of 40
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.

Modelling the ecosystem as a hierarchical system

Category:

School Work

Publish on:

Views: 2 | Pages: 40

Extension: PDF | Download: 0

Share
Description
Rencontres de la Chaire Modélisation mathématiques et biodiversité 15/9/2015, Paris Modelling the ecosystem as a hierarchical system Jacques Gignoux1 Shayne Flint2 Ian Davies2 Guillaume Chérel3 Eric Lateltin1
Transcript
Rencontres de la Chaire Modélisation mathématiques et biodiversité 15/9/2015, Paris Modelling the ecosystem as a hierarchical system Jacques Gignoux1 Shayne Flint2 Ian Davies2 Guillaume Chérel3 Eric Lateltin Ecology is an «integrative» or «synthetic» science pedology [bio]geography systematics geology genetics hydrology demography ecology [bio][geo]chemistry physiology cell biology [micro]meteorology climatology graph theory thermodynamics game theory system dynamics As a result of this «integration», ecological modelling is far from united - PDE type models - cellular automata - individual-based models / multi-agent systems Consequences: - incompatible time and space representations - problems with model comparisons - and «the scaling problem» : changing from one type of model to another for the same real-world system is costly when possible. Can ecological concepts help us designing a broadly applicable framework for ecological modelling? Which concepts used in ecology are truely 'ecological', i.e. were not borrowed from another scientific field?? Which concepts used in ecology are truely 'ecological', i.e. were not borrowed from another scientific field? The ecosystem... What can we build upon the ecosystem definition? Problem: 'conceptual drift' The ecosystem was defined in 1935 and has undergone many transformations since then. Jax K., Can We Define Ecosystems? On the Confusion Between Definition and Description of Ecological Concepts. Acta Biotheoretica, 55: «A major problem, which impedes the solution to these questions, is a common confusion between definitions and additional descriptions of concepts» example: O'Neill R.V., Is it time to bury the ecosystem concept? (with full military honors, of course). Ecology, 82: There is no proof of ecosystem showing stability, resilience, etc. These properties were never part of the initial definition. Back to the roots! The ecosystem Tansley (1935) The use and abuse of vegetational concepts and terms. Ecology 16: : [...] But the more fundamental conception is, at is seems to me, the whole system (in the sense of physics), including not only the organismcomplex, but also the whole complex of physical factors forming what we call the environment of the biome the habitat factors in the widest sense. Though the organisms may claim our primary interest, when we are trying to think fundamentally we cannot separate them from their special environment, with which they form one physical system. 1. Ecosystem = biological system + physical system physical environment organisms organism: plant organism: predator physical environment organism: herbivore Organism Organism: slime mold physical environm Biological or physical system: a creative ambiguity ecosystem physical system 'organismic complex' ecosystem physical system biological system the ecosystem has a physical and a biological part the ecosystem is a physical and a biological system 'the whole system (in the sense of physics), including not only the organism-complex, but also the whole complex of physical factors' 'when we are trying to think fundamentally we cannot separate them from their special environment, with which they form one physical system.' ecosystem = a mixture of organisms and 'physical factors' ecosystem = a system studied with the methods of physics and biology physics-dominated ecosystem organism-dominated ecosystem The dual nature of ecosystems Ecosystems as biological systems : birth, death, reproduction, demography, discrete states, decision, stochasticity Ecosystems as physical systems : matter and energy fluxes, thermodynamics, continuous states, determinism, conservation laws microbe population dynamics carbon and water fluxes 1 system, 2 representations representation = description of a system using a particular method [...] But the more fundamental conception is, at is seems to me, the whole system (in the sense of physics), including not only the organismcomplex, but also the whole complex of physical factors forming what we call the environment of the biome the habitat factors in the widest sense. Though the organisms may claim our primary interest, when we are trying to think fundamentally we cannot separate them from their special environment, with which they form one physical system. There is no idea of space, time, or scale in Tansley's definition 2. The ecosystem is a scale-independent concept a large ecosystem a small ecosystem Further down in Tansley's paper: '... a system we isolate for the purpose of the study'. 