task will integrate their theoretical, laboratory and field experience to provide insights into the evolution of life history traits in metapopulations. Task 3a: To identify empirically the effects of landscape structure on the evolution of migration in central and marginal populations in more and less fragmented landscapes. Task 3b: To model the effects of landscape structure and habitat fragmentation on the evolution of migration rate, and to model the effects of evolving migration rate for metapopulation persistence and changing geographical distributions in response to environmental change. In combination, 3a and 3b represent a highly novel approach to the evolutionary effects of environmental change on species and the composition of biological communities, an area largely ignored in conservation and landscape management.

Description of task. Task 3a: Sample, rear and conduct genetic crosses within and between metapopulations to relate differences in flight morphology and life history characters to (1) measured migration rates (from task #1) and (2) to the position of the metapopulation along the continuums from central to marginal ones, and from much to less fragmented habitat. Task 3b: Develop models examining (1) how landscape fragmentation is likely to affect the evolution of migration rate, (2) how evolutionary changes in migration may affect metapopulation persistence (contributions from task #2), and (3) evolution of migration rate at expanding and contracting species range margins, with implications for changing patterns of European biodiversity in the future (task #5).

Objectives and approach. We will examine how population genetics theory applies to metapopulations and how evolutionary and genetical processes (especially inbreeding) influence the persistance of metapopulations and their components. We will do this by integrating three different approaches: (1) Monitoring genetic variances and fitness within manipulated laboratory populations, (2) collecting data on genetic variation, fitness variation and the persistance of local populations in natural metapopulations and (3) examining how these results fit with the predictions of population genetics theory and simulation models. One major problen in (meta)population biology is lack of data on effective population size, which is critical in determining losses of genetic variation. Initially, we will use captive populations of B. anynana to explore the differences between census number and effective population size (breeding population) under different conditions of inbreeding histories.

Description of task. Captive (meta)populations will be established and run for 5-6 generations to address the unresolved issues associated with inbreeding depression and effective population size in butterfly metapopulations. Some of these populations will be monitored at the levels of individual pedigrees (using established methodology) so that known effective population sizes and inbreeding coefficients of individuals can be linked to effects on performance and fittness. Other populations will be run under different conditions to explore the variation in effective population size with inbreeding history and environmental stress. The result from these experiments will be examined by integrating them with data collected from natural populations and with specific modelling studies.

Objectives and approach. To develop a general framework based on the theory and concepts of metapopulation biology and allowing (1) identification of the maximum level of habitat fragmentation compatible with metapopulation viability and (2) the restoration of extinct metapopulations by reintroductions. Using models developed in task #2 and species studied in task #1, we aim at establishing rules of habitat network design that are practical for conservation. The novel aspect of this project is the collaboration between teams with theoretical/empirical and ecological/genetic expertise in a joint effort to develop a methodology for species conservation applicable for other mobile organisms living in fragmented landscapes.

Description of task. Develop guidelines to assess the effects of metapopulation structure and habitat fragmentation on population persistence and the maintenance of genetic variation within and between metapopulations, drawing on data from tasks #1, 3 and 4, and analyses in task #2. Evaluate empirically the demographic and genetic consequences of population reintroductions, and assess the implications for species recovery programs. Test in the field the predicted optimal design of patch networks (from task #2) for metapopulation restoration. UCL, LDN and HKI have past experience in population genetics. UCL, HKI and LDS have previous experience in large- scale introduction experiments.

HKI - Department of Ecology and Systematics, Division of Population Biology, University of Helsinki; FI

Qualifications and role of the team in the network:

Network coordinator, leader of task #1 (M. cinxia). The metapopulation research group of prof Hanski has made significant contributions to the theoretical and empirical literature on metapopulation dynamics, including the latest text book in this field (ref. #1). High level of activity is demonstrated by two European workshops organized in 1994 (one funded by the European Science Foundation). An international workshop is planned for 1998. This team has collaborated previously with the team from Leeds (LDS, Dr Thomas). In the new text book on metapopulation biology (ref. #1), chapters are contributed by members of the research teams in Leeds (LDS, Dr Thomas) and Montpellier (MNP, Dr Olivieri).

