Quantifying the benefits ecosystem services to people is a central goal of ecosystem science. “Big data” can have an important role in increasing this understanding of the interactions of humans and nature. Prof Simon Willcock (Rothamsted Resarch & Bangor University) delivered a DSNE seminar on the importance of surveying ecosystem services, and innovate methods to do so.

What is an Ecosystem Service?

In general terms, ecosystem services are defined as natures contribution to people. Ecosystem services are primarily viewed through an anthropocentric lens – what is nature doing for us and how is it helping people – and the study of such combines both natural and social sciences. There are generally four types of ecosystem service:

· Provisioning services - where we get some type of material good i.e. food;

· Regulating services - that give us a safe and liveable environment, i.e. regulation of atmospheric temperature;

· Supporting services – often associated with regulating services, they make it possible for ecosystems to continue providing services i.e. nutrient cycling

· Cultural services – unique to the individual, how does the environment make you feel i.e. a sense of place, spiritual value.

Understanding the benefits of an ecosystem for people requires an understanding of how people access the service. Therefore, there is a need to think about the different flows of ecosystem services – and break down the process of people to nature and nature to people. This process highlights just how complex the study of ecosystem services is – you have to consider the flows of people, the ways in which different socio-economic groups (rural dwelling vs urban-dwelling, age, wealth, gender) access and utilise nature and green space, as well as other provisioning services. There is also often movement of goods across supply chains to consider e.g. rural dwellers, whilst located closer to food production activities, often have to further to travel to access shops that sell the food.

Modelling Ecosystem flows

The spatial distribution of ecosystem flows can be predicted using different models. Different patterns of flows can be modelled and compared to real world observations, to help determine which theories are best to interpret how people use different services.

However - modelling the spatial distribution of ecosystem services flows is often easier said than done!

Data on the natural science systems that underpin ecosystems is plentiful, it’s relatively standardised, at high resolutions at vast spatial extents and at high temporal frequencies. Understanding what’s happening in these natural science systems is therefore much easier to do in comparison to socio-economic aspects. A substantial hinderance in understanding interactions of people and nature is social systems data is not collected in a comparable manner to the natural, and doesn’t exist to same extent across these scales. Social data is collected on more infrequent timescales, and whilst at best they attempt to be geographically representative, these aren’t as vast as the natural science data. As a discipline, ecosystem service science does not appear to have a good handle of the socio-economic drivers and social impacts of a natural event (e.g. in a flooding event we can model the environmental impacts, but the detail on how the event impacts socially is more difficult to model).

Therefore, high spatial and temporal resolution data across consistent areas is needed to meaningfully improve understanding of the services ecosystem provide us, and the capability to analyse these “big data” quickly and efficiently.

Using Smartphones for surveying ecosystem service use

Internet connectivity has increased drastically over the last 10 years in all parts of the world, and costs in data have fallen. Smartphones can be used to make surveying large numbers of participants across different time scales more feasible. There has been increasing use of the open-source ODK (Open data Kit) on smartphones to collect data on ecosystem service use, which has its advantages over traditional methods of survey collection. An ODK survey can include the usual survey questions, as well as images, videos and audio, as well as having multiple language capabilities, encryption of results, and almost instant data sets of results.

Using a “microtasks for micropayments” platform, the app notifies users when a survey is available, and respondents are paid for the survey with credits towards their phone bill – a big motivation for those in the global south. There is a large buy-in with the approach, and large retention rates, which allows for the collection of data at frequent timescales over long periods, to better coincide with the natural science data.

One example of the benefits this approach has is when looking at weather events, particular in relation to where water is collected from. Participants recorded, either daily, weekly, or monthly, where they sourced water from, and when there was a drought event, it was evident in the data how bad it would need to be before participants changed water source. Without the high frequency of data collection, that change wouldn’t be so evident.

This method has been used by Simon and his team in Bangledesh, India, Malawi, Cambodia, with new projects starting in Peru and Haiti. They have also conducted online surveys in Wales looking at cultural services of nature.

Convincing social scientists

Big data has a huge role to play in the understanding of social science data, but the capabilities to use data science techniques are somewhat lacking in ecosystem service science. Machine learning techniques expands traditional data analysis, offering more rapid processing of big data and enabling advancement in modelling of ecosystem services. Convincing social scientist to use Machine Learning techniques that could greatly benefit and research is a big challenge.

Research suggests that machine learning is under-utilised in ecosystem services studies, and that its methods aren’t currently implemented with high scientific rigour. Refinement of ML techniques is needed to address shortcomings in current research. As social data in ecosystem services increases as the field matures, techniques to manage larger data sets is imperative, and can aid the understanding of ecosystem flows. This understanding can lead to better, more sustainable, management of natural environments and improve equity of access to the services nature provides, better ensuring benefits for generations to come.

Related Blogs

• 

  Annual meeting of the Society for Decision Making under Deep Uncertainty, 2022

• 

  The Cryosphere 2022 Conference

• 

  DSNE Reseachers receive best poster award at CDE Conference 2022

Disclaimer

The opinions expressed by our bloggers and those providing comments are personal, and may not necessarily reflect the opinions of Lancaster University. Responsibility for the accuracy of any of the information contained within blog posts belongs to the blogger.