'The isolation is partly artificial, but is the only possible way in which we can proceed' 'The mental isolates we make are by no means all coincident with physical systems, though many of them are, and the ecosystem among them.' 3. The ecosystem is an arbitray construct, a representation of the real world cf. Carnot (1824) : system = the part of the world under consideration The holocoen [(Friederichs 1927) = 'a naturally delimited part of the biosphere'] never achieved the success of the ecosystem concept. Jax 2006: criteria for a good definition: clarity, consistency, applicability Good news: Everything is an ecosystem! 1. ecosystem = physical + biological system 2. scale independent 3. arbitrary construct (almost) anything can be studied as an ecosystem The ecosystem is the basic building block of ecology. Ecology consists in viewing everything as ecosystems. The ecosystem as a self-similar object If anything can be treated as an ecosystem, any part of an ecosystem is still an ecosystem: Ecosystems can be nested 0..* ecosystem physical system 1 biological system 2 problems The boundary problem: In practice, how do we delineate ecosystems in the field? If I want to work on «the forest», where do I sample and take measurements? The abstraction problem: Can the sub-systems of an ecosystem be represented at the same abstraction level? Is the ecosystem a consistent representation of the real world? The boundary problem: an easy case outside world ecosystem 1 ecosystem 2 A not so easy case Should the lake be isolated from its water catchment? The decision to consider the lake or the lake within its catchment is a choice. It is usually motivated by 'the purpose of the study', although scientific tradition also interfers. What is the link with 'the purpose of the study'? Examples: - considering the water catchment around the lake means we are dealing with water runoff (a particular ecological process). - if we were interested by the full trophic network of the lake, we might consider migratory birds as top predators. We would then extend the spatial domain differently (eg where is the lake on a migration route). That's another ecological process. The spatial domain we consider depends on the ecological processes we want to consider in the study (which depend on the purpose of the study). Consequence: since spatial domain of interest depends on ecological processes, there may be as many different spatial domains as processes considered in the ecosystem. 1 : lake Vertical migration Plankton demography Trophic cascades 2 : lake + catchment Organic inputs Eutrophication Hydrology 3 : migratory route Trophic web Waterbird demography Nutrient exports 1 ecosystem, 3 spatial domains / representations The ecosystem and the landscape A landscape is an 'ecosystem' within an area (Lepczyk et al. 2008) landscape ecosystem physical system area biological system space is central to the definition of the landscape, while it is absent from that of the ecosystem Lepczyk C.A., C.J. Lortie & L.J. Anderson, An ontology for landscapes. Ecological Complexity, 5: The ecosystem concept and the ecosystem object The practical problem of the field ecologist : There is no obligate need to refer to space or scale when thinking or modelling the ecosystem the ecosystem used in this case is just a concept There is a need to locate a place where to sample when experimenting on an ecosystem in the field the ecosystem in this case is an object representative of the ecosystem concept. This is known as the class-instance relation in object progamming. Here, to get an instance 'my_ecosystem_for_experimentation' of the class 'Ecosystem', we need an operation on space. But once we have the instance, we might not need to refer to space anymore. Ecosystem (class) constructor = operation on space ecosystem (instance) Application for modelling 1 associate processes to spatial representations of an ecological object (based on computation optimisation) 2 manage interaction between processes through spatial overlaying of spatial representation of an ecological object ecological processes local competition, population dynamics variability of the physical environment radiation absorption, photosynthesis seed dispersal, competition for nutrients ANR Project : the 3Worlds modelling platform for ecosystem simulation carbon allocation, morphologic plasticity spatial representation The abstraction problem 'Abstraction is used to reduce and factor out details so that one can focus on a few concepts at a time' [wikipedia] Do we need to know every organism in an ecosystem? What is the correct level of detail for an ecosystem? - parsimony