Nicola Arndt und Matthias Pohl - Neobiota
Nicola Arndt und Matthias Pohl - Neobiota Nicola Arndt und Matthias Pohl - Neobiota
Russia. Information on rural land use stems from KOSTROWICKI (1984) and CSATI (1980) for Russia. The various sources have been manually overlaid, visually compared and analysed to determine the final typology and delineation for the landscape map by Meeus. The overlay comparison between the red contour lines of the Map of European Landscapes and the Map of European Natural Vegetation (Map 6) shows that the main resemblance can be found between the vegetation units estuaries and riverine forests along the Waddensea coastline and what Meeus has identified as the landscape type ‘polder landscapes’. There is some overlap between the natural vegetation unit of species poor oak and mixed oak forests in the NW-European lowlands and the socalled ‘Kampen’-landscapes, a term that originates from Flandern and refers to undulating plains with brooks and sandy soils. The hemiboreal zone of southern Sweden appears to correspond with the landscape type ‘southern taiga’, though its shape and geographic location is hard to match. Difficult is the interpretation of the landscape type ‘central collective open fields’ in the region southwest/around? of Prague that is actually covered with substantial stands of mixed forest. It appears that there the resemblance between the two maps is highest for azonal vegetation units such as coastal zones and riparian corridors. 4.3 Overlay with the European Landscape Typology and Map (LANMAP 1) Given the increasing demand for high-accuracy landscape information at the European level (WASCHER 2003), and the observation, that existing approaches fall short of using state-of-the-art technology and addressing cultural attributes (e.g. land cultivation patterns, historical features, landscape elements, land use characteristics), there is a clear need to establish a classification and map for Landscape Character Types at the European level as a main point of reference in support of both research and policy implementation at the European and national level. The strategic objectives are as follows: • Establish a European-wide neutral and culturally unbiased typology of landscape types that is based on high-quality data of European coverage and which can be linked to existing national approaches while linking up with the European bio-climatic regions; • Make sure that the proposed landscape types provide a meaningful reference base for policy application, e.g. the European Landscape Convention (Council of Europe), Agenda 2000 (rural development), reporting according to the DPSIR framework (Driving Force - Pressure - State - Response); ESPON spatial planning, etc. A European landscape mapping project should provide a practical and easy tool for European policy implementation. Important applications are integrated environmental assessment, monitoring and reporting, especially indicator-based approaches. After formulating user requirements and possible target groups (see above), a critical review of the main European environmental data sets has been undertaken in order to select the following suitable core data sources for the delineation of the major landscape units to develop the first draft version of a European Landscape Typology and Map (LANMAP 1): 98 • Topography (GTOPO30, grid data, 1km resolution) • Parent material/ Ecological site conditions (ESDB 1:1M, vector data) • Land use / Land cover (CORINE land cover database, vector data, 1:100 000)
The choice of data sets reflects that landscapes are a product of natural and cultural driving forces. Since a reliable European map on geo-morphological aspects was not available, information on topography and parent material has been chosen as the adequate substitute. These three core data sets determine the matrix for a European Landscape Map. Specific landscapes, such as wetlands or bocages will be delineated within this matrix on basis of additional data sources. For the segmentation of the major landscapes the software package eCognition has been used. ECognition is an object- oriented image classification software for multi-scale analysis of Earth Observation data of all kinds. The image classification is based on attributes of image objects (semantic information) rather than on the attributes of individual pixels. Before the segmentation could take place, the legend units of the three core data layers (topography, parent material and land cover) have first been reduced to represent