Ecological modeling

from Wikipedia, the free encyclopedia

The Ecological modeling is a part of scientific ecology . It can be used to analyze and simulate developments and scenarios of individuals, in habitats or on a global level. B. to be able to predict future developments. Ecological modeling developed from theoretical ecology , paired with increasing computing power and an ever increasing amount of data from natural systems.

In the international scientific landscape, the trend towards increasingly complex models for answering global questions from the environmental sciences, macroecology, climatology and other natural sciences can be observed. Sometimes the usefulness, validity and necessity of modeling approaches are not always clear.

history

Theoretical approaches to the development of models of natural or near-natural habitats have existed in ecology since it was recognized as a scientific discipline.

As early as the 1960s, the Department of Forestry and Rural Development in Ottawa developed a forerunner of the geographic information systems (GIS) used today . Roger Tomlinson developed the GIS "Canada Geographic Information System" (CGIS) for the state development authority, with which data from the "Canada Land Inventory" could be analyzed and processed.

The use of habitat models was institutionalized for the first time in planning by the US Fish & Wildlife Service in 1981 with the development of habitat suitability index models (HSImodels). It was part of the so-called habitat evaluation procedure HSI . Initially, the HSI models were based more on expert knowledge and general statements on habitat preferences of the respective species (e.g. Schroeder 1982; Conway and Martin 1993; Reading et al. 1996).

In the 1980s, many scientists were able to use computers to process large amounts of discrete data. At the same time, more and more remote sensing data from satellites and aerial photographs became available for science after the end of the Cold War . In addition, it became possible for normal users to obtain precise position data using the Global Positioning System .

The combination of geo-referenced remote sensing data, field data with relatively exact position information and computer-aided spatial analysis opened up great opportunities for ecologists to predict local to global trends in habitats from the mid-1990s. Modeling has become an important tool in ecosystem research . The relatively young branch of ecology, macroecology , is based on the modeling of ecological processes and has strong overlaps with geographically shaped biogeography .

approaches

In ecological modeling, a distinction is made between two main types of ecological concepts, the application of which depends on the question: analytical models and simulations . Analytical models are often mathematically complex and are best applied to relatively large simple (often linear) systems. Simulation models are used in a broader field and are considered more ecologically realistic. They are based on a mathematically more sophisticated foundation. The models are created using different programming languages. Accessible besides Java and Delphi applications of R .

Applications

Habitat models are a common tool in applied ecology. They are used in marine, marine and terrestrial ecology. Both synecological approaches and the environmental conditions for individual species can be modeled.

Habitat models are used by authorities to forecast the environmental impact of interventions. They are also used for compensation and maintenance measures in environmental planning and can improve the effects of nature conservation management. In nature conservation, occurrence forecasts can be made. The models can be used to carry out habitat connectivity analyzes, with which statements can be made about the connection between habitats of less mobile species.

Predominantly analytical approaches

Habitat models

Habitat models usually aim at two main questions:

  • Which biotopes are suitable as habitats?
  • What are the features of the biotopes or what habitat requirements of the species are?

The modeling process is used to analyze data sets . Special modeling software such as Maxent in combination with geographic information systems (ArcGIS, DIVA GIS, etc.) are used in the actual habitat modeling. Remote sensing data is obtained from the Landsat program , for example . Digital terrain models using SRTM data are usually also created in GIS. Special packages from R , sometimes also SPSS, are used to process ecological data .

Statistics and filter cascade Generally, generalized linear models are used for macroecological analyzes. The basic idea of ​​most habitat models is to predict settlement probability values ​​using a set of habitat gradients and presence / absence data.

Filter cascades are often used for analysis.

