The role of combining national official statistics with global monitoring to close the data gaps in the environmental SDGs

1.Introduction

The 2030 Agenda for Sustainable Development (2030 Agenda) provides a holistic framework for development which aims to transcend the pursuit of siloed interventions related to social development, economic development and environmental protection and resource use [8]. The 2030 Agenda includes 17 Sustainable Development Goals (SDGs), 169 targets and 244 indicators which form the basis for monitoring and implementing the 2030 Agenda [9]. At the core of the 2030 Agenda is an attempt to identify the most pressing issues for global development while keeping within planetary boundaries [10]: How can natural resources be sustainably managed while at the same time providing food, energy and water for the growing global population? What are the underlying governance and partnership requirements for sustainable development? What is the interaction between human health and the natural environment? How can we protect biodiversity and terrestrial and marine environments while still achieving economic growth? In order for the SDG framework to be useful for answering these key questions, data, analysis and science for each of the 244 SDG core development indicators are required. Unfortunately, there is a dearth of information for understanding the environment. An assessment of the global environmental goals during the 5th Global Environment Outlook process, which included a review of more than 320 environmental goals which are included in legally and non-legally binding multilateral environmental agreements (MEAs), revealed that data availability, comparability and fragmentation impede the ability of countries to report and fulfil their obligations [11]. The data gaps are even more pronounced when trying to understand specific local contexts or the interactions between gender, poverty and vulnerable populations and the environment [4, 12, 13].

The use of comparable data, which follow harmonized definitions and standards, for financial and economic decision making has existed in every civilization that has a system of writing and trade – as early as 3300 BC in Egypt [14]. By comparison, the work to measure, account for and value the environment is recent in comparison with the first attempt of a global environment statistics framework, the Framework for the Development of Environment Statistics published in 1984, the first environmental accounting framework, the System of Environmental Economic Accounting published in 2012, and the initial establishment of the Group on Earth Observation in 2005 [15, 16]. Despite efforts to work toward comparable definitions, methodologies and standards for environmental data, geospatial data and statistics, there are still gaps in the existing frameworks and much of the environment-related data products which have been developed with a specific purpose in mind are not comparable. In comparison to economic and social statistics, monitoring of the environment is a recent development and is the weakest area of monitoring in the SDG framework based on the number of indicators with available data by type.

2.Methodology

This analysis is based on the 244 SDG indicators which have been adopted for global monitoring by the UN Statistical Commission [9]. This indicator list includes some duplication of indicators which are listed under multiple targets and thus there are 232 unique indicators; for the purpose of this analysis, all 244 indicators were included as this provides a method for understanding the availability of data for assessing each of the 169 targets and 17 goals.

The SDG Tier Classification was developed by the IAEG-SDGs and endorsed by the UN Statistical Commission and the Tiers are defined as: “Tier I: Indicator is conceptually clear, has an internationally established methodology and standards are available, and data are regularly produced by countries for at least 50 percent of countries and of the population in every region where the indicator is relevant. Tier II: Indicator is conceptually clear, has an internationally established methodology and standards are available, but data are not regularly produced by countries. Tier III: No internationally established methodology or standards are yet available for the indicator, but methodology/standards are being (or will be) developed or tested” [17]. As of March 2020, all indicators have now been reclassified to Tier I or II; however, the analysis presented by Tier change between 2016 and 2019 aims to show the progression of the SDG indicator methodological development.

There is no agreed definition of which indicators and targets should be included in the environmental dimension of sustainable development. Different definitions have been used in different analysis, with some analysis of which goals are strictly related to the environment; for example, focusing on the SDG indicators with a direct link to the physical environment, focusing on climate change (SDG 13), water (SDG 6), land (SDG 15) and oceans (SDG 14) [18, 19] or the physical environment plus the goals related to the sustainable use of natural resources, thus adding energy (SDG 7) and sustainable consumption and production (SDG 12) [20]. The analysis in this paper is based on a classification of SDG indicators included as a background document at the 20 September 2018 meeting of the UN Environment Assembly Committee of Permanent Representative subcommittee [21] and the list used by UN Environment in the Global Environment Outlook process [1, 22] (included in Annex 1). The list is subjective and should not be interpreted as a comprehensive list of all SDG indicators which interact with the environment, but as an initial attempt to identify the SDG indicators which are directly related to environmental policy, environmental state or trends, sustainable consumption and production and natural resource use or interactions between people and the environment.

