Data capture, recycling and sharing: The multiscalar and multidiscipline challenge

By Natalie Small, Sustainable Places Research Associate

Over time I am finding myself listening, reading and getting involved with discussions on the same concerns surrounding data. It doesn’t matter at what scale, what location, what discipline, or even whether the focus is on industry, government or academia, it appears that everyone is facing the same issues that revolve around data capture, data sharing and data recycling.  I hear, “we don’t know what we have”, or “we’ve collected lots of data”, to “it’s not all in one place” or alternatively “we have not got the capacity or capability to interpret these data!” Not surprisingly, the most common comments made relate to data accessibility and how data are often restricted by licensing or cost.

Existing data could be considered priceless (especially for long term monitoring and historical trends), and they are difficult to locate/identify.  You need to know where they are, then get access to them. Once access is granted, it comes down to the skills and expertise of individuals or teams to overcome problems interpreting and combining the data. It is also important to develop mechanisms for information exchange and management.

To address the pressing issues relating to sustainability, we need to give attention to the issues surrounding access to quality-assured datasets, and their long term management and preservation.  These data needs are particularly important as they enable us to monitor ongoing natural and human induced changes to our environment.  Data are the key component of an evidence base which will support future strategies to deliver holistic and integrated resource management.

So how is it that although we recognise the need for an evidence base, to underpin future resource management, we still don’t know the true extent of what data we have, or how to combine them for re-use? We all need these data to be more accessible. The continuing efforts of the ‘Open Data Movement’ are a significant and powerful force endorsed by many. In 2010, Ordnance Survey launched OS Open Data. This service made a selection of mapping datasets of the UK freely available for use. Following this, other organisations across the UK began to do the same.  There have been advances on making public data more widely available through open government licences but the vast opportunities the ‘Open Data Movement’ could offer have not been fully realised.  Progress is being made, but we are still not there yet, why?

Will the recent unveiling of the Environment Secretary’s vision for open data to transform food, farming and the natural environment help us move forward, and speed up the process of collectively increasing the availability and accessibility of scientifically-based information?

Barriers, and challenges to future data collection, data sharing and recycling

As previously mentioned, there are a variety of challenges to face when collecting, recycling and interpreting data.  There are also a number of barriers which prevent organisations and individuals sharing their information to wider audiences.  Figure 1 illustrates some of the key questions raised at Scidatacon 2014 last year.  Similar questions are being addressed at the ‘Eye on the Earth’ summit this autumn.   We need to ask ourselves these questions if we are to increase the longevity of existing data, foster collaborations to increase data sharing and what to consider when capturing future data.

The UK is a data rich country but not surprisingly we are not fully aware of the plethora of data that already exists. Much of this existing data are archived on servers/stores, restricted by licensing or not published.  To overcome these challenges, awareness relating to existing data needs to be raised.  Coordinating and conducting ‘data audits’ within our work, departments, organisations etc., provide a clearer picture of what data exists.   Knowing what is already out there reduces the likelihood of unnecessary duplication of data.

When using existing data, considerations should be made on data suitability.  For instance, the final results of any analysis will depend on the quality of the data being fed into the analytical process.  Following data collection, all data should be subjected to a suitability and appropriateness assessment to address questions surrounding whether the data are fit for the purpose that it’s intended for.  Factors that should be considered include (but are not limited to); data quality, precision and accuracy of data, metadata quality, the age of the dataset and geographical variability of the data.

By knowing what we have also enables us to identify any data gaps we may have.  Associated with data gaps, are more questions on what are the most suitable data collection methods for our future data capture needs, especially at a time where financial budgets are being restricted.

Talking from an environmental perspective, when it comes to working out the most efficient and cost effective way of capturing new information on, for example habitat extent over the long term, would involve weighing up between, 1)  in-situ field data collection or 2) Earth observation techniques, and deciding what will provide you with the most useful dataset.  For long term monitoring, in-situ field studies are difficult to use for mapping national land-cover distribution or for vegetation dynamics.  This is because traditionally, in-situ data are collected at relatively small scales and vary in type and reliability.  Additionally, field dataoften comefrom a single time period in a year which make it difficult to gather information on temporal change.  To collect in-situ data at larger scales is time consuming and costly. So what is the alternative? For long term land-surface monitoring, recent research has shown that Earth observation techniques can assist current surveillance and monitoring requirements. Earth observation can provide a verifiable means of collecting environmental information with complete spatial coverage at a variety of temporal resolutions. Imagery from satellite sensors is relatively inexpensive and programmes like the Landsat Program have collected information about our land surfaces for decades and is freely available. Excitingly, the Copernicus programme collects information from both a network of in-situ sensors and satellites. This opens up opportunities for future studies where data from both in-situ and satellite sensors could be combined to answer gaps in our knowledge on land-surface processes.  There is already a lot of excitement surrounding the data that are currently being collected on land surface by the recently launched Sentinel 2 satellite, which will provide high resolution optical imagery for land services.

Alternatively, it may be more appropriate and cost effective to use existing data in future studies.  Whereby these studies utilise methods such as geographical information systems (GIS) or statistical models to manipulate and derive new data sets. Utilising GIS helps to facilitate the analytical process by being able to link up all contextual information at a range of spatial and temporal scales. It provides the information and analytical capabilities needed for making place based decisions. Spatial analysis helps to visualise, question, analyse, interpret and communicate complex data, giving a clearer understanding of any potential relationships, patterns and trends.

Why are data important for policy needs?

The forthcoming Environment Bill (Wales) plans to put in place the legislation that is needed to manage Wales’s natural resources in a sustainable, proactive and integrated way.   Fundamental to the success of an integrated approach will be the unification of the evidence base.  By taking a holistic approach to evidence gathering will ensure that the supporting evidence base is not just restricted to environmental data, but also, cultural, social and economic data. The data and information we currently have, and will continue to gather, is helping us to understand our environments, how they are changing, and the role being played by human activities in driving these changes.  To be able to make the right decisions for the future, and to coordinate natural resource management, requires access to reliable and up-to-date information on how human and natural environments are evolving and interacting with each other.

Data challenges: it is on the global agenda

The topic of data capture, recycling and sharing is not just a local hot topic of discussion, it is also happening at the global level Last year, the international conference on data sharing and integration for global sustainability, ‘SciDataCon2014’³ took place in New Delhi, India.  The forthcoming, ‘Eye on the Earth’ Summit takes place in autumn 2015 in Abu Dhabi. Both are synergistic in their view that pressing issues relating to sustainability and resource management cannot be properly addressed without attention being given to the issues relating to data⁹.  There is an urgent need to bridge the information gap that policy makers face in designing plans for sustainable development.

I look forward to the outputs from the ‘Eye on the Earth’ summit and hopefully to see what answers the community come up. Engagement with a large community of interdisciplinary stakeholder organisations is pivotal if the summit is to be successful in reaching its goal of pushing forward the common agenda on collectively increasing the availability and accessibility of scientifically-based information.

By Natalie Small, Sustainable Places Research Associate

@Orcanat