  • Level 1: Resources
  • Level 2: biotic intraspecific & space
  • Level 3: biotics interspecific

Niche models

Niche models are used to answer questions about its distribution (biogeography) based on the ecological niche of a population or a species, and on this basis also about population trends and their possible endangerment. First, the fundamental niche is modeled on the basis of the known physiological and ecological demands of the species, provided that these are known from independent data (e.g. laboratory tests, experience from the culture or keeping of the species, autecological studies). Since there are usually no or far too few data, the species' claims are usually only derived from the correlation of their known occurrences with environmental factors, e.g. B. Their temperature claims from the northernmost or southernmost occurrences, the temperature profile of which can be read from climate atlases. This data is then validated against the known data on actual diffusion. By comparing the actual distribution with that simulated according to the modeled niche, the plausibility of the model can be checked (of course, the distribution data used to calibrate the model must not be used a second time! If sufficient data is available, part of it is usually used as a Test data used). With the aid of the calibrated and tested model, it is now possible to assess the effects of simulated changes in the basic data, e.g. B. to the climate, to predict the distribution and abundance of the species.

Mainly simulation approaches

Agent-based models

In contrast to other types of modeling (for example System Dynamics ), in agent-based modeling, many small units (agents) have decision-making or action options, such as B. is the case with animals in a real environment. In these models, the system behavior results from the behavior of the individual agents and is not specified at the system level. If there are effects at the system level (e.g. a biome) that cannot be directly derived from the decision-making algorithms of the individuals, this is called emergence . In addition, system behavior that is separate from the individual decisions can be implemented.

Two decisive aspects of agent-based modeling are the ability to explicitly depict heterogeneous behavior and dependencies on other individuals.

This type of modeling is mainly used when the focus of a question is not the stability of an equilibrium or the assumption that a process will return to equilibrium, but rather the question of how a system can adapt to changed framework conditions ( robustness or resilience ). Practical applications can be found e.g. B. in the question of which resilience behavior coral reefs exhibit and to what extent they can compensate for negative environmental influences. Agent-based models take into account the fact that complex problems require the micro-level, i.e. the decisions of individuals, their heterogeneity and their interactions, to be examined directly.

Limitations of modeling approaches

Since the advent of practical statistical applications (R etc.) and geographical information systems, more and more models are being used in modern ecological research. Existing databases are often used, for example to model climate, vegetation and other factors. Critics point out that not everything that can be modeled technically has added value in ecological research. Particularly in the case of predominantly analytical approaches, factors are often taken into account for which data are available, but other gradients with a greater influence are sometimes ignored. Mobile species (e.g. birds) are difficult to model.

literature

Reference books

  • Yup, Fred; Reuter, Hauke; Breckling, Broder (Eds.) (2011) Modeling Complex Ecological Dynamics. An Introduction into Ecological Modeling for Students, Teachers & Scientists. Springer ISBN 9783642050282
  • Horning et al .: Remote Sensing for Ecology and Conservation. A Handbook of Techniques. Oxford University Press, Oxford u. a. 2010, ISBN 978-0-19-921995-7 .
  • Lang, Blaschke: Landscape analysis with GIS . Ulmer
  • Marie-Josee Fortin, Mark Dale (2005): Spatial Analysis. A Guide for Ecologists. Cambridge University Press . ISBN 9780521009737

Specialist journals

Technical article

Individual evidence

  1. ^ Sven Erik Jørgensen: Handbook of Environmental and Ecological Modeling. ISBN 1-56670-202-X , limited preview in Google Book Search
  2. ↑ Boris Schröder: Habitat models for modern nature conservation management. ( Memento from April 19, 2008 in the Internet Archive ) In: Albrecht Gnauck (Ed.): Theory and Modeling of Ecosystems - Workshop Kölpinsee 2000. Shaker, Aachen 2002, pp. 201–224.
  3. ↑ 2. Basics. In: Boris Schröder: Habitat models for modern nature conservation management. ( Memento from April 19, 2008 in the Internet Archive ) In: Albrecht Gnauck (Ed.): Theory and Modeling of Ecosystems - Workshop Kölpinsee 2000. Shaker, Aachen 2002, ISBN 3-8322-1316-3 , p. 202.
  4. ↑ Page no longer available , search in web archives: brandenburg.geoecology.uni-potsdam.de page 4@1@ 2Template: Toter Link / brandenburg.geoecology.uni-potsdam.de