• No results found

Development of a conceptual framework for including frugivory and its effects on ecosystem dynamics in the Madingley Model

N/A
N/A
Protected

Academic year: 2021

Share "Development of a conceptual framework for including frugivory and its effects on ecosystem dynamics in the Madingley Model"

Copied!
15
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Research Proposal

Development of a conceptual framework for including frugivory

and its effects on ecosystem dynamics in the Madingley Model

Amalia Llano

Contact email: amallanbo@gmail.com

Student number: 12162337

Supervisor / Examiner: dr. rer. nat. Daniel Kissling Assessor: dhr. prof. dr. ir. Willem Bouten

Universiteit van Amsterdam | IBED | BIOMAC May – June 2019

(2)

1

Contents

Summary ... 2

Introduction ... 2

Research Aim and Objectives... 4

Research Questions ... 5

Theoretical Framework ... 5

Methods ... 10

Literature Review and Data Collection ... 10

Conceptual Framework ... 11

Implementation ... 11

Expected Results ... 11

Timeline ... 13

Data Management Plan ... 13

Knowledge Utilisation ... 13

Funding and Equipment ... 14

(3)

2

Summary

The rapid changes being experienced by the Earth system as a result of human activity in the Anthropocene are creating unprecedented pressures on ecosystems and the services they provide. These new circumstances and interactions are complex and nonlinear in nature and understanding them requires multidisciplinary approaches that have both an empirical component and accurate predictive power. The Madingley Model, referred to as a General Ecosystem Model by its authors, was developed as an answer to this demand for a tool that can use existing empirical research to model, and eventually predict, ecosystem dynamics under different scenarios. Madingley was designed as a flexible, individual-based model that can be modified to include different ecosystem processes, organisms, climates, and situations. Frugivory, a process essential to the establishment and regeneration of many populations of plants with fleshy fruits in tropical forests, is not yet included in the Madingley Model. Aside from being the link between plants and their seed banks, frugivory has also been found to be affecting the carbon storage capacity of tropical forests by playing a key role in determining the composition of plant communities in these ecosystems. Hence, we propose to explore the question of how this process could be eventually included within the model, not only to enrich the simulated dynamics of terrestrial autotrophs, but to ultimately explore the effects of frugivory on ecosystem services like carbon storage, and to evaluate the consequences that the defaunation of large-bodied frugivores has on ecosystem dynamics.

Introduction

The presence of humans on the planet has had profound consequences. From ocean acidification to landscape fragmentation, biodiversity loss, and climate change, anthropogenic activities have altered the Earth’s ecosystems and geological cycles so significantly that they have come to define a new epoch, the Anthropocene (Otto, 2018; Dirzo, et al., 2014). The Anthropocene is characterised by what is known as ‘global change’, which broadly refers to the biophysical and socioeconomic changes that alter the functioning of the Earth system. It involves land use and land cover changes, urbanisation, globalisation, changes in atmospheric composition, climate, and geochemical cycles, biodiversity loss, resource use, and pollution, among others. It is noteworthy that most of these changes are complex in nature, interrelated, and show strong nonlinearities (Steffen, Crutzen, & McNeill, 2008).

(4)

3 Global change and human impact have degraded ecosystems and the life-supporting services they provide so extensively, that it is now accepted that life on Earth and the future of contemporary societies are at risk (Steffen, Crutzen, & McNeill, 2008). It has therefore become paramount to understand, predict, and mitigate anthropogenic impacts through the use of a variety of tools, including policy instruments, conservation measures, and scientific research (Harfoot, et al., 2014). Particularly, research needs to address the way in which small-magnitude ecological processes scale up to Earth systems like climate, ocean circulation, and the water cycle (Mace, 2013), while registering the complexity and nonlinearities typical of these processes.

This scenario demands the development of mechanistic ecological models that use accessible measurements and parameters to explain and, most importantly, to predict, the effects and outcomes of different global change scenarios on ecosystems and their services. Therefore, these models, embedded within a systems theory framework, ought to span more than one level of complexity. They should also aim to capture how processes at the individual scale, including species interactions, can lead to emergent behaviour at higher levels of complexity (especially at a global scale). Additionally, they should combine the predictive power of process-based models with empirical information and the accuracy of conventional measurement-based models (Landsberg, 2003). Models like these, which explicitly represent the biological, physiological, and ecological mechanisms that underly a system’s functioning, are and will be instrumental to policy and decision-making in the context of global change (Harfoot, et al., 2014).

In 2014, Harfoot et al. published the first version of the Madingley Model. The authors refer to this model as the “first General Ecosystem Model” because it attempts to use fundamental ecological theory (concepts and processes) to simulate both terrestrial and marine ecosystems at various spatial scales. The modelling framework is designed so that the fate of each individual organism can be captured and modelled using an encoded version of its behaviour and biology, which also captures its interactions with the abiotic environment. This model uses core biological and ecological processes to predict emergent properties at the ecosystem level that arise as a result of the behaviour of individuals. Due to its unique modelling framework, its inherent flexibility, its modularity, and its global nature, the Madingley Model is well-suited for studying questions about sustainability and for predicting human impact

(5)

4 on both terrestrial and marine ecosystems, and thus for addressing some of the critical challenges of global change.

Here we propose using Madingley to focus on the effects of frugivory on ecosystem dynamics and ecosystem services in tropical forests. Particularly, we intend to pay special attention to the recent evidence that suggests that the defaunation of large frugivores may be indirectly affecting the carbon storage capacity of tropical forests (Bello, et al., 2015; Osuri, et al., 2016). Since carbon storage is a crucial ecosystem service, especially in times of global change, understanding the temporal and spatial dynamics of complex interactions like frugivory and its effects on overall ecosystem functioning is paramount. Furthermore, because frugivory is the link between many fleshy fruiting plants and their seed banks in ecosystems like tropical forests, the regeneration and dynamics of these plant populations are largely dependent on seed dispersal by frugivores (Jordano, 2000). This makes frugivory a central process in plant population ecology and considering it in Madingley could offer greater resolution and accuracy in modelling land autotroph ecology. Therefore, the goal of this project is to construct a conceptual modelling framework based on empirical research that can eventually be inserted within the Madingley Model to study the effects of frugivory on ecosystem dynamics, with a special focus on defaunation and the carbon storage capacity of tropical forests.

Research Aim and Objectives

The aim of this project is to identify relevant parameters, relationships, and knowledge gaps in literature and use them to propose a conceptual framework that can later serve as a basis to develop a modified version of the Madingley Model that captures the effects of frugivory on ecosystem dynamics.

The specific objectives related to this aim are:

i. Conduct a literature review to identify any parameters and relationships that can be used to construct the new conceptual framework, as well as any knowledge gaps that need to be filled to incorporate frugivory interactions within the Madingley Model ii. Identify, define, and organise the necessary elements of the new conceptual

(6)

5 iii. Propose new equations and relationships that can be later expanded and joined to include frugivory interactions and their effects on ecosystem dynamics in Madingley iv. Run partial implementations of the proposed relationships and/or parameters to

assess their properties or their scope

Research Questions

This project aims to answer the following research question:

What is an appropriate conceptual framework to include frugivory and its effects on ecosystem dynamics in the Madingley Model?

To answer this question, the following sub questions are considered:

i. What information is required to include frugivory and its effects on ecosystem dynamics in the Madingley Model?

ii. What information and data are available in literature which can be used to parametrise a conceptual framework that includes frugivory and its effects on ecosystem dynamics within the Madingley Model?

iii. What knowledge gaps exist in literature which need to be addressed to include frugivory and its effects on ecosystem dynamics in the Madingley Model?

Theoretical Framework

The Madingley Model

The Madingley Model was initially developed as a joint effort between UN Environment World Monitoring Centre and Microsoft Research in Cambridge. It was designed as a ‘simple’ model expected to evolve and to be improved over time by a growing community of users and developers. Therefore, the model’s code, parameters, and equations are openly available online at https://madingley.github.io/. Since its publication in 2014, the model has been applied to investigate the effects of habitat loss and fragmentation on ecosystems (Bartlett, Newbold, Purves, Tittensor, & Harfoot, 2016), and several institutions are working as partners of Madingley in topics such as conservation, the mapping of functional groups to species, the improvement of the ecological realism of the model, agricultural systems, the improvement of the model’s software, the implications of projected scenarios of biofuel production, tipping points and ecosystem change, fisheries, nutrient dispersal, and the biodiversity planetary boundary, among others (for more information, see

(7)

6

https://madingley.github.io/partners/). The numerous projects working with or on the Madingley Model exemplify the model’s adaptability and flexibility, as well as its potential to be further improved and used to investigate a vast array of ecological questions.