The two main contributions of this team to the network are the modelling of metapopulation dynamics and the field study of one of the four focal butterfly species (Melitaea cinxia). Ref. #2 is a representative example of the pioneering and state-of-the-art modelling work conducted in this team. The large-scale field study of Melitaea cinxia is unique in terms of the spatial scale and detail covered. This system now represents an 'ecological field facility' which facilitates the study of many ecological, genetic and evolutionary questions that would be hard to study with any other system.

Two key publications:

Hanski, I.A. & Gilpin, M.E. (eds.) 1996. Metapopulation Biology: Ecology, Genetics and Evolution. Academic Press, London, 512 pp.

Wahlberg, N., Moilanen, A. & Hanski, I. 1996. Predicting the occurrence of species in fragmented landscapes. Science. 273:1536-1538.

UCL - Unite d'Ecologie et de Biogeographie, Universite de Louvain, Louvain-la-Neuve; BE

Qualifications and role of the team in the network:

Leader of tasks #1 (P. eunomia), 4. UCL has accumulated extensive experience on metapopulation reestablishment through participation in a demographic and genetic study of an established metapopulation of Proclossiana eunomia. The group has additionally wide experience in behavioural ecology. The research team of UCL was created in 1992. Since that time the team has made two significant contributions to metapopulation studies (ref. #1 and 2). The work conducted in the team focuses on the causes and consequences of migration: why and how do individuals move, and what are the consequences of population isolation at different time/space scales.

Two key publications:

Baguette, M., Neve, G. 1994. Adult movements between populations in the specialist butterfly Proclossiana eunomia. Ecological Entomology 19: 1-5.

Neve, G., Barascud, B., Hughes, R.M., Aubert, J., Descimon, H., Lebrun, Ph., Baguette, M. 1995. Dispersal, colonisation power and metapopulation structure in the vulnerable butterfly Proclossiana eunomia (Lepidoptera, Nymphalidae). Journal of Applied Ecology 33 (1):14-22.

LPZ - Department of Ecological Modelling, Centre for Environmental Research, Leipzig; GE

Qualifications and role of the team in the network:

Network coordinator, leader of task #2. The research team of prof Wissel has long-standing experience in modelling stochastic population dynamics (ref. #1) and they have made significant contributions to the mathematical and methodological basis of population modelling. They have determined the key factors in the extinction processes and deduced a way of quantifying the risk of extinction in conservation applications. They have unified different modelling approaches (analytical, simulation and individual based models), and constructed a spatially explicit stochastic simulation model for determining the extinction risk of a locust metapopulation in a dynamic landscape (ref. # 2). In recent stochastic models, key factors for the survival of metapopulations were determined and the linkage to the local within-patch dynamics was clarified. The contribution of the team to the network is the development of stochastic metapopulation models which are butterfly specific, the determination of key factors for the survival of these metapopulations under different environmental condition, and the deduction of conservation strategies.

Two key publications:

Wissel, C., Stephan, T. & Zaschke, S.-H. 1994. Modelling extinction and survival of small populations. In: H. Remmert. Minimal animal populations. Ecol. Studies 106: 67-103, Springer.

Stelter, C., Reich, M., Grimm, V., Wissel, C. 1997. Modelling survival in dynamic landscapes: lessons from a metapopulation of the grasshopper Bryodema tuberculata. J. Anim. Ecol. (in press)

LDS - Department Biology, University of Leeds; UK.

