Framework for the Development of Environment Statistics (FDES 2013) 14 Emergency Events Database (CRED EM-DAT); classifications of protected areas and threat ened species developed by the United Nations Environment Programme’s World Conservation Monitoring Centre (UNEP-WCMC) and the International Union for Conservation of Nature and Natural Resources (IUCN); ecosystem reporting categories used by the Millennium Eco system Assessment; source categories for greenhouse gas (GHG) emissions from the Intergov ernmental Panel on Climate Change (IPCC); or the United Nations Framework Classification for Fossil Energy and Mineral Reserves and Resources (UNFC). Ensuring harmonization of these classifications and building bridges among them are among the most important roles of environmental statisticians. 1.43. For more information on classifications used in environment statistics, see Chapter 3 and Annex A, which contain the Basic Set of Environment Statistics. The Basic Set includes a column that lists commonly used classifications and categorization. Annex D contains relevant classifications and groupings in the field of environment statistics. 1.7. Temporal considerations 1.44. While it is important to align the temporal aggregations of environmental data with those used in economic and social statistics to ensure their proper integration, a uniform cal endar or fiscal year often does not correspond to the diversity of natural phenomena. There fore different time scales—or longer or shorter time periods—must also be used to aggregate environmental data over time. 1.45. The environmental data used in environment statistics are measured or monitored at various frequencies. Certain features of natural growth of biomass (e.g., in a natural, slow- growing forest that is not subject to logging) or processes such as changes in land cover or soil erosion do not justify or require frequent, diligent monitoring because the most relevant changes may be observed on an annual, or even much less frequent, basis. Other environmen tal processes, however, change so quickly that measurements are needed hourly or even more frequently. One example of frequent monitoring is air quality 15 in urban settings. 1.46. Determining the appropriate temporal aggregation of environment statistics often involves a variety of considerations. For example, fluid environmental phenomena call for care ful consideration of the temporal dimension because ebbs and flows, droughts and floods, snow and runoffs can occur, which all influence measurements. Variations may be daily and, at other times, seasonal depending on what is being measured. Seasonal variations may be seen in the fluctuations in certain types of fish biomass, surface water levels, ice cap surface or the inci dence of fires. In such cases, monitoring must focus more on certain months than others. Given these temporal aspects, statistics often point out the maximum, minimum and/or other ways of describing the relevant phenomenon and its levels below or above certain benchmarks and are not limited to a sum or average over a longer period. In addition, even when environmental data are produced at irregular intervals, environment statistics based on these data can still be produced at regular intervals if there are enough data points in each period to do so. 1.8. Spatial considerations 1.47. The occurrence and impacts of environmental phenomena are distributed spatially without regard for political-administrative boundaries. The most meaningful spatial units for environment statistics are: natural units, such as watersheds, ecosystems, eco-zones, landscape or land cover units; or management and planning units based on natural units, such as pro tected areas, coastal areas or river basin districts. 15 Air quality is measured by the concentrations of particulate matter (PM10, PM2.5), also known as suspended particulate matter (SPM), ground-level ozone (O3) or other pollutants specific to a particular city.
15 Overview of Environment Statistics—Characteristics and Challenges 15 1.48. Economic and social statistics are aggregated traditionally according to administra tive units. This difference can complicate the collection and analysis of environment statistics, particularly when they must be combined with data originating from social and economic statistics. However, there is a trend towards producing more georeferenced data, which would overcome some of the spatial complications of analysis. 1.49. While environment statistics are usually collected and aggregated for natural physical, geographical and administrative areas, the concept of economic territory is used for environmental-economic accounting. This involves a geographic boundary that defines the scope of an economy. Economic territory is the area under the effective control of a single government. It includes the land area of a country, including islands, airspace, territorial waters and territorial enclaves in the rest of the world. Economic territory excludes territorial enclaves of other countries and international organizations located in the reference country. 1.9. Geospatial information and environment statistics 1.50. Geospatial information presents the location and characteristics of different attributes of the atmosphere, surface and subsurface. It is used to describe, display and analyse data with discernible spatial aspects, such as land use, water resources and natural disasters. Geospatial information allows for the visual display of statistics in a map-based layout, which can make it easier for users to work with and understand the data. The ability to overlay multiple data sets using software, for instance on population, environmental quality and environmental health, allows for a deeper analysis of the relationship among these phenomena. 1.51. The complexity of current environmental issues (e.g., climate change, biodiversity loss, ecosystem health, natural disaster frequency and intensity, population growth and food and water shortages) increasingly calls for the integration of geospatial information, statistics and sectoral data to achieve more effective and efficient monitoring of progress in strengthening the environmental pillar of sustainable development. GIS can help establish the links between different types and layers of data by providing powerful tools to store and analyse spatial data and by integrating databases from different sectors in the same format and structure. 1.52. Geospatial information adds significant value and utility to environment statistics. Ide ally, geographic aspects of data should always be collected, represented and analysed at the most detailed scale possible, based on national capacities and priorities. Geospatial information enables better analysis of environmental issues as environmental, social and economic statistics can be aggregated or disaggregated according to a wide range of scales and zones that address diverse analytical and policy demands, such as natural units (e.g., watersheds and ecosystems); administrative units (e.g., municipalities, districts, counties and regions), management units (e.g., protected areas and river basin districts), planning units (e.g., coastal zones and urban areas); legal property units (e.g., cadastral units) and analytical units (e.g., land cover units, socioecological landscape units, eco-complexes, geosystems and eco-zones). 1.53. Geospatial data may be obtained using a variety of technologies such as Global Posi tioning Systems (GPS) and remote sensing satellites. Land surveyors, census takers, aerial photographers, police and even average citizens with a GPS-enabled cell phone can collect geospatial data using GPS or street addresses that can be entered into GIS. The attributes of the collected data, such as land-use information, demographics, landscape features or crime scene observations, can be entered manually or, in the case of a land survey map, digitized from a map format to a digital format by electronic scanning. The final representation of the data is constructed by superimposing different layers of information as required by the analytical and/or policy requirements.
