February 2001

LCCS: Land cover classification system

John Latham

Officer, Environment and Natural Resources Service

Despite the high demand of natural resources information, many existing maps/digital databases are not developed to really meet multi-user requirements. One of the basic causes (generally underestimated) of such situation is the type of classification/legend used to describe basic information such as land cover and/or land use: many of these classifications/legends are generally not comparable one to another and very often single project oriented. Although there are many classification systems in existence throughout the world, there is no single internationally accepted land cover classification system.

Analysis of the physical and human resources information in many countries has revealed the following difficulties:

• basic information on environment and natural resources are often limited in extent;
• the quality of existing information is extremely variable and the way it is presented is often not optimal to support decision-making;
• topographical maps and thematic maps, when available, are heterogeneous;
• the information provided by the various institutions are often duplicated or inconsistent within the same country;
• it is extremely difficult to integrate or compare the existing information made available by departments and organizations.

Decision-makers at local, national, sub-regional, public or non-governmental levels are now realizing the necessity to have access to reliable geo-referenced information systems on natural resources. Technically, the production and the rational and harmonized use of these information systems should necessarily go through the establishment of a geo-referenced and standardized thematic base which meets the common needs of natural resources information producers and users.

The FAO/UNEP Land Cover Classification and its software program (developed and finalized) is a comprehensive standardized a-priori classification system designed to be able to meet specific use requirements, and created for mapping exercises, independent of the scale or means used to map; any identified land cover anywhere in the world can be readily accommodated. The proposed classification uses a set of well-defined independent diagnostic criteria that allow correlation with existing classifications and/or legends. Therefore, this system could serve as a basis for a reference system.

Historical background

The AFRICOVER initiative on the definition of a Land Cover Classification was initiated during the expert consultation held in Addis Ababa, in July 1994. The Working Group on Classification and Legend had the task to define a standardized classification, which could be used for mapping land cover in all African countries. This classification had to meet the following requirements:

• imagery to aerial photography;
• oriented to the preparation of multi-user databases;
• compatible with existing classifications/legends in Africa;
• practical and adapted to existing African capacities; and
• liaise with ongoing initiatives on the classification and definition of land cover and land use (e.g. IGBP, FAO/UNEP/ITE/ITC/WCMC, EU Corine program, etc.).

Following this expert consultation, a user definition study has been held in 27 African countries to evaluate the information requirements and the priority classes to be mapped within the project. The latter survey clearly indicated the importance of cultivated cover types. This survey also included an extensive view of existing maps and databases on land cover and land use existing in Africa. Several papers and preliminary reports (Negre, Barisano) were prepared, analysing the existing classification/legends; drafts and nomenclatures were submitted and proposals made for the standardized classification.

At a meeting in Rome in 1995, the approaches of AFRICOVER and FAO/UNEP/ITE/ITC/WCMC, which were developed in parallel, eventually merged into one common approach.

Objectives/potential utility

The main objective of the initiative for definition of a reference classification is to respond to the need for standardization (or harmonized collection of data, as mentioned in UNCED's Agenda 21 Chapter 10, for which FAO is Task Manager within the UN system) and to develop a common integrated approach to all aspects of land cover. This implies a methodology that is applicable at any scale, and which is comprehensive in the sense that any land cover identified anywhere in the world can be readily accommodated.

The approach developed for land cover could serve as the basis for a reference classification system with links to specific expertise, because it describes and allows correlation of land cover with a set of independent diagnostic criteria, the so-called classifiers, rather than being nomenclature based. Also, existing classifications and legends can be "translated" into the reference system, thus facilitating the use of existing historical materials. Re-arrangement of the classes, based on re-grouping of the used classifiers, facilitates the extensive use of the outputs by a wise variety of end-users.

The main objective is the development of a unified methodological approach and relevant tools for land cover classification and mapping, that could serve as a unified land cover reference basis. More in details, the aim is to:

• respond to the need for land cover data of a variety of end-users;
• apply the methodology in mapping exercises, independent of the means used, which may range from high resolution satellite imagery to aerial photography;
• link with existing classifications and legends, allowing comparison and correlation;
• support international ongoing initiatives on classification and definition of land cover; and
• harmonize principles and methodology for land cover mapping.

This international standard will be applied to other modules of AFRICOVER at regional and national levels in Africa, and are already in use in several other country programs outside of Africa, e.g. Afghanistan, Azerbaijan, Bulgaria, Lebanon and Yemen, and are being considered for inclusion in the forthcoming Asiacover project.

The utilization of this new worldwide common language of land cover between different countries and different type of final users, in support to the development of a reliable environmental baseline information source, will lead to a basic source of detailed and homogeneous environmental information as baseline for more complex systems. The resulting land cover data set will be able to supply a large amount of detailed, homogeneous and standardized information for a wide range of applications. The homogeneity of the data will allow an immediate, scientifically sound and economic replicability of each application at a national, regional and international level.