principle. Can we describe an ecosystem to the same level of, e.g., biological organization (e.g. population or individual)? What is the consequence of inconsistency in the level of abstraction of ecosystem components? At first sight, it makes sense to decribe an ecosystem at the same level of abstraction for all its components (as a complex system) But this seems impossible: whatever ecosystem is studied, some of its parts will always be more detailed than others. Examples: - population level: too many & unknown species! - individual level: sizes of individual organisms span 7-8 orders of magnitude. The common practice is to focus on dominant species, features, traits. Which may affect resilience and other traits, eg response to climate change This is a BIG issue! Importance of the level of abstraction: an example from trophic networks Hulot et al. 2000; Lazzaro et al species *** 20 ** ** * 10 0 genera functional groups *** ** 30 6 species **** **** **** 4 2 lumped by species genera functional groups lumped by 80 ** ** * genera functional groups 0 species genera functional groups %B - Percent of basal species **** gizzard shad (filter feeder) %Bi - Percent of inedible basal species **** %I - Percent of intermediate species 12 0 S - Overall species richness ** 16 H - Herbivore species richness %T - Percent of top-consumers bluegill (visual feeder) 60 species genera functional groups **** Properties of trophic network as a function of the lumping of trophic species (=abstraction level of the network) **** **** species genera functional groups lumped by System-level (emergent) properties depend on the level of abstraction The abstraction problem: solutions? 1 There is no fully consistent representation of an ecosystem 2 The level of abstraction impacts the system-level properties Possible fixes: - simulation platforms that allow to play with the level of abstraction (using eg the self-similarity of ecosystems) - base the high-level representation of emergent properties on simulated emergence at lower levels e.g. Boulain N, Simioni G, and Gignoux J. (2007). Changing scale in ecological modelling: a bottom up approach with an individual based vegetation model. Ecological Modelling, 203: Is the ecosystem a complex system? A complex system is a system made of interacting parts, which displays emergence due to interaction between parts. Problem: many definitions of emergence, with no agreement among them Emergence as novelty or limit to knowledge (Chérel 2013) Chérel, G Détection et abstraction de l'émergence dans des simulations de systèmes complexes : application aux écosystèmes de savane. Thèse de doctorat, UPMC/ UNA. Dessalles, J.-L., J.-P. Müller, and D. Phan (2007) Assad, A. and N. H. Packard (1992) Ronald, E., M. Sipper, and M. Capcarrère (1999) Forrest, S. (1990) Bedau, M. A. (2003) Kim, J. (1999) Bonabeau, E. and J.-L. Dessalles (1997) Bedau, M. A. (2008)Searle, J. (1992) Müller, J.-P. (2003) Zwirn, H, and Delahaye, JP (2013) The 'hallmarks of emergence' (Bedau 1997) : (1) Emergent phenomena are somehow constituted by, and generated from, underlying processes; (2) Emergent phenomena are somehow autonomous from underlying processes. Bedau, M.A., Weak emergence. In: Malden, M.A., ed., Philosophical perspectives: Mind, causation, and world, pp Wiley-Blackwell Is anything useful derivable from this mess? Only one common point : emergence arises in systems with a «microscopic» description and a «macroscopic» description. This defines a hierarchical system. What is the best representation of a hierarchical system? a dynamic graph : system +descriptors component +descriptors relation +descriptors The ecosystem as a hierarchical system The ecosystem is self-similar = hierarchical Components = objects relevant to the purpose of the study ; must include biological components and an environment (possibly the whole graph) Relations = rules for interactions between components 4 types of «emergence» in a hierarchical system: - 'naive' : arises from neglecting/ignoring interactions - 'discovery' : arises from missing descriptors - 'weak' : due to computational irreducibility between micro- and macro-state - '?': due to causal loops (feedbacks) within the system... conceptual work in progress Simulation as an integration tool the 3Worlds platform ANR-CIS in collaboration with ANU realises as computer software code - multifactorial and multi-scale - versatile abstraction level 39/40 Conclusion Broad applicability only comes from general concepts but we need good concepts to do so The ecosystem definition is rich enough to constrain a simulation platform Emergence is linked to hierarchical systems, which are easy represented as graphs... which are easy to implement as simulation tools (3Worlds)
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks