Forecasting Change in Vegetation Dependant Ecosystem Functions.

Project Information

EcoFunc4Cast

Grant agreement ID: 703258

DOI

10.3030/703258

Project closed

EC signature date 3 February 2016

Start date 1 April 2017

End date 31 March 2019

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 200 194,80

EU contribution

€ 200 194,80

Coordinated by

AARHUS UNIVERSITET
Denmark

Periodic Reporting for period 1 - EcoFunc4Cast (Forecasting Change in Vegetation Dependant Ecosystem Functions.)

Reporting period: 2017-04-01 to 2019-03-31

Summary of the context and overall objectives of the project

Realizing the potential threat of climate change on ecosystem functioning requires a capability to make reliable forecasts of species range shifts in response to environmental change . For this reason, species distribution modelling (SDM) has received much attention . Static SDMs, also referred to as ‘niche based’ models relate field observations to environmental predictor variables. They are widely used among ecologists with relative success; nevertheless ‘ecological realism’ is not always well considered . Key theoretical limitations are accountable: (1) observed species patterns are assumed to be in equilibrium with the environment, and (2) several processes crucial to species range shifts, notably propagule dispersal , biotic interactions and establishment probability are disregarded, often adopting an ‘all or nothing’ model parameter for dispersal capability.
Evaluating the effects of species range shifts on European vegetation-related ecosystem functions is of upmost importance for improving our understandings of climate change impacts upon European societies. However, this can only be achieved through accurate species range forecasts, for which the ‘ecological realism’ of species’ response models to climate must be improved. This can be achieved through use of hybrid mechanistic models; models that consider both standard niche-based habitat suitability projections and mechanistic simulations of local demography and propagule dispersal.

This project’s overall objective was to address exactly this challenge: develop a novel mechanistic modelling framework in which the ecological realism of forecast models can be improved.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

(1) A novel climate downscaling model with the aim to improve the environmental realism of current climatic downscaling procedures when applied across highly heterogeneous landscapes.

Preliminary results of the works have been showcased at the British Ecological Society (BES) Annual Conference (2018) and discussed in a broader context of species distribution modeling in the Anthropocene at the Macroecological Special Interest Group Conference in St Andrews, Scotland (2018). A methods manuscript is in preparation, pending evaluation of validation statistics.

(2) A novel pseudo-absence selection method was developed, one which broadly considers the species pool of cells prior to generating background localities.

Results have been presented in the context of presence only SDM at two international conferences: International Biogeography Society,, Arizona 2017 and the British Ecological Society (BES) Conference (2018), UK. A methods manuscript is in preparation.

(3) We develop a framework for computing dispersal distance (building on extant published works: e.g. Tamme et al. 2014; Thompson et al. 2010). The result will be an R toolbox (under development) which enables users to estimate dispersal syndrome and ‘binned dispersal kernel’ for their own list of species. Users can add their own habitat-specific trait data to enhance dispersal estimations or the toolbox can use relevant available data from open access databases.

(4) After a long century of debates around Reids paradox, we aimed to review the current situation. Migration has been now estimated for numerous species in different continents, and local dispersal has been measured in a lot of different landscapes around the world. We ask if there are particular species, or biogeographical contexts, where the population dynamics is understood accurately enough that the paradox vanishes. What is today the residual difference between species migration and landscape spread estimates and the new hypothesis to explain it ? Given that migration and landscape spread do not occur at the same spatio-temporal scale, we finally provide guidelines to compare the two kinds of estimates.

Two manuscripts are in preparation, one a methods paper and another a theoretical essay. The latter is very much work in progress. Upon acceptance of all manuscripts discussed above, open access will be guaranteed in accordance to Article 29.2 of the Grant Agreement. Further, all forthcoming publications directly relating to work implemented under this Horizon 2020 Fellowship will acknowledge the funding received from the Marie Sklodowska-Curie programme (Article 29.4 of the grant agreement).

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

Understanding how biodiversity and ecosystems will respond to the striking rates of climate and land-use change that characterize the Anthropocene world is one of the most important questions for contemporary science. In this endeavour, species distribution modelling (SDM) in response to anthropogenic changes have been widely applied, yet their ‘ecological realism’ is not always well considered . Practical and theoretical limitation are accountable: e.g. coarse-grain environmental data and inept downscaling methodologies, arbitrary absence estimates, processes crucial to species range shifts, notably propagule dispersal and establishment probability being disregarded .

As part of this EU funded Marie Sklodowska-Curie Fellowship, EcoFunc4Cast has specifically targeted much of these limitations extending the state of the art in view of a novel conceptual framework for mechanistic SDMs. The framework adheres to suggestions set out to improve biodiversity forecasts , namely: obtain good information and make better use of it; improve widely used modelling methods; account for multiple causes of changes in biodiversity and crucially; evaluate models before using them. Given the weight of importance of improving modelling approaches prior to applying them, EcoFunc4Cast invested vastly more resource towards framework development over application. The application of this framework is now the logical extension of this work, demonstrating the utility of improved ecological realism in biological forecasts, both taxonomic and functional.

To summarize, a conceptual paper discussing these methodological tools for facilitating SDMs in the Anthropocene is in prep. Models being only as reliable as the input data we address climate downscaling and validation through Geographical Weighted Regression, explicitly accounting for non-stationarity relationships between microclimate and geodiversity . We introduce an improved approach to pseudo absence generation, one that utilises the dark-diversity concept, and we address multiple causes of changes in biodiversity through the application of dispersal dynamics, introducing and exemplifying in itself one novel framework for estimating plant species dispersal dynamics. We discuss the utility and application of our framework within highly heterogeneous landscapes and microclimatic research. Our hope is to open a new chapter of discussion for explicitly consideration of the ecological realism of forecast modelling and see wide adoption and uptake of our developed framework for improving our understanding of anthropogenic driven environmental change at timescale immediately relevant to today societies.

This work has been presented at the BES Macroecology Special Interest Group Annual Conference, St Andrews, Uk 2018; and was invited for pre-submission to Ecography under their E4 award for excellence in Ecology in Evolution. Upon acceptance, open access this work will be guaranteed in accordance to Article 29.2 of the Grant Agreement and will acknowledge the funding received from the Marie Sklodowska-Curie programme (Article 29.4 of the grant agreement).

Macro2MicroClim