Simplifying complexity: applications of stochastic dynamic methodology (StDM) in terrestrial ecology
Ficheiros
Data
2008
Autores
Santos, Mário Gabriel Santiago dos
Título da revista
ISSN da revista
Título do Volume
Editora
Resumo
A utilização de modelos ecológicos tem, nas últimas três décadas, conhecido um crescimento exponencial. Ao mesmo tempo, os avanços no hardware e no software têm vindo progressivamente a entrar no nosso quotidiano, tornando os programas informáticos cada vez mais fáceis de utilizar. Sabemos também que o desenvolvimento de modelos ecológicos necessita de conhecimentos essenciais, nomeadamente acerca do funcionamento dos ecossistemas e principais problemas ambientais. Quando os modelos são desenvolvidos e testados, estes deverão ser aplicados considerando o seu âmbito e princípios fundamentais. Os modelos poderão então simular condições que são difíceis ou impossíveis de compreender de outro modo. No entanto, a modelação ecológica apresenta ainda muitas dificuldades, nomeadamente devido ao facto do conhecimento acerca do funcionamento dos ecossistemas ser ainda deficiente. Numa perspectiva reducionista, a estimativa dos valores dos parâmetros é considerada o elemento mais problemático da modelação. Isto resulta da evidência que a caracterização completa de um ecossistema é provavelmente irrealizável. Considerando que muitos dos aspectos fenomenológicos dos ecossistemas são holísticos, a metodologia apresentada pretende de uma maneira mecanicista captar as propriedades globais dos sistemas, tendo como base métodos estatísticos de estimativa dos parâmetros. A metodologia estocástico-dinâmica (StDM) é um processo sequencial de modelação que foi desenvolvido com o intuito de prever o estado ecológico de ecossistemas em alteração. Este procedimento tem com enfoque interacções entre componentes chave conceptualmente isolados, nomeadamente interacções entre componentes biológicos e variáveis ambientais, que apresentem relevância ecológica e holística, reduzindo assim os preconceitos adicionados à estrutura e funcionamento dos modelos. A metodologia proposta é expedita e facialmente aplicável a diversos tipos de ecossistemas afectados por gradientes de mudança. O contexto de dos modelos StDM é a aplicação ou uso de modelos de gestão com o intuito de monitorizar o estado dos ecossistemas. Assim o StDM é compatível com a maioria de actividades da ecologia, como sejam a procura de padrões, a capacidade de explicar nos ecossistemas estados passados, presentes e futuros. Quando aplicado em contextos de gestão ambiental, as aplicações de StDM poderão ser valorizadas em termos de custos, rapidez e precisão. Em contextos relacionados com educação e comunicação, os resultados dos StDM aplicados à monitorização, ciência e actividades de gestão são intuitivos e por isso facilmente comunicáveis a não especialistas. (desde estudantes a gestores de recursos, políticos e legisladores). A primeira parte desta tese descreve a teoria e procedimentos base do StDM. A segunda parte foi escrita para mostrar com detalhe várias aplicações da metodologia organizadas de acordo com um maior realismo (sem no entanto sacrificar a simplicidade): aplicações de StDM na problemática das políticas agro-ambientais; aplicações de StDM com o intuito de melhorar a qualidade de censos com o método da contagens pontuais; aplicações de StDM para prever os efeitos de infra-estruturas humanas sobre a dinâmica de populações animais; aplicações de StDM em paisagens agrícolas em mudança; aplicações da StDM para prever os impactos dos parques eólicos na fauna terrestre. O último capítulo integra os resultados obtidos, discutindo alguns dos aspectos metodológicos. Em síntese o maior objectivo deste trabalho é aprofundar a discussão em torno de metodologias holísticas a partir das quais se poderão delinear estratégias para recuperar o funcionamento dos ecossistemas e comunidades biológicas afectadas por perturbações de origem antropogénica.