Qualifications and role of the team in the network:

Leader of task #1 (P. argus). Dr Thomas, his research group and collaborators (including HKI, prof Hanski), have been responsible for major empirical contributions to the study of metapopulation dynamics, evolution within metapopulations, butterfly ecology and conservation. Dr Thomas coordinates a rapidly-growing and already the largest university research group working on butterfly population ecology in the UK. Dr Thomas has recently contributed chapters on butterfly metapopulations for several books and review articles on butterfly conservation and metapopulation biology. Dr Thomas is an editor of the international journals Journal of Animal Ecology and Ecological Entomology.

The main contributions of this team to the network are: distributions of, and migration in, butterfly metapopulations; evolution of dispersal; unique field study systems of P. argus with survey data extending back to 1982-83, and prior historical information. The important empirical contributions by this group are already influencing the theoretical development of metapopulation biology. The group makes major contributions to conservation biology, and has shown how landscape change has resulted in rapid evolution in a butterfly. The historical knowledge of P. argus and other butterfly metapopulations gathered by Dr Thomas in the UK provide a unique background to the study of metapopulations of these species and, as with M. cinxia and P. eunomia (HKI and UCL, respectively), can be regarded as "ecological field facilities," will be used for practical work, and to guide development of theoretical models in all tasks of the network.

Two key publications:

Thomas, C.D., Hanski, I. 1996. Butterfly metapopulations. In: Metapopulation dynamics: ecology, genetics and evolution (I.A.Hanski & M.E.Gilpin, eds.). pp 359-386. Academic Press, London.

Thomas, C.D., Singer, M.C., Boughton, D.A. 1996. Catastrophic extinction of population sources in a butterfly metapopulation. American Naturalist 148:957-975.

COR - Department of Plant biology and Ecology, Division of Ecology, University of Cordoba; ES

Qualifications and role of the team in the network:

Leader of task #1 (M. jurtina).The research group has gained a wide expertise in butterfly ecology through intensive empirical work. Combining field observations with carefully designed laboratory and field experiments, the group has made significant contributions to the current knowledge of factors underlying patchy spatial patterns of several butterfly species. The role of both abiotic factors and biotic relationships, including their interactions with plants, ant mutualists, vertebrate herbivores, and parasitoids, have been investigated (Refs. #1-2).

The contributions of this team to the network are detailed ecological knowledge of the Donana's Plebejus argus population. This population occurs in protected area, the Donana National Park. The Donana Biological Field Station is currently implementing the "European Large Installation Programme at the Donana Biological Station" (sept-1994/Aug 1997). Our group is included in this program, and other members of network could also benefit of it. Besides of this, the group has recently accumulated a valuable experience in using a computerized image analyzer system to measure morphometric variables related with wing shape, size and design; and analyzing allometric relations to study resource allocation.

Two key publications:

Jordano, D., Rodriguez, J., Thomas, C.D. and J. Fernandez Haeger 1992. The distribution and density of a lycaenid butterfly in relation to Lasius ants. Oecologia, 91: 438-446.

Rodriguez, J. Fernandez Haeger, J. and D. Jordano. 1994. Spatial heterogeneity in a butterfly-host plant interaction. Journal of Animal Ecology, 63: 31-38.

MNP - Institut des Sciences de l’Evolution, Universite Montpellier 2; FR

Qualifications and role of the team in the network:

Leader of task #3.The research group of prof Olivieri, along with collaborators from other labs in Montpellier and Paris, has made significant contributions to the theory of the evolution of migration rate (Ref#1), sex allocation, and host-pathogen relationships in metapopulations. They have also contributed to empirical studies of dispersal and mating systems in natural plant metapopulations, as well as to metapopulation biology of endemic, endangered plant species, and fine-scale population structure of widespread species, for both molecular markers and quantitative characters.

The team also includes D. McKey (Professor), 4 associated professors and 3 PhD students not involved in the project.

Two key publications:

Olivieri, I., Michalakis, Y.& P.H. Gouyon. 1995. Metapopulation genetics and the evolution of dispersal. American Naturalist 146:202-228.