Madingley’s great potential to be used in ecological research is a result of its trait- and individual-based design and its mechanistic, process-based nature. This means, first, that in Madingley’s modelling framework organisms are characterised by their functional traits, which are divided into categorical traits (like feeding mode) and continuous traits (like body mass). This approach allows taxonomic identity to be excluded from the model, thus eliminating the need for detailed taxonomic literature and data (which is often incomplete, outdated, or completely unavailable).

Second, an individual-based approach is used for all organisms except autotrophs, which are modelled as stocks (which are defined as total biomasses of these organisms). All groups of heterotrophs are modelled in cohorts, which are defined as groups of organisms that occur within the same spatial unit and have identical functional traits (like body mass, feeding mode, reproductive strategy, etc.). These cohorts, then, act as ‘individuals’, so this grouping serves the double purpose of maintaining an individual-based model while keeping computation times manageable.

Third, Madingley is defined as a mechanistic model because it relies on basic ecological processes to simulate the individual behaviour of heterotroph cohorts and autotroph stocks. Explicitly, the processes it considers are primary production for autotrophs, and eating, metabolism, growth, reproduction, and mortality for heterotrophs. Ecosystem function and structure emerge from the combination of these processes, which operate at the individual level within each spatial unit of the model. In each simulation, autotroph stocks and heterotroph cohorts interact through space and time. For each time step and spatial unit (grid cell), the model runs a series of organised steps, starting with the implementation of the growth function that changes the autotroph biomass (for detailed information, refer to Harfoot et al. 2014).

The ecology of the heterotroph cohorts is modelled next. The order in which the processes governing heterotroph cohort behaviour are modelled is random for every time step. Per time step, each cohort of heterotrophs performs the processes illustrated in Figure 1. In the order exemplified in the figure, individuals in the cohort: metabolise biomass to gain energy (i); eat

(8)

7 biomass according to their feeding mode (they eat autotrophs if they are herbivores, eat other heterotrophs if they are carnivores, or eat both if they are omnivores) (ii); use net biomass gain to grow (for juveniles) or to store it as reproductive potential (if their biomass is enough to qualify them as reproductively mature) (iii); reproduce (give birth to a separate offspring cohort) if they’ve accumulated enough biomass (iv); die as a result of various processes (starvation, background mortality, or senescence) (v); and disperse to an adjacent grid cell (vi). The interaction between these processes in space and time results in the self-assembly of communities of cohorts and in the emergence of ecosystem structure and function (Figure 1 B).

Figure 1. Figure and caption taken from Harfoot et al. (2014). Schematic of the model. Ecosystem structure and function (B) emerges from a combination of processes operating on individual organisms within a grid cell (A), and exchange of individuals among grid cells via dispersal (C).

As briefly mentioned before, the model also includes a spatially explicit representation of the environment. For both land and water, this is defined as a two-dimensional layer that is divided into grid cells within which organisms are assumed to be well-mixed. Publicly

(9)

8 available datasets are used to assign abiotic environmental conditions to each grid cell. For the terrestrial cells, these variables are: air temperature, precipitation, soil water availability, number of frost days, and seasonality of primary productivity. For the marine cells, the model uses sea-surface temperature, net primary productivity (NPP), and ocean current velocity. Although the authors select specific sources for each of these parameters and use them in their initial simulations, the model is flexible with respect to the data used, especially to allow for the simulation of future environmental conditions. Additionally, heterotrophs can disperse through neighbouring cells, thus making the dynamics of one cell dependent on its interactions with its neighbours.