The comprehensive use of ecological models has shown a constant and remarkable growth in the last three decades. At the same time, hardware and software advances have progressively entered in our quotidian, making computer-modelling frameworks easier to use. We learnt that development of ecological models requires a consistent knowledge to the functioning of ecosystems and focal environmental problems. When properly developed and tested, they must be applied with insight and with regard to their underlying assumptions. These requirements could result in models capable of simulating conditions that are difficult or impossible to understand otherwise. Nevertheless, we are still facing serious problems in ecological modelling, namely because basic deficiencies exist in ecosystem science as we really do not know how ecosystems work. In a reductionistic analytical perspective, the parameter estimation is often the weakest point in modelling. This results from the evidence that the characterization of an ecosystem cannot be complete (at most it can only be very partial). Since many of the ecosystem phenomenological aspects are holistic, whole-system properties, the main proposal of this methodology is a mechanistic understanding of the holistic ecological processes, based on a statistical parameter estimation method. The stochastic-dynamic methodology (StDM) is a sequential modelling process developed in order to predict the ecological status of changed ecosystems. These procedures are focusing on the interactions between conceptually isolated key-components. Our approach includes any kind of interactions between biological components and environmental variables, with holistic and ecological relevance, and reduces the number of pre-conceptions added to the modelling procedures. The methodology proposed is expeditious and easily applicable to several types of ecosystems affected by gradients of change. The context of StDM refers to the application or final use of management models in the scope of programs to monitor status of the ecosystems. In this context, the StDM is compatible with most activities undertaken by conventional ecological science, i.e., pattern seeking, the ability to explain past and present states, and the ability to predict future ecosystem states. When applied to contexts relating to environmental management, the applications of StDM are valued in terms of cost and speed of reliable ecological assessment results. In contexts relating to education and communication, the results of StDM applied to monitoring, science, or management activities are intuitive and can be easily communicated to non-experts (ranging from students to resource users to senior policy makers). The first part of this thesis describes the theory and procedures of the StDM. The second part was written to provide with a detailed description and a consistent discussion of several StDM ecological applications organized with increasing realism without sacrificing parsimony: StDM applications in the scope of the agri-environmental measures problematic; StDM applications for improving the quality of point count censuses; StDM applications for the prediction of the infrastructural effects on the dynamics of animal communities; StDM applications in changing traditional agricultural patterns and landscape; StDM applications for predicting the impacts of wind energy facilities in faunal communities. In the last part the chapters’ results are integrated and several aspects of the methodology are discussed. In summary the ultimate objective of this work is to encourage the discussion about holistic methodologies from which management strategies can be designed to restore ecosystems functions and biological communities that have been damaged by anthropogenic disturbances. We hope that the present work constitutes a step forward in this direction.
The comprehensive use of ecological models has shown a constant and remarkable growth in the last three decades. At the same time, hardware and software advances have progressively entered in our quotidian, making computer-modelling frameworks easier to use. We learnt that development of ecological models requires a consistent knowledge to the functioning of ecosystems and focal environmental problems. When properly developed and tested, they must be applied with insight and with regard to their underlying assumptions. These requirements could result in models capable of simulating conditions that are difficult or impossible to understand otherwise. Nevertheless, we are still facing serious problems in ecological modelling, namely because basic deficiencies exist in ecosystem science as we really do not know how ecosystems work. In a reductionistic analytical perspective, the parameter estimation is often the weakest point in modelling. This results from the evidence that the characterization of an ecosystem cannot be complete (at most it can only be very partial). Since many of the ecosystem phenomenological aspects are holistic, whole-system properties, the main proposal of this methodology is a mechanistic understanding of the holistic ecological processes, based on a statistical parameter estimation method. The stochastic-dynamic methodology (StDM) is a sequential modelling process developed in order to predict the ecological status of changed ecosystems. These procedures are focusing on the interactions between conceptually isolated key-components. Our approach includes any kind of interactions between biological components and environmental variables, with holistic and ecological relevance, and reduces the number of pre-conceptions added to the modelling procedures. The methodology proposed is expeditious and easily applicable to several types of ecosystems affected by gradients of change. The context of StDM refers to the application or final use of management models in the scope of programs to monitor status of the ecosystems. In this context, the StDM is compatible with most activities undertaken by conventional ecological science, i.e., pattern seeking, the ability to explain past and present states, and the ability to predict future ecosystem states. When applied to contexts relating to environmental management, the applications of StDM are valued in terms of cost and speed of reliable ecological assessment results. In contexts relating to education and communication, the results of StDM applied to monitoring, science, or management activities are intuitive and can be easily communicated to non-experts (ranging from students to resource users to senior policy makers). The first part of this thesis describes the theory and procedures of the StDM. The second part was written to provide with a detailed description and a consistent discussion of several StDM ecological applications organized with increasing realism without sacrificing parsimony: StDM applications in the scope of the agri-environmental measures problematic; StDM applications for improving the quality of point count censuses; StDM applications for the prediction of the infrastructural effects on the dynamics of animal communities; StDM applications in changing traditional agricultural patterns and landscape; StDM applications for predicting the impacts of wind energy facilities in faunal communities. In the last part the chapters’ results are integrated and several aspects of the methodology are discussed. In summary the ultimate objective of this work is to encourage the discussion about holistic methodologies from which management strategies can be designed to restore ecosystems functions and biological communities that have been damaged by anthropogenic disturbances. We hope that the present work constitutes a step forward in this direction.
Descrição
Tese de Doutoramento em Ciências do Ambiente
Palavras-chave
Ecossistemas , Sustentabilidade , Ecologia , StDM