The list of environment-related SDG indicators was then categorized into four indicator types: indicators related to (1) mechanisms, enabling environment or policy; (2) change in behaviour or consumption or production patterns; (3) environmental state and trends; and (4) linkages between people and the environment (access to natural resources to meet basic needs or for livelihoods, vulnerability to climate change and disasters, environmental mortality, etc.). This classification done by the authors is shown in Annex 1.

To assess data availability, UN Environment maintains a database of indicators related to the environment which includes more than 1000 indicators [23]. These indicators cover all 93 indicators which are included in UN Environment’s list of environment-related SDG indicators. Global and regional aggregations are made available in the UN Environment database when possible. When global and regional aggregations are not available, UN Environment uses an algorithm to aggregate data to the regional and global level [24]. Note that the aggregation procedure is only completed if at least 30% of the total population, area, GDP or countries (depending on the weight of the indicator) have a data point for the year being aggregated and 70% must have either a valid data point or a data point within 5 years of the year being aggregated which can be used for extrapolation or interpolation. This is the methodology that UN Environment uses for all SDG indicators [24].

For each country, region and at the global level, a determination on sufficient or insufficient data was made based on the inability to attempt to extrapolate data based on a regression model. Thus if there were no available data points or, in the case of regional and global aggregates, if there was not sufficient data for the aggregation procedure, between 2000 – 2018 or the most recent data point was prior to 2010 then the indicator was classified as ‘no data available’. If an indicator has data after 2010, but only has a single data point, then this represents insufficient data for assessing progress and the indicator was classified as ‘some data, but not sufficient data available’. This assessment was based on data available as of June 2019. For the analysis by income group, the World Bank classification of income groups for 2018 was used [25].

Country-level analysis was done for the 193 United Nations Member States based on data available in the Global SDG Indicators Database as of June 2019. For each indicator, data was deemed available for a country if it had any observations. Countries were then aggregated into four groups: OECD, non-OECD, Small Island Developing States (SIDS), Least Developed Countries (LDC), and Landlocked Developing Countries (LLDC) using UNSD definitions [26]. The reporting coverage (percentage) for each group is the average of the percentage of indicators for which there is data for each country.

3.Result

The IAEG-SDGs was established in 2015 to develop the SDG indicator framework [27]. In order to do so, the IAEG-SDGs conducted an open consultation involving a wide range of stakeholders from across government, civil society, academia and regional and international organizations [28]. From a conceptual perspective, the SDG indicators were developed in order to capture an ideal, ambitious monitoring framework for development, including environmental state, trends and impacts. However, from a practical perspective, the SDG indicator framework took into account an interest in increasing synergies with existing processes (e.g. the Millennium Development Goals, UN Framework Convention on Climate Change, the Sendai Framework on Disaster Risk Reduction) in order to reduce the reporting burden for countries, increase feasibility, capitalize on prior experiences of national statistical offices in collecting official statistics on a topic and use a well-developed plan for developing and rolling out indicator proposals [29]. Despite the interest in using existing indicators, when the original tier classification was developed in 2016, 50% of environment-related SDG indicators were classified as Tier III as compared to 28% of the remaining indicators. There has been significant progress to develop methodologies and reporting mechanisms for the SDGs and as of May 2019, 26% of the environment-related SDG indicators are Tier III compared to 7% of the other indicators (Table 1).

In terms of the environmental dimension of the SDGs, holistic measurement of the environment is complicated by a lack of existing globally-agreed methodologies related to specific SDG targets and the fact that many statistical offices do not have experience in compiling environment statistics or environmental economic accounts [30]. The System of Environmental Economic Accounts provides a useful conceptual framework which can underpis more 40 of the environment-related SDGs [17, 31]. Out of the 244 indicators in the SDG framework, less that 5% (11 indicators) are related to environmental state and trends (Fig. 1).

Figure 1.

Total number of environmental Indicators by type and data availability, 2019.

3.1Country level data availability

Country level reporting depends on various factors like the capacity of national statistical offices, data availability and political interest. Some indicators are reported independent of national reporting capacity based on international estimates and models (proxy data) such as SDG8.4.1 on material footprint and SDG6.6.1 on water-related ecosystems. Others rely more heavily on national capacity and interest (national data) such as SDG12.1.1 on sustainable consumption and production action plans and SDG11.6.1 on urban solid waste management. Analyzing country-level data availability can help identify gaps in national capacity or highlight areas where political interest may be lacking.