Acknowledging the model’s built-in flexibility, Harfoot et al. (2014) give a set of initial recommended pathways of improvement for the model and they mention that the ecological processes it captures are expected to evolve based on the community’s specific needs. Here we consider three additional processes that can be included and analysed in Madingley, although they are not explicitly mentioned by the authors: frugivory, fruit production, and the carbon storage capacity of tropical forests. The inclusion of these processes responds to a growing amount of evidence that suggests that defaunation of large frugivores has significant consequences on the carbon storage capacity of tropical forests and, therefore, indirectly affects climate change and its mitigation.

Frugivory, defaunation, and changes in carbon storage

A study published in 2015 found that in tropical forests where animal-dispersed plants are abundant, the extinction of large frugivores will result, over time, in decreased carbon storage (Bello, et al., 2015). The rationale behind this is that eliminating large fruit-eaters limits the recruitment of species with large seeds and therefore induces compositional changes in the plant community. This, in turn, affects the community-aggregated values of wood density and height, and causes a reduction in the carbon storage capacity of the forest. By simulating extinctions of large frugivores in a model built using plant data from the Atlantic forest, Bello et al. found that defaunation scenarios led to a decrease in carbon storage capacity. In their study, the authors also propose a functional relationship between seed diameter and traits related to carbon storage, such as wood density and maximum tree height.

A similar study published in 2016 re-examined the relationship between carbon storage capacity and defaunation of large frugivores in the global tropics and found that the effects

(10)

9 of defaunation vary depending on the local species composition of the tree communities of tropical forests (Osuri, et al., 2016). Therefore, to evaluate the effects of large-frugivore extinctions, it is important to consider the identity (at least trait-wise) of the tree species that replace the individuals that are lost in defaunated forests. Osuri et al. also raise questions related to the time scales of the relationship between defaunation and carbon storage and propose that more long-term empirical data paired with temporally explicit models are required to determine the magnitudes and time frames of these changes. Although their results vary in some significant respects, both studies agree on the fact that in forests composed mainly of animal-dispersed plant species, the defaunation of large frugivores has a negative effect on aboveground carbon storage. By using a temporally and spatially explicit mechanistic model like Madingley, these processes could be analysed further.

Inclusion of frugivory in the Madingley Model

Including these processes in Madingley implies adding an additional level of complexity to the autotroph model to account for differences in fruit size, plant adult size, and dispersal mode, and introducing a new group of heterotrophs (frugivores). This would require, among other things, redefining and refining the allocation of plant primary productivity on land to different parts of the plant, like fruits and seeds, and explicitly structuring terrestrial plant communities. Currently, terrestrial autotrophs in Madingley are modelled as stocks using a model known as the climate driven-terrestrial model (Smith, Purves, Vanderwel, Lyutsarev, & Emmott, 2013; Harfoot, et al., 2014).

In Madingley, plant growth on land is a function of NPP, the proportion of NPP produced by deciduous leaves, and the fraction of NPP allocated to structural tissue. All of these are dependent on local climate. Biomass loss in terrestrial plants is determined by the rate of leaf mortality, which is a function of the climate and of the consumption of leaves by herbivores. Overall, the biomass of terrestrial plant stocks in each time step of the model is determined by the biomass in the previous time step, the gain in biomass (growth) and the mortality. This modelling approach avoids structuring the plant community into functional groups, and only considers biomasses of deciduous leaves, evergreen leaves, and the allocation of NPP to structural tissue. Fruits are not explicitly considered in the model, although they are implicitly included within the plant biomass that isn’t structural tissue and that is available as a food source for herbivorous and omnivorous heterotrophs.

(11)

10 Moreover, the Madingley Model considers several functional groups of terrestrial heterotrophs that are assembled into cohorts. These groups are classified according to their feeding mode (herbivore, carnivore, or omnivore), their mobility, their reproductive strategy (iteroparous, semelparous), and their thermoregulation mode (ectotherm, endotherm). Currently, a frugivore group is not explicitly included, since herbivores are assumed to consume leaves and therefore influence the mortality of autotroph stocks, and omnivores are both predators (carnivores) and herbivores, which means they consume other heterotrophs and also have an effect on plant mortality. Frugivores, on the contrary, would have a neutral effect on plant mortality and would be important in plant reproduction and dispersal, which aren’t directly included in the Madingley Model.