Olivieri, I. and P.-H. Gouyon, 1996. Evolution of migration rate and other traits: the metapopulation effect. In Hanski, I. and Gilpin, M.E. (eds.), Metapopulation biology : Ecology, Genetics and Evolution. Academic Press.

LDN- Institute of Evolutionary and Ecological Sciences, Leiden University, Leiden; NL

Qualifications and role of the team in the network:

Leader of task #4. The conservation genetics group of prof. Brakefield has used captive populations of the African butterfly Bicyclus anynana to examine the effects of genetic drift (in bottlenecks) and inbreeding on fittness and population persistence. This butterfly is extremely sensitive to inbreeding. The most recent experinets have shown how we can use a highly sensitive pedigree approach in small laboratory populations to explore the details of the relationships between inbreeding, population size and gene flow. We have, however, become increasingly aware that much more understanding is needed of the factors which influence effective population size (relative to census number) in metapopulations. Experiments to investigate the genetical processes within metapopulationswill be established and run in an interfacing manner with the field studies and modelling approach. The work in Leiden is also performed to interact with evolutionary, developmental and physiological studies of both morphological (wing pattern) and life history traits of butterflies.

Two key publications:

Brakefield, P. M., et al. 1996. Development, plasticity and evolution of butterfly eyespot patterns. Nature 384:236-242

Brakefield, P. M., Kesbeke, F. 1997. Genotype x environment interactions for insect growth in constant and fluctuating temperature regimes. Proc. R. Soc. Lond. B. (in press)

The research conducted in this network is structured into five research tasks as detailed in the Work Plan (Section 5). Task #1 forms the core of the network, with close collaboration between five teams to carry out the field work on the four species. Tasks #2 to 5 involve 2 to 3 teams each, with the network coordinator (HKI) being involved in each task.

The following charts specify the teams taking part in different tasks:

The two less experienced teams (UCL, COR) take part in three tasks each, #1, 3a and 5. The post doc researchers in these teams will spend at least two months in at least two other laboratories belonging to the network to receive training in relevant techniques (Section 10). The directors of these teams belong to the Steering Committee of the network and can thereby facilitate the collaboration of their respective teams with the other teams in the network.

As the above chart specifies, each research task involves collaboration among several teams. This involves jointly planned field research and central data collection (task #1), in which a post doc employed for the scientific coordination of the project will assume major responsibility. The modelling task (#2) involves joint work with empirically-oriented researchers and modellers, and the use of data collected in tasks #1 and 3 to parameterize models. The breeding facility in LDN can be used to study inbreeding (task #4) of the species on which field work is carried out in task #1. Finally, the results of all tasks #1 to 4 will be needed for the development of practical conservation procedures. Here the modelling results (task #2) play a pivotal role.

Habitat loss and fragmentation is widely considered to be the greatest threat to biodiversity in Europe and elsewhere. Though the solutions to the problems created by habitat destruction are often straightforward, and depend more on political will than new scientific discoveries, there is also an urgent need to extend our knowledge of the biological responses of species to habitat fragmentation, and a need to train biologists in this area. For instance, we have only a limited understanding of the time delays involved in the ecological, genetic or evolutionary responses to habitat fragmentation.

The relatively young field of metapopulation biology tackles these issues from a fresh perspective. The approach is so novel that most universities in Europe have no adequate training in this field of biology. The first text books are only now appearing. The methodology is developing rapidly. We have identified training need for young European researchers especially in the following areas:

Methods of field work. There is little methodological standardization of field techniques employed in metapopulation biological studies, which makes comparative studies difficult. The core task (#1) of this network will help to disseminate sound methods of field work. Students and young post doc researchers examining the current international literature will find almost as many slightly different field techniques as studies published on spatial population structure. It it essential that we develop comparable methods that can be applied widely with minimum modification.