Framework for the Development of Environment Statistics (FDES 2013) 16 Figure 1.1 Example of GIS data layers or themes16 1.54. Remote sensing gathers information about an object without coming into physical con tact with it. It involves the quantitative analysis of digital information where measurements can be taken from sensors on the ground, in aircraft or on orbiting satellites. The information is car ried by electromagnetic signals. Remote sensing calls for skills in digital image analysis when computer programming, image display tools and statistics are required for interdisciplinary work that may involve scientists and experts in fields including biology, climatology, geology, atmospheric science, chemistry and oceanography. Satellite remote sensing can address global issues by detecting, monitoring and measuring regional and global changes. 1.55. Remote sensing data from satellites are obtained digitally and communicated to cen tral facilities for processing and analysis in GIS. Digital satellite images, for example, can be analysed in GIS to produce land cover and land use maps. When geospatial data are combined in GIS (e.g., combining satellite remote sensing land use information with aerial photographic data on housing development growth), the data are transformed so that they are coincident and fit the same coordinates. GIS uses the processing power of a computer, together with geographic mapping techniques (cartography), to transform data from different sources onto one projec tion and one scale so that the data can be analysed and modelled together. 1.10. Institutional dimension of environment statistics 1.56. The institutional dimension of environment statistics refers to the institutional factors necessary to develop and strengthen the sustained production, dissemination and use of envi ronment statistics. It comprises the legal framework that establishes the mandates and roles of the main partners, the institutional setting and institutional development level of environ 16 Government Accountability Office (2004). “Geospatial Information: Better Coordination Needed to Identify and Reduce Duplicative Investments”, available from www.gao.gov /assets/250/243133.pdf (accessed 4 August 2017).
17 Overview of Environment Statistics—Characteristics and Challenges 17 ment statistics units, and the existence and effectiveness of inter-institutional cooperation and coordination mechanisms at the national level and with specialized international agencies. The institutional dimension of environment statistics is fundamental when developing environ ment statistics at the national level. Given the multidisciplinary and cross-cutting nature of environment statistics, the production of environmental data and statistics involves numerous stakeholders, actors and producers. The challenges of insufficient institutional development, overlapping mandates and functions, inadequate inter-agency coordination and other institu tional issues are very common in many countries. The problems of coordination and heteroge neous development can also escalate to the regional and global levels, where multiple partner agencies operate under different mandates, work programmes and production timetables. 1.57. Identifying the primary institutional obstacles that impede the production of environ ment statistics and developing a strategy to overcome them is essential for countries that seek to develop or strengthen their environment statistics programmes. The following are four key elements pertaining to the institutional dimension that should be considered and dealt with simultaneously while developing environment statistics. 1.58. The legal framework. In most countries, the legal framework for the production of envi ronment statistics commonly consists of statistical, environmental and other relevant sectoral legislation, such as for water, energy and agriculture. Each of these laws defines the mandate and competencies of the institutions in charge of the relevant sectors. 1.59. Under national statistical legislation, the NSO is usually the authority responsible for creating and coordinating the national statistical system. However, in most cases, these laws do not explicitly refer to environment statistics, as this is a relatively new statistical domain. More over, in many cases it neither provides explicit guidelines for statistical coordination among the relevant statistical parties at the national level nor spells out responsibilities and obligations. Nevertheless, since the environment is becoming increasingly important in the development agenda, NSOs have included the production of environment statistics in their programmes, though sometimes without clarifying the supporting institutional arrangements. 1.60. Overlapping mandates, duplication of efforts, and other coordination difficulties may exist in this complex institutional context. In fact, it is often difficult to determine the official figures for a specific statistic when different agencies produce the same or similar statistics but with different values. 1.61. Institutional development. A well-defined mandate and the designation of a specific unit responsible for producing environment statistics is critical for the successful organization of a national environment statistics programme within the official institutions that are respon sible for producing statistics. This unit requires a regular operations budget and a minimum number of trained personnel for the tasks involved. Environment statistics units thus need a capacity-building programme for staff, together with the financial resources to implement it. 1.62. Inter-institutional collaboration. Environment statistics cover several topics for which the data, whether in the form of administrative records, remote sensing, scientific measure ments or survey results, are generated by NSOs, specialized agencies, ministries, provincial and municipal governments and scientific institutions. This requires these stakeholders to col laborate, both at the strategic and technical level. 1.63. The collaboration of national and subnational institutions can take the form of a multi- stakeholder or inter-agency platform tasked with coordinating the strategic development and production of environment statistics. These inter-agency platforms bring together users and producers of environment statistics to identify users’ needs and ensure the coordinated produc tion of the necessary environment statistics from a variety of data sources. One of the tasks of the platform is to ensure that a common statistical methodology or protocol is used to ensure