Within this context, therefore, relatively indirect threats to ecosystems and ecosystem services like the defaunation of large frugivores could be studied using the Madingley Model by: 1) incorporating changes in the autotroph modelling framework that capture the way in which plant communities are structured with respect to functional traits related to fruit production, allocation of NPP to different parts of the plant, and carbon storage; 2) adding a new functional group of heterotrophs that act as frugivores; 3) establishing parameters that link specific plant traits to carbon storage capacity; and 4) linking the frugivore functional group to the assemblage of plant communities through a reproductive term related to seed dispersal. To effectively incorporate these aspects in Madingley, it is essential to first conduct a thorough literature review aimed at identifying empirical parameters and relationships that can be included in the model. These data can later be used to propose a modelling framework that clearly outlines how to include frugivory and its effects on ecosystem dynamics in the model.

Methods

Literature Review and Data Collection

To create a conceptual modelling framework that describes a comprehensive strategy to incorporate frugivory in the Madingley Model, it is first necessary to carry out a targeted literature review. In this revision, any relationships, data, parameters, and concepts that are pertinent to the development of the final conceptual framework will be identified and extracted. The revision, although not systematic, will encompass papers and books that contain empirical information on topics such as plant biomass allocation, fruit production

(12)

11 measurements, allometric relationships in plants, carbon storage in tropical forests, fruit-frugivore relationships, and any useful data that can be used to parametrise the model. This review will also encompass theoretical papers on the subject, as well as anything that can hint to knowledge gaps that need to be addressed before fully incorporating frugivory in the Madingley framework. Findings resulting from this review will be collected in appropriately cited Excel tables (which will contain parameter values and knowledge gaps) and in a written theoretical framework that will serve as the base for the new conceptual framework.

Conceptual Framework

The conceptual framework will be constructed from the already existing Madingley framework and the information gathered from the literature review. The final product will be a written strategy that describes how to incorporate frugivory in the Madingley Model, and that includes parameters, data, mathematical relationships, figures (including modelling schematics), and an accompanying theoretical framework (that includes the knowledge gaps, if any, that need to be addressed before fully incorporating frugivory in Madingley).

Implementation

Depending on the information that is gathered from the literature review, partial implementations of the mathematical relationships can be carried out to evaluate how plausible it would be to include them in the Madingley Model. These implementations could vary, but they would be primarily aimed at exploring the behaviour of the proposed parameters and/or mathematical relationships and providing feedback with which to improve the conceptual modelling framework.

Expected Results

This project is designed to result in a collection of tables, schematics, figures, mathematical relationships, and written text that will serve as a base to include frugivory and its effects in ecosystem dynamics in the Madingley Model. These results will also include discussions on significant knowledge gaps that need to be addressed to fully incorporate frugivory in the model. Specifically, such discussions should focus on how these gaps can be bridged by either making assumptions in the modelling approach, or by suggesting unexplored empirical research pathways aimed at collecting the missing information.

(13)

12

Workflow

Figure 2. Workflow. This workflow illustrates the steps that will be taken to address the project's questions and objectives, as well as the expected products and results.

(14)

13

Timeline

The project has been programmed to be completed in seven months (42 EC project), starting on the 12th of June 2019. During this time, the workload will be divided as illustrated in Table

1.

Table 1. Timeline. This table illustrates how the workload for the project will be divided over the seven months available for its completion.

Data Management Plan

The data collected in this project will be managed according to the guidelines suggested by van Loon (van Loon, unpublished document). The parameters and data that result from the literature search will be stored in Excel tables, as discussed in the method section, which will be stored in a repository. This repository will also contain the conceptual framework, figures, and other information or results relevant to the project. The repository will be shared with the project’s supervisor and examiner, as well as with Harfoot and collaborators, so that it can be eventually included in the open-source Madingley Project repository.