Molecular biology applied to metapopulation studies. Though there is no lack of general expertise in the use of molecular techniques in population biology in Europe, there is an ever present need to help field ecologists acquire a better understanding of the opportunities presently available via the application of current molecular techniques. The main problem has been that most molecular biologists do not obtain sufficient field experience to allow them to exploit the field potential of the methodology to its limits. Similarly, field ecologists should have greater first hand experience of molecular techniques. Through the exchange programme (Section 10), we will address this issue.

Mathematical modelling. Questions about habitat fragmentation are typically quantitative questions about the responses of populations and species to changes in their environments. Such questions cannot be properly answered without resorting to mathematical models. The basic training of most biologists involves little if any instruction in the construction, analysis and interpretation of mathematical models. Though we cannot undertake to train new fully fledged modellers in this network, we can help the young researchers to become familiar with the purpose and use of models in this area of biology. The key issue is that modellers must have some experience of field conditions, and field ecologists must get some exposure to modelling of field systems (see Section 10, exchanges).

In this network, young post doctoral researchers will develop an ability to work in groups and, progressively, they will be given management responsibilities and will play a significant role in project development. On completion of the project, transferable and specific skills will enable the post docs to conduct independent or collaborative research to understand and predict the direct and indirect effects of human activities on biodiversity, or to enter environmental agencies with a view to the management of the impacts of environmental change on biodiversity.

TRAINING PROGRAMME

Procedure to hire visiting researchers. The vacancies will be advertised in mailing lists directed to population biologists and in the annual meetings of European population and evolutionary biologists.

Each young category 30 (post doc) researcher employed in this program will receive a contract for 1-3 years in one of the seven research teams in the network. Additionally, over the period of the contract, each post doc will spend at least two months in at least two other laboratories to receive training in the following techniques:

methods of field work	LDS, HKI, UCL, COR
laboratory experiments on captive populations	LDN, HKI, LDS, UCL
molecular biology	HKI, LDN, MNP
mathematical modelling	HKI, LPZ, MNP
population genetics	MNP
land use and conservation issues	UCL, LDS, HKI

Other post doctoral researchers in the network (i.e. those working in the research projects but not employed by this program) will be given the opportunity to visit one other laboratory in the network for one month per year of their contract. This mobility is essential to the success of the research collaboration as well as to the training of young post doctoral staff. During periods of intensive field work, post doctoral staff will work together at particular locations in association with the local task leader, post graduate students, and undergraduate assistants. During winter months, post docs will visit other laboratories and work closely with staff involved in genetic and mathematical analyses of material and data gathered in the field season. This will provide invaluable assistance to the staff involved in these studies, and will provide excellent training for visiting post docs. Local young post docs and post graduate students will benefit of the visits by other post docs and task leaders to their laboratories. Two meetings will be organised (years 2 and 3, Section 8.2) in which all young post docs and post grads in the network give presentations and discuss progress and conclusions. All post docs will be encouraged to participate in staff development programs in the institutions where they are employed, annual career development appraisals will be carried out, and training progress will be subject to annual reports.

The research teams have the following facilities that provide unique training opportunities in the form of state-of-the-art research:

Ecological field facilities. Well-studied and in many ways unique metapopulations of Melitaea cinxia in Finland, Plebejus argus in Britain, and Proclossiana eunomia in Belgium represent "ecological field facilities" for the study of ecological, genetic and evolutionary questions, enabling the research projects and training programs to start immediately.

Laboratory facilities. The butterfly laboratory in Leiden has purpose-built facilities for large-scale rearing and experimentation on butterflies, with up-to-date rearing chambers. Leeds ecologists occupy a new purpose-built building with extensive ecology and molecular genetic laboratories, a wide range of controlled-temperature rooms, temperate greenhouses and image analysis facilities. The Helsinki team has a modern molecular genetics laboratory and is developing a butterfly rearing facility at a field station close to Helsinki.

Training capacity in mathematical modelling. The Leipzig team has extensive experience in training students in metapopulation dynamics, including lectures and practical training as well as practice-oriented courses for conservation biology.