only key thematic classes. This was necessary in order to keep the data management processes feasible and to arrive at results that reflect a European aggregation level. For the LANMAP 1 typology construction remain 5 altitude classes, 13 parent material classes and 8 land use classes. Please note that this aggregation has been undertaken purely for facilitating the identification of broad and coherent landscape units, but that the underlying information (the complete legend details for each layer) are nevertheless fully available for further analysis of these units. For the urban, marine and freshwater landscapes the information was directly derived from the land use layer. (This was also necessary because for these landscape types there were data gaps in the soil database). So in principle there are (5×13×8)+3 = 523 combinations, however in reality there are 202 existing combinations, read landscape types, represented in LANMAP 1. The LANMAP 1 typology consists of 202 landscape types featuring a 3 digit code: the first capital letter is used for the topographic class, the second capital letter for the parent material and the third letter (lower case) for the land cover class. This is also illustrated in Figure 3. As an extra attribute the environmental zone (e.g. Alpine south, Nemoral, Pannonian) has been attached to each landscape mapping unit. The Environmental Zones (13 zones in total) have not been used in the typology, but will be used in the description of the landscape type. For the urban landscapes the information was derived from the CORINE land cover database. However, some extra processing was done to derive only the larger urban agglomerations. For this purpose a 5 km by 5 km majority filter was used in ERDAS Imagine. This map was integrated within the landscape map. After this there were additional post-processing steps necessary to upgrade the European Landscape Map, being summarised below. A large advantage of the European Landscape Classification is that its selection of boundaries is consistent, crisp and transparent based on the underlying layers: topography, parent material and land cover. However, if misclassifications do occur in one of the three underlying layers this is reflected in the European Landscape Classification. The fact that the European Landscape Classification lacks information on the land use history is a limiting factor but was so far difficult to collect at the European scale Maps 8/9 present a side-by-side comparison between LANMAP 1 with the same area selection of the European Natural Vegetation Map. 99
- Seite 49 und 50: 0 200 400 600 0 500 1000 1500 2000
- Seite 51 und 52: Abb. 5: Karte der landschaftsökolo
- Seite 53 und 54: Die Analyse der geostatistischen Re
- Seite 55 und 56: DINTER, W. (1999): Naturräumliche
- Seite 57 und 58: Application and Analysis of the Map
- Seite 59 und 60: 2 FAO Requirements Many environment
- Seite 61 und 62: In practical terms, delineation of
- Seite 63 und 64: 3.3 FAO Global Ecological Zone clas
- Seite 65 und 66: Table 2: LUT for Europe, showing th
- Seite 67 und 68: forests (F), 7 subgroups (F1- F7) h
- Seite 69 und 70: Figure 2: Map of Global Ecological
- Seite 71: Annex Table 4: Source maps used for
- Seite 74 und 75: DMEER-Projekt (Digitale Karte der
- Seite 76 und 77: Figure 1: The ecoregions are catego
- Seite 78 und 79: An example of the relationship betw
- Seite 80 und 81: DASMANN, R.F. (1973): A system for
- Seite 82 und 83: THACKWAY, R. & CRESSWELL, I.D. (eds
- Seite 84 und 85: 1 Background 1.1 Increasing interes
- Seite 86 und 87: While regional and national activit
- Seite 88 und 89: forest systems are supposed to reac
- Seite 90 und 91: According to SCHLÜTER (1991), hete
- Seite 92 und 93: derived from an analysis of soils a
- Seite 94 und 95: Map 1: Landscape Character Areas of
- Seite 96 und 97: in terms of successions and replace
- Seite 98 und 99: Map 5: Landscape map of Europe by M
- Seite 102 und 103: 100 Map 7: European Landscape Typol
- Seite 104 und 105: G.3.1/37 F.1.1/7 & F 1 2/15 F.1.1/8
- Seite 106 und 107: Map 10: Comparison of the vegetatio
- Seite 108 und 109: 106 Map 11: Landscape Typology and
- Seite 110 und 111: � Natural vegetation data will al
- Seite 112 und 113: LUC (1999): Glasgow and the Clyde V
- Seite 115 und 116: Application and Analysis of the Map
- Seite 117 und 118: characterized by zonal vegetation i
- Seite 119 und 120: Table 1: FAO Global Ecological Zoni
- Seite 121 und 122: The name of each type includes the
- Seite 123: References LAVRENKO E.M. (1964): Al
