Ecosystem model

Ecological modeling is a part of scientific ecology. With her ​​can developments and scenarios of individuals, are analyzed and simulated in habitats or at the global level in order to predict the effects of eg future developments. The Ecological Modelling developed from theoretical ecology, coupled with stronger computing power and an ever-increasing amount of data available from natural systems.

In the international scientific landscape of the trend towards ever more complex models for responding to global issues from the environmental sciences, the macro ecology, climatology and other sciences can be observed. Some of the meaningfulness, Valididät and need for modeling approaches is not always clear.

• 5.1 Agent-based models

• 7.1 Reference Books
• 7.2 journals
• 7.3 Articles

History

Theoretical approaches to the development of models of natural or semi-natural habitats there are in ecology, since it is recognized as a scientific discipline.

Already in the 1960s, a forerunner of geographic information systems in use today (GIS ) was developed in Ottawa by the Department of Forestry and Rural Development. Roger Tomlinson developed for the Regional Development Authority, the GIS "Canada Geographic Information System" ( CGIS ), with which data could be the "Canada land Inventory" analyzed and processed.

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

In the 1980s opened up the possibility for many scientists, computer-aided process large amounts of discrete data. In parallel, an increasing number of remote sensing data from satellites and aerial photography for science became available after the end of the Cold War. In addition, it was possible for ordinary users, accurate position data using the Global Positioning System to win.

The combination of geo-referenced remote sensing data, field data, with relatively accurate location information and computer-based spatial analysis opened ecologists from the mid- 1990s, great ways to predict local to global trends in habitats. Constructing models have become an important tool in ecosystem research. On the modeling of ecological processes based the relatively young branch of ecology Macroecology, which has strong overlaps between geographically dominated biogeography.

Approaches

In the ecological modeling two common main types are distinguished by ecological concepts, their application depends on the question: analytical models and simulations. Analytical models are often mathematically complex and are best on relatively large simple ( often linear ) systems apply. Simulation models are used in a wider field and are considered ecologically realistic. They are based on a mathematically sophisticated basis. The models are created using different programming languages. Consistently, in addition to Java and Embarcadero Delphi, applications of R.

Applications

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

Habitat models are used by authorities to predict the environmental effects of interventions. Even with compensation and care measures in environmental planning, they are used and can improve the effects of conservation management. In the nature reserves forecasts can be made. On the basis of models habitat connectivity analysis may be performed, which indication of the compound of habitats less mobile species can be made.

Mostly Analytical Approaches

Habitat models

Habitat models are aimed mostly depend on two main questions:

• What habitats are suitable as habitats?
• Because the equipment used and the habitats which habitat requirements of the species is that?

For the analysis of data sets is moved according to the method of modeling. Special modeling software, such as Maxent in combination with Geographic Information Systems ( ArcGIS, DIVA GIS uaa ) come in the actual habitat modeling used. Remote sensing data are obtained for example from the Landsat program. Digital terrain models using SRTM data are usually also created in GIS. For processing biological data from special packages of R, partly SPSS is mostly used.

Statistics and filter cascade Most are used generalized linear models for macro- ecological analyzes. The basic idea of most habitat models is the probability values ​​of the settlement on the basis of a set of Habiatgradienten and " presence / absence - data " to predict.

Frequently filter cascade can be used for analysis.

• Level 1: Resources
• Level 2: Biotic intraspecific & Space
• Level 3: Biotic interspecific

Niche models

Niche models are used, and then answered on the basis of the ecological niche of a population or a species issues affecting their distribution ( biogeography ), on this basis, to population trends and their potential hazard. First, the fundamental niche is modeled based on the known physiological and ecological requirements of the species, provided that (eg laboratory testing, experience of the culture or attitude of the kind, autecological studies) are of independent data known. Because usually there are no or too few data, the claims of the type such as their temperature requirements of the northernmost or southernmost occurrences in which the temperature profile can be read from climate atlases are usually derived only from the correlation of their known occurrences with environmental factors. This data is then validated against the known data on the actual distribution. By comparing the actual distribution with the simulated according to the modeled niche the plausibility of the model can be checked (of course, may not here the distribution data used for calibration of the model can be used a second time! Loungers enough data before, is usually a part of it as test data is used). Based on the calibrated and tested model, it is now possible to predict the effects of simulated changes in the basic data, such as the climate on the distribution and abundance of the species.

Mostly simulation approaches

Agent-based models

In contrast to other types of modeling (for example, System Dynamics ) have in the agent-based modeling many small units (agents) decision-making or action possibilities, as for example in animals in a real environment is the case. In these models the system behavior results from the behavior of individual agents and is not imposed at the system level. If this leads to effects on the system level (eg, a biome ) that are not derivable directly from the decision algorithms of individuals, it is called emergence. In addition, a separate from the individual decisions system behavior can be implemented.

Two key aspects of the agent-based modeling are the ways to explicitly represent heterogeneous behavior and dependencies on other individuals.

This type of modeling is especially used when the focus of a question is not the stability of an equilibrium and the assumption that a process returns to an equilibrium, but the question of how a system can adapt to changing conditions (robustness or Resilenz ). Practical example Anwendnungen find themselves the question which Resilenzvehalten have coral reefs and the degree to which they can compensate for negative environmental influences. In agent-based models of knowledge is taken into account that complex problems require, the micro - level, ie the decisions of individuals, their heterogeneity, and their interactions, to investigate directly.

Limitations of modeling approaches

Since the advent of practical statistical applications (R, etc.) and geographic information systems is modeled more and more in modern ecological research. Frequently use is made of existing databases to model, for example, climate, vegetation, and other factors. Critics point out that not everything what could technically model, have an added value in ecological research. Especially with predominantly analytical approaches factors would often considered are provided for the data, but others remain gradients with greater influence partly unnoticed. Difficult to model, mobile species (e.g., birds).