Knowledge Utilisation

The results of this project will contribute to the growth and development of the Madingley project. As discussed in the data management plan, sharing the data repository with Madingley’s authors will contribute to further enriching the existing open-source, collaborative general ecosystem model in order to enhance its predictive capacity, realism, and scope. These results are meant to serve as a base to develop a more comprehensive modelling framework that includes frugivory, so that relevant ecosystem properties, like carbon storage capacity, or relevant processes like defaunation, can be better studied, understood, and simulated. Moreover, these results can also serve as the starting point for empirical research that bridges existing knowledge gaps on frugivory.

No. Task/Date* 12 19 26 3 10 17 24 31 7 14 21 28 4 11 18 25 2 9 16 23 30 6 13 20 27 4 11 18 25

1 Literature review 2 Studying Madingley Model 3 Information gathering (tables) 4 Development of conceptual model

5 Implementation of new modelling framework 6 Writing

7 Revising

8 Work on presentation

*Each box represents one week (for a total of 28 weeks, corresponding to 42 EC)

(15)

14

Funding and Equipment

Since this is a theoretical project based on literature reviewing and modelling, and because all Madingley material is open source, no additional funding or equipment are needed.

References

Bartlett, L. J., Newbold, T., Purves, D. W., Tittensor, D. P., & Harfoot, M. B. (2016).

Synergistic impacts of habitat loss and fragmentation on model ecosystems . Proc. R.

Soc. B, 283: 20161027.

Bello, C., Galetti, M., Pizo, M. A., Magnago, L. S., Rocha, M. F., Lima, R. A., . . . Jordano, P. (2015). Defaunation affects carbon storage in tropical forests . Sci. Adv. , 1:e1501105. Dirzo, R., Young, H. S., Galetti, M., Ceballos, G., Nick, I. J., & Collen, B. (2014).

Defaunation in the Anthropocene. Science, 401-406.

Harfoot, M. B., Newbold, T., Tittensor, D. P., Emmott, S., Hutton, J., Lyutsarev, V., . . . Purves, D. W. (2014). Emergent Global Patterns of Ecosystem Structure and Function from a Mechanistic General Ecosystem Model. PLoS Biology.

Jordano, P. (2000). Fruits and Frugivory. In M. Fenner, Seeds: the ecology of regeneration in

plant communities (pp. 125-166). Walingford, UK: CABI Publ.

Landsberg, J. (2003). Modelling forest ecosystems: state of the art, challenges, and future directions. Canadian Journal of Forest Research, 385-397.

Mace, G. M. (2013). Global change: Ecology must evolve. Nature, 191-192.

Osuri, A. M., Ratnam, J., Varma, V., Alvarez-Loayza, P., Hurtado Astaiza, J., Bradford, M., . . . Sankaran, M. (2016). Contrasting effects of defaunation on aboveground carbon storage across the global tropics. Nature Communications, 7:11351 .

Otto, S. P. (2018). Adaptation, speciation and extinction in the Anthropocene . Proc. R. Soc.

B, 285: 20182047.

Smith, M., Purves, D., Vanderwel, M., Lyutsarev, V., & Emmott, S. (2013). The climate dependence of the terrestrial carbon cycle including parameter and structural uncertainties. Biogeosciences, 583–606.

Steffen, W., Crutzen, P. J., & McNeill, J. R. (2008). The Anthropocene: Are Humans Now Overwhelming the Great Forces of Nature? AMBIO A Journal of the Human

Referenties

GERELATEERDE DOCUMENTEN

Named Entity Extraction and Linking Challenge: University of Twente at #Microposts2014..

This allows hybrid g8p proteins with quite large foreign peptides to be displayed on the virion surface, even though the hybrid protein by itself cannot support phage

A!very!visible!similarity!to!Ruddle!and!Chesterfield’s!theory!is!that!knowledge!and!skills!were!in!all! domains! acquired! through! learning! and! cumulatively! practicing! a!

In order to measure the variables a survey was conducted. Existing scales for these variables used and/or adapted and translated from English to Dutch. This because the

Performance of micro gas turbines is governed by certain operating parameters, and the effect these parameters have on the turbine's performance will be proven by

With respect to the regression results with only the 2 outliers excluded, the significant coefficients of the control variables are the one for the openness to trade

has demonstrated that there is a significant difference between pro-environmental consumption, simple EOA as well as difficult EOA; not only in terms of how likely individuals are