Technical specifications

Land cover classes are defined by the combination of a set of independent diagnostic criteria, the so-called classifiers, which are hierarchically arranged to assure a high degree of geographical accuracy (also referred to as "mapability"). Because of the heterogeneity of land cover, the same set of classifiers cannot be used to define all land cover types. The hierarchical structure may also differ from one land cover type to another. Therefore, the classification is designed according to two main phases:

1. an upper dichotomous phase where eight major land cover types are distinguished;
2. a lower modular-hierarchical phase where the set of classifiers and their hierarchically arrangement are tailored to the major land cover type.

This allows the use of the most appropriate classifiers and reduces the total number of impractical combinations of classifiers. Because of the complexity of the classification and the need for standardization, a software program of which the beta version has been developed, assists the interpretation process: it reduces heterogeneity between interpreters, and with interpretations over time. Because of the flexible manner in which the classification is set up (creation of classes at different levels of the system and the optional use of modifiers and/or environmental and/or specific technical attributes) and the tremendous number of classes possible, this innovative software program assists the user by selecting the right class going stepwise classifier by classifier. This software will be integrated into a digital image interpretation software which will allow interpretation of imagery followed by labeling of the mapping units with the land cover classes. The classification system provides a mutually exclusive Land Cover Class which comprises: (1) a unique Boolean formula (a coded string of classifiers used); (2) name/nomenclature; and (3) a unique numerical code. Both the numerical code and nomenclature can be used to build an automatically generated Legend with the created classes grouped according to the main land cover categories and their level of detail. The nomenclature can be linked to a user-defined name.

Further definition of the Land Cover Class can be achieved by adding attributes. Two types of attributes, which form separate levels in the classification, are distinguished:

1. environmental attributes: these attributes (e.g. climate, landform, geology) influence land cover but are not inherent features of it and should not be mixed with "pure" classifiers.
2. specific technical attributes: these attributes refer to the technical discipline. As an example, for (Semi-)Natural Vegetation floristic can be added (the method how this information was compiled and a list of occurring species), for Cultivated Areas the crop types can be added according to broad categories commonly used in statistics or at the detailed level of species.

Advantages

In order to homogenize and take into account the different nomenclature/legend existing in the different Countries, AFRICOVER E.-Africa will be the first project to use the FAO Land Cover Classification system LCCS and the related software.

From a conceptual point of view, the advantages of the proposed classification are:

1. A real classification system in the sense that it covers all possible combination of classifiers. Some combinations are excluded due to some conditions, which are elements of the classification system.
2. A given land cover class is clearly defined by a set of independent classifiers. The classifiers are clearly differentiated in: pure land cover classifiers, environmental and other classifiers and discipline specific classifiers. This avoids an unclear mixture.
3. The classification is truly hierarchical. The difference between a land cover class and a further sub-division of this class is given through the addition of new classifiers. The more classifiers used, the greater the detail of the defined land cover class.
4. The classification can be used as reference classification system. In fact, the emphasis given to the set of classifiers defining the class allows easy correlation between existing classification/legend and the proposed one.
5. The specific design of the classification allows incorporation into GIS and databases. The pure land cover classes can be used in overlay procedures to make combinations with e.g., climate and physiography, to create new classes.

From a practical point of view:

1. The classification is designed for mapping. The hierarchical arrangement of classifiers is set up to assure a high level and precise mapping accuracy (clear definition of boundary between two land cover classes).
2. It facilitates the integration of different types of data.
3. It rationalizes the field data collection. As the classes are defined by a combination of classifiers, field surveyors should detect the single classifiers and not deal with the final class name. This means that the field survey can be done independent of, or parallel to, the interpretation process.
4. It is highly flexible, responding to the information available or gathered in a given area or for the time and budgetary constraints of a project. This means that within one land cover map, the mapping units will contain the maximum available information but this quantity of information may differ between mapping units. This will not affect the homogeneity of the resulting map.
5. It facilitates the standardization of the interpretation process contributing to its homogeneity. In fact, the interpreter is not dealing with a final class name but is dealing with one classifier at the time. This reduces heterogeneity between interpreters and with interpretations over time.
6. It is multi-user oriented. Because the class is defined by a set of classifiers, every user can make a re-selection based upon the classifier(s) of interest.
7. It is designed to map at a variety of scale, from small-to large-scale.

Validation and distribution

This product, developed as part of the normative process, will be the subject of peer review by a panel of internationally recognized high-level experts, in 2001, in order to promote its worldwide application.

The organizations involved in the validation process of LCCS are:

• GLC 2000 (Global Land Cover mapping) - Joint Research Center (JRC) ISPRA
• UNEP - United Nations Environment Program
• FGDC - Federal Geographic Data Committee
• Global Land Use Cover Classification (GLUCC) working group of IGBP (International Geosphere/Biosphere Program)