EnsemblES - Using ensemble techniques to capture the accuracy and sensitivity of ecosystem service models

Lead Research Organisation: Bangor University
Department Name: Sch of Natural Sciences


If the United Nations sustainable development goals (SDGs; https://sustainabledevelopment.un.org/) are to be achieved, it is vital to understand the interactions between people and nature. A significant aspect of these interactions can be classed as 'nature's contributions to people' (termed ecosystem services; ES). However, the empirical ES data needed to quantify these relationships are sparse in all parts of the World. Using recent advances in data availability from remote sensing, models are increasingly able to provide credible information where empirical data are lacking. Specifically, ES models produce maps of estimated ES (typically based on land cover and other driving variables) and so can provide the understanding of the spatial distribution and heterogeneity ES required to aid planning and optimisation of land use decisions. However, most ES modelling applications rely on a single model for each ES and few applications explicitly validate ES models against independent datasets. As a consequence, the uncertainties associated with each application of ES models (and the datasets that underpin) them remain largely unknown. This is a particular issue as the results of local-scale validation are likely not to be transferable to new locations or to the regional and national scales at which ES model outputs are most widely used.

EnsemblES seeks to address these issues by: 1) investigating ES model input sensitivity, varying initial conditions at the start of model simulations; 2) combining the outputs of multiple ES models (from multiple initial conditions) into 'ensembles' of models using a variety of techniques including when data on individual model performance is vs is not available; and 3) validating these model ensembles against independent data, highlighting a) the accuracy of ES ensembles, and b) whether coefficients of variation of the ensemble is a good predictor of ensemble uncertainty.

Planned Impact

The societal impact of EnsemblES will be to contribute to the attainment of the Sustainable Development Goals (SDGs; https://sustainabledevelopment.un.org/), by enhancing the tools available to decision-makers to manage ecosystem services (ES). Specifically, the findings of EnsemblES will enable decision makers to: 1) best use existing ES models to inform national and regional land use/cover change policies supporting ES management and promoting equality and justice amongst the beneficiaries of these ES; and 2) set priorities determining where scarce resources should be invested to improve effective management of ES. The primary beneficiaries will be regional and national decision-makers within the UK, and the secondary beneficiaries include academics and researchers working in cognate fields, as well as similar decision-makers with international remits.


10 25 50
Title Ensembles of ecosystem service models 
Description We produced a 90sec animation describing the benefits of using ensembles of ecosystem service models. This was released as the video abstract of a paper, but also shared with key decision makers 
Type Of Art Film/Video/Animation 
Year Produced 2021 
Impact None to date 
URL https://www.sciencedirect.com/science/article/pii/S221204162100156X
Description The project found that ensembles of ecosystem service models can indeed improve accuracy and indicate uncertainty. Researchers used validation data across a data-deficient area to test the accuracy of ensembles of models against the accuracy of individual models. This comparison found evidence that had at minimum a 5-17% higher accuracy than a randomly selected individual model and, in general, ensembles weighted for among model consensus provided better predictions than unweighted ensembles.

Although sometimes individual models can give good predictions, without validation against data or other models there is the potential for serious negative consequences to occur if estimates deviate significantly from the truth. For this reason, ensembles of models should be widely adopted so that sustainable development decisions can be made to protect the ecosystem services we all rely on. Our analysis suggests various ensemble methods should be applied depending on data quality, for example if validation data are available.
Exploitation Route Future scientists and decision-makers using ecosystem service models should use multiple different models together to look at the average outcomes. We suggest which methods of combining models are best.
Sectors Digital/Communication/Information Technologies (including Software),Environment

URL https://landscapedecisions.org/ensembles/
Description Our findings are currently being summarised in a ~2min animation. We have plans to disseminate this animation via virtual workshops with key stakeholders (as detailed in our proposal). In addition, via Landscape Decisions, we plan real-time panel sessions of 'national' level key stakeholders (i.e., Defra, FR, HS etc, EN, National Trust, NFU, CLA etc), and we have applied to showcase this animation at the COP26.
First Year Of Impact 2021
Sector Digital/Communication/Information Technologies (including Software),Environment
Impact Types Societal

Description Between environmental concerns and compliance: How does media messaging affect motivation and choice between disposable versus reusable facemasks?
Amount £343,974 (GBP)
Funding ID AH/W003813/1 
Organisation Arts & Humanities Research Council (AHRC) 
Sector Public
Country United Kingdom
Start 05/2021 
End 05/2022
Description Scaling up Off-Grid Sanitation
Amount £1,749,830 (GBP)
Funding ID ES/T007877/1 
Organisation Economic and Social Research Council 
Sector Public
Country United Kingdom
Start 06/2020 
End 09/2023
Description The impact of Covid-19 restrictions on recreation and use of green space in Wales
Amount £10,877 (GBP)
Funding ID ES/V004077/1 
Organisation Economic and Social Research Council 
Sector Public
Country United Kingdom
Start 05/2020 
End 11/2021
Title Ensemble outputs among contemporary ecosystem service models for water supply and aboveground carbon storage in the UK following 10 different methods 
Description This data set contains UK-wide maps of ten different among-model ensemble approaches for two services: above ground Carbon stock and water supply. The data for Carbon comes as fourteen TIF maps for above ground carbon storage at a 1-km2 resolution with associated world files: ten approaches, with a double option for two of those, together with maps of variation among models and among ensembles. For water, the data comes as one shapefile with polygons per watershed, each polygon containing these fourteen estimates. For all maps, 600dpi jpg depictions are added to the supporting information. Directory location independent layer files are included to aid scaling and providing the colour palettes. Ensemble output maps were calculated with different approaches following the supporting documentation and associated publication. Uncertainty estimates for these services are included as variation among contributing model outputs and among the employed ensemble approaches. The work was completed under the 'EnsemblES - Using ensemble techniques to capture the accuracy and sensitivity of ecosystem service models' project (NE/T00391X/1) funded by the UKRI Landscape Decisions programme. 
Type Of Material Database/Collection of data 
Year Produced 2021 
Provided To Others? Yes  
Impact None to date 
URL https://catalogue.ceh.ac.uk/id/a9ae773d-b742-4d42-ae42-2b594bae5d38
Title Ensemble outputs from Ecosystem Service models for water supply, aboveground carbon storage and use of water, grazing, charcoal and firewood by beneficiaries in sub-Saharan Africa 
Description This dataset contains the gridded estimates per 1 km2 for mean and median ensemble outputs from 4-6 individual ecosystem service models for Sub-Saharan Africa, for above ground Carbon stock, firewood use, charcoal use and grazing use. Water use and supply are identically supplied as polygons. Individual model outputs are taken from previously published research. Making ensembles results in a smoothing effect whereby the individual model uncertainties are cancelled out and a signal of interest is more likely to emerge. Included ecosystem service models were: InVEST, Co$ting Nature, WaterWorld, Monetary value benefits transfer, LPJ-GUESS and Scholes models. Ensemble outputs have been normalised, therefore these ensembles project relative levels of service across the full area and can be used, for example, for optimisation or assignment of most important or sensitive areas. The work was completed under the 'EnsemblES - Using ensemble techniques to capture the accuracy and sensitivity of ecosystem service models' project (NE/T00391X/1) funded by the UKRI Landscape Decisions programme. 
Type Of Material Database/Collection of data 
Year Produced 2020 
Provided To Others? Yes  
URL https://catalogue.ceh.ac.uk/id/11689000-f791-4fdb-8e12-08a7d87ad75f