WISER Which Ecosystem Service Models Best Capture the Needs of the - - PowerPoint PPT Presentation

WISER Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

James Bullock NERC Centre for Ecology and Hydrology jmbul@ceh.ac.uk

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

Ecosystem services: a complex concept

The UKNEA

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

The need for ecosystem service modelling

• moving beyond mapping proxies
• Link variables measured in the field to

ecosystem services & goods

• Map ES & G over large regions
• Provide information on ES & G in poorly

reported regions

• Allow future scenarios to be explored
• Assess possible outcomes of policy actions
• As with all models – provide a test of our

understanding of the processes driving ES & G levels

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

Issues in modelling ecosystem services

Simple models Complex models Less data (type, resolution, frequency, etc) More data Less accurate (predictions, representation, etc) More accurate Less utility (scenario/policy assessment, etc) More utility Greater ease of use (training, software, etc) Lesser ease of use

• User can run Co$tingNature online with ease and no need for own

data

• For anywhere in the world – uses 150+ intrinsic input maps and a

spatial model

Simple ecosystem service model – Co$ting Nature

http://www.policysupport.org/costingnature

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

• Focused on certain ecosystem services

(carbon, water, hazard mit., tourism) biodiversity metrics.

• Provides scenario tools for climate change

and land use change

• But, limited range of ES and simplistic,

proxy-based modelling

• Cannot assess uncertainty

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

More complex ecosystem service model – InVEST

www.natural capitalproject.org/InVEST.html

• GIS-based spatial modelling of ES
• Wide range of ES
• Range of model complexities

and data needs (Tier 0 very simple)

• Biophysical & economic

value outputs

• Can be used with

stakeholders to explore scenarios

• Poor at dealing with

uncertainty

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

InVEST pollination model

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

More complex ecosystem service model – ARIES

www.ariesonline.org

• Web-based artificial intelligence system that customizes models to

user goals

• A mapping process for ES provision, use, and flow
• 3 elements: provision

areas, flow paths, areas of use

• Probabilistic models

provide likelihood of all possible outcomes (uncertainty)

• Adaptable to amount

and type of data available & user needs

Example: using poverty, population density, pollution, habitat suitability and harvest data.

• 1. total demand for subsistence fisheries
• 2. met demand fraction
• 3. unmet demand fraction

1 2 3

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

Overarching Aim: To identify what constitutes the simplest adequate ecosystem service modelling framework to inform effective policy and management interventions for poverty alleviation at appropriate spatial and temporal scales

Obj1: To explore the level of model complexity required to map, in sufficient detail to inform policy, ES of importance to poverty alleviation in sub-Saharan Africa.

RQ 1: Which models of ES accurately map the distribution of biophysical stocks of ES of relevance to the poor? RQ2: Are there ES for which particular models perform better than others? RQ3: Are there poverty contexts for which the models perform better (e.g. forests vs drylands, subsistence farming vs. harvesting systems)? RQ4: What are the key gaps in existing data and model capacity needed to model the biophysical stocks of ES that are most relevant for poverty alleviation?

Objective 2: To explore the potential and synergies of existing models of ES to make explicit the links between services, benefit flows and wellbeing of the poor.

RQ5: Are current models able to capture use of and demand for ES by poor beneficiaries? RQ6: Are there ES for which particular models perform better than others? RQ7: Are there poverty contexts for which the models perform better in terms of capturing use and demand by poor beneficiaries? RQ8: What are the key gaps in existing data and model capacity needed to model the use of and demand for ES by poor beneficiaries?

Project team

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

CEH, UK: James Bullock & Danny Hooftman Southampton University, UK: Felix Eigenbrod, Simon Willcock, Terry Dawson, Malcom Hudson, Kate Schreckenberg. CSIR, S. Africa: Belinda Reyers & Patrick O’Farrell BC3, Spain: Ferdinando Villa & Elena Pérez-Miñana

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

Country Biophysical data Beneficiary data Crop production Stored carbon Water availability NTFPs Grazing Pollination Crop production Water availability NTFPs Grazing Pollination Benin   Burkina Faso   Cameroon   Equatorial Guinea   Ethiopia   Ghana     Kenya       Malawi           Namibia   Nigeria   Sierra Leone   South Africa        Tanzania        Uganda   Zambia    

Target Ecosystem Services & Countries

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

Project Partners & Investigators – obtaining data for ES modelling

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

Source Data & location ESPA ASSETS General ES and beneficiaries data for Zomba region of Malawi Center for International Forestry Research NTFP data from 17 African case studies CIAT Soils Research Area Water availability data for Burkina Faso and Ghana Valuing the Arc ES data for the Eastern Arc mountains in Tanzania CHIESA Data on crop yields and food security in Tanzania, Kenya, and Ethiopia Indiana University Household data on rural livelihoods and water use from Kenya and Zambia Malcolm Hudson Rangelands Trust Household data on trends in well-being and benefits from ES in Kenya Charlie Shackleton Household-level NTFP availability and use in S. Africa and

• ther sub-Saharan nations

CSIR

• S. African data

e.g. S. African data

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

Biodiversity Water availability

Household use of dung & wood for fuel Dispersed rural communities

Methods to compare models in terms of complexity, data needs & fit to stakeholder needs

NB: we are comparing model complexities, not modelling platforms per se

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

• Common units of comparison, by normalising outputs and comparing

ranks

• Intensity of data requirements; data availability, spatial resolution and

uncertainty

• Quantify model discrepancies statistically
• accuracy in predicting the distributions of ES based on field data
• uncertainty, i.e. changes in accuracy with location, service and spatial

resolution

• sensitivity of model outputs to parameter values
• Evaluate the impact of model discrepancies on the usefulness of the

model for the decision-making process

• including acceptable limits of error for policy contexts

Work plan

WISER: Which Ecosystem Service Models Best Capture the Needs of the Rural Poor?

• Phase 1 – Stakeholder consultation. Engagement with African and

global stakeholders (governmental and non-governmental

• rganisations), to identify their needs for ES models to guide and

inform policy and interventions - e.g. what outputs are needed and at what temporal and spatial scales? + obtain data from partners. Months 0-3

• Phase 2 – Biophysical model comparison. Run InVEST, ARIES and

Co$ting Nature for biophysical modelling of ES provisioning in target regions and for target ES. Months 3-15

• Phase 3 – Beneficiary model comparison. Create and compare

different tiers of models estimating the social aspects of ES provisions, namely access and utilisation. ARIES, InVEST and local

• S. African models. Months 15-27
• Phase 4 – Dissemination. To stakeholders. Months 27-30