- Seite 126 und 127: 1 Introduction A dominant feature i
- Seite 128 und 129: 10° 65° 70° SBZ 75° MBZ 42 20°
- Seite 130 und 131: As shown in Table 2 there is a good
- Seite 132 und 133: What is interesting is that the tem
- Seite 134 und 135: the driest (like Ulmus and Corylus
- Seite 137: Gliederung und Inhalte einzelner Fo
- Seite 140 und 141: 1 Gliederung der borealen Wälder D
- Seite 142 und 143: Die Assoziation Empetro nigri-Pinet
- Seite 144 und 145: WOJTERSKI, T. (1964): Bory sosnowe
- Seite 146 und 147: Attempts to characterize and delimi
- Seite 148 und 149: The schematic map presented here ca
The choice of data sets reflects that landscapes are a product of natural and cultural driving forces.<br />
Since a reliable European map on geo-morphological aspects was not available, information on<br />
topography and parent material has been chosen as the adequate substitute. These three core data sets<br />
determine the matrix for a European Landscape Map. Specific landscapes, such as wetlands or<br />
bocages will be delineated within this matrix on basis of additional data sources. For the segmentation<br />
of the major landscapes the software package eCognition has been used. ECognition is an object-<br />
oriented image classification software for multi-scale analysis of Earth Observation data of all kinds.<br />
The image classification is based on attributes of image objects (semantic information) rather than on<br />
the attributes of individual pixels.<br />
Before the segmentation could take place, the legend units of the three core data layers (topography,<br />
parent material and land cover) have first been reduced to represent only key thematic classes. This<br />
was necessary in order to keep the data management processes feasible and to arrive at results that<br />
reflect a European aggregation level. For the LANMAP 1 typology construction remain 5 altitude<br />
classes, 13 parent material classes and 8 land use classes. Please note that this aggregation has been<br />
<strong>und</strong>ertaken purely for facilitating the identification of broad and coherent landscape units, but that the<br />
<strong>und</strong>erlying information (the complete legend details for each layer) are nevertheless fully available for<br />
further analysis of these units. For the urban, marine and freshwater landscapes the information was<br />
directly derived from the land use layer. (This was also necessary because for these landscape types<br />
there were data gaps in the soil database). So in principle there are (5×13×8)+3 = 523 combinations,<br />
however in reality there are 202 existing combinations, read landscape types, represented in<br />
LANMAP 1.<br />
The LANMAP 1 typology consists of 202 landscape types featuring a 3 digit code: the first capital<br />
letter is used for the topographic class, the second capital letter for the parent material and the third<br />
letter (lower case) for the land cover class. This is also illustrated in Figure 3.<br />
As an extra attribute the environmental zone (e.g. Alpine south, Nemoral, Pannonian) has been<br />
attached to each landscape mapping unit. The Environmental Zones (13 zones in total) have not been<br />
used in the typology, but will be used in the description of the landscape type.<br />
For the urban landscapes the information was derived from the CORINE land cover database.<br />
However, some extra processing was done to derive only the larger urban agglomerations. For this<br />
purpose a 5 km by 5 km majority filter was used in ERDAS Imagine. This map was integrated within<br />
the landscape map. After this there were additional post-processing steps necessary to upgrade the<br />
European Landscape Map, being summarised below.<br />
A large advantage of the European Landscape Classification is that its selection of bo<strong>und</strong>aries is<br />
consistent, crisp and transparent based on the <strong>und</strong>erlying layers: topography, parent material and land<br />
cover. However, if misclassifications do occur in one of the three <strong>und</strong>erlying layers this is reflected in<br />
the European Landscape Classification. The fact that the European Landscape Classification lacks<br />
information on the land use history is a limiting factor but was so far difficult to collect at the<br />
European scale Maps 8/9 present a side-by-side comparison between LANMAP 1 with the same area<br />
selection of the European Natural Vegetation Map.<br />
99