Overall, there is about a 40% country-level data coverage for the 93 environment-related indicators (including Tier III indicators) (Fig. 2). There is more difference in reporting across the four indicator types than the five country categories (OECD, non-OECD, SIDS, LDC and LLDC). Indicators in the “Environmental State or Trend” category have the highest country-level coverage (about 59%). Indicators in the “Mechanisms, enabling environment and policy” category have the lowest data coverage (about 32%). This is consistent with the proportion of Tier III indicators in each of these indicator types (Fig. 1).

While the biggest differences in country-level data availability are based on the indicator types, there are also some differences in reporting level based on country categories. OECD countries (all of which are high-income countries except for Mexico and Turkey) have higher levels of data availability than non-OECD countries for all indicator categories except for “Mechanisms, enabling environment and policy” indicators. On the other hand, SIDS have the lowest levels of country-level data reporting across all indicators. The biggest gap being for indicators in the “Behavior and Sustainable Consumption and Production” category where country-level reporting in SIDS countries (41%) is 4% lower than the global average (45%) and 7% lower than OECD countries (48%) (Fig. 2). SIDS small size, remoteness, narrow resource and exposure to global environmental change may be factor to their low levels of country-level reporting [2].

Figure 2.

Data availability by indicator category and country group, 2019.

4.Discussion and conclusion

Faced with the SDGs’ 169 targets and 244 indicators – a number of which still lack clear measurement methodologies – countries are stymied in both action and monitoring of SDG progress. This is no more true than in regards to environmental indicators which are outpaced among all indicators in globally-agreed methodologies and, regardless of a country’s stated priorities, will impact progress on other targets [34, 35, 36].

In order for the SDGs to be an effective tool for galvanizing action, there is a need to urgently uplevel monitoring and build a digital ecosystem for the planet [37]. The idea behind a digital ecosystem is to improve accessibility and access to traditional statistics along with data from citizen science, remote sensing, transactional data and other forms of data in a decentralized data ecosystem. This would include improving the accessibility and interoperability of raw data and data products on the environment while at the same time ensuring that privacy and security. This would provide a foundation to fill data gaps for targets lacking data disaggregated by crucial specificities and disparities (including by gender, income quintile, education, disability and other vulnerable groups) and even lacking data at all. Such an ecosystem would also help build the tools and capacity to conduct integrated analysis that can be used to generate actionable insights [1, 38]. Global data products and global modelling can form a basis for filling data gaps and ensuring that there is some data for all countries. These products also can demonstrate environmental concerns that should be further monitored through high-frequency or high-resolution remote sensing or through in situ collection or other means. Existing global data products like those for SDG target 6.6 (https://www.sdg661.app/) and SDG targets 8.4 and 12.2 (material flows) can provide an example of how such products can and have already been accepted for official SDG monitoring [16, 39].

Complementing global products and global modelling is citizen science, which can also help fill data gaps in some countries. Citizen science is already contributing to SDG indicators, e.g. 15.5.1, but further efforts are needed to investigate the potential of citizen science for other environmental SDG indicators, particularly Tier III. Citizen science projects and representatives from the citizen science community need to be brought into the high-level discussions on methodology development and data collection to explore the best way forward. Making progress on one indicator, e.g., 14.1.1. on marine plastic debris, may be provide a blueprint for integrating citizen science in other places.

In economic policy, a single headline indicator – Gross Domestic Policy – is used as a flag for economic progress. There has been some discussion on developing an equivalent indicator, with corresponding sub-indicators, for measuring sustainable development [40, 41]. Building out the portfolio of country-level data on globally-agreed environmental indicators could support consensus on a headline indicator on the environmental dimension of development. However, building the capacity of countries to collect the underlying basic data and better utilizing existing data – including from non-traditional data sources such as satellites, drones and citizen science – are at the crux of being able to monitor the SDGs and develop integrated analyses or indices [42].

The current limitations on understanding environmental indicator progress illustrate a “catch-22 situation” when it comes to achieving the SDGs for the sustainability of the planet and for humanity: We use existing data to identify priorities, but priorities for data collection are identified on the basis of which topics are priorities. It is the role of the international statistical community to disrupt this dynamic, leverage the commitment to monitoring the SDGs to better monitor across all sustainable development issues and to ensure that development occurs within planetary boundaries.