Upload
charles-johnson
View
215
Download
2
Tags:
Embed Size (px)
Citation preview
copyright Eurostat 2011
LUCAS 2009 Survey
copyright Eurostat 2011
Coverage 2008/2009Member States Sampling size
Member States Sampling size
Austria 4,969 Latvia 3,827
Belgium 1,808 Lithuania 3,864
Bulgaria * 8,100 Luxembourg 153
Czech Republic 4,674 Malta ** 80
Denmark 2,554 Netherlands 2,461
Estonia 2,680 Poland 18,530
Finland 19,946 Portugal 5,426
France 32,417 Romania * 20,600
Germany 21,157 Slovakia 2,895
Greece 7,819 Slovenia 1,201
Hungary 5,513 Spain 29,917
Ireland 4,165 Sweden 26,665
Italy 17,851 United Kingdom 14,508
Total 263,780
*Survey took place in 2008
** Survey financed with national funds
copyright Eurostat 2011
Informative contentParameters:•Land cover (1 & 2)•Land use (1 & 2)•Land Cover percentage•Area size•Height of trees•Width of features•Transect (250 m to east) with Land Cover and linear features•Land management (grazing)•Water management
WestEast
North
South
SOIL SAMPLE
copyright Eurostat 2011
LUCAS 2008/2009: Nomenclature
Fully hierarchical (3 levels)
Land cover and land use clearly distinguished
Comparable with other main international nomenclatures (e.g. FFS agriculture, FAO, forestry)
– Some new parameters introduced
– Forest types have been introduced for forestry areas, in line with the EUNIS classification on forests (http://eunis.eea.europa.eu/about.jsp)
copyright Eurostat 2011
A10 Built-up areas
A20 Artificial non built-up areas
B10 Cereals (+ triticale)
B20 Root crops
B30 Non permanent industrial crops
B40 Dry pulses, vegetables and flowers
B50 Fodder crops
B70 Fruit trees & berries
B80 Other Permanent Crops
C10 Broadleaved and evergreen woodland
C20 Coniferous woodland
C30 Mixed woodland
D10 Shrubland with sparse tree cover
D20 Shrubland without tree cover
E10 Grassland with sparse tree/shrub cover
E20 Grassland without tree cover
E30 Spontaneous vegetation
F00 Bare Land
G10 Inland water bodies
G20 Inland running water
G30 Coastal water bodies
G50 Glacier, permanent snow
H10 Inland wetlands
H20 Coastal wetlands
Nomenclature LUCAS 2008/09: Land Cover
copyright Eurostat 2011
U110 Agriculture ( + Kitchen garden + Fallow land)
U120 Forestry
U130 Fishing
U140 Mining, Quarrying
U150 Hunting
U210 Energy production
U220 Industry & Manufacturing
U310 Transport, communication, …
U320 Water & waste treatment
U330 Construction
U340 Commerce, Finance, Business
U350 Community Services
U360 Recreation, Leisure, Sport
U370 Residential
U400 Unused
Nomenclature LUCAS 2008/09: Land Use
copyright Eurostat 2011
Sampling strategy: sampling design
1 100 000 points
LAND COVER classes1 ARABLE LAND2 PERMANENT CROPS 3 GRASSLAND4 WOODED AREAS AND SHRUBLAND5 BARE LAND, RARE VEGET.6 ARTIFICIAL LAND7 WATER
First phase sample for stratification: orthophoto interpretation
2km squared grid
Ground survey
Parameters•Land cover•Land use•pictures•etc.
Sample of around 260,000 pts
Second phase sample: in-situ data collection
copyright Eurostat 2011
Aims : obtaining a reasonable balance among the strata (the
agricultural strata were over-represented in the previous rounds of the survey) since the focus of the survey changed from a merely agricultural to a broadly agro-environmental one;
providing reliable estimates at a geographical level more detailed than the EU one for the most relevant land cover classes in the various countries;
gathering longitudinal data on land cover and land use to monitor their changes (for the countries involved in the 2006 round).
Sampling design :
copyright Eurostat 2011
Sampling design: expected errors at NUTS2 level
LAND COVER CLASSES UPPER BOUND EXP. ERRORS
Cereals 15%
Root crops, Vegetables, floricolture, ornamental plants 25%
Fibre and oleaginous crops, non permanent industrial crops 25%
Fodder and temporary grassland 25%
Permanent crops and nursery 25%
Grassland 7.5%
Broadleaved woodland 20%
Coniferous woodland 20%
Mixed woodland 20%
Shrubland 20%
Bare land 20%
Artificial areas 15%
Water 20%
copyright Eurostat 2011
Selection of points: criteria– LUCAS 2006 sample points are included as
much as possible (panel approach)– Points above 1000m and very difficult to access
are excluded– Strategy to reduce autocorrelation among points
Sampling drawing:
copyright Eurostat 2011
Points selection strategy: recourse to replicates
9 23 66 44 68 10 48 16 42 76 23 66 44 68 10 48 16 42 76
8 71 12 69 25 51 29 60 37 7 71 12 69 25 51 29 60 37 7
7 20 79 18 72 78 3 31 63 70 20 79 18 72 78 3 31 63 70
6 33 11 80 11 59 32 38 9 64 33 1 80 11 59 32 38 9 64
5 28 40 26 49 55 17 53 50 77 28 40 26 49 55 17 53 50 77
4 45 27 41 67 6 65 15 73 5 45 27 41 67 6 65 15 73 5
3 35 39 13 36 62 21 57 24 47 35 39 13 36 62 21 57 24 47
2 8 58 74 46 14 75 2 56 34 8 58 74 46 14 75 2 56 34
1 43 30 4 54 61 19 81 22 52 43 30 4 54 61 19 81 22 52
9 23 66 44 68 10 48 16 42 76 23 66 44 68 10 48 16 42 76
8 71 12 69 25 51 29 60 37 7 71 12 69 25 51 29 60 37 7
7 20 79 18 72 78 3 31 63 70 20 79 18 72 78 3 31 63 70
6 33 11 80 11 59 32 38 9 64 33 11 80 11 59 32 38 9 64
5 28 40 26 49 55 17 53 50 77 28 40 26 49 55 17 53 50 77
4 45 27 41 67 6 65 15 73 5 45 27 41 67 6 65 15 73 5
3 35 39 13 36 62 21 57 24 47 35 39 13 36 62 21 57 24 47
2 8 58 74 46 14 75 2 56 34 8 58 74 46 14 75 2 56 34
1 43 30 4 54 61 19 81 22 52 43 30 4 54 61 19 81 22 52
Row in block
1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
Column in block →
To reduce the autocorrelation within and between strata:
1. GRID (2km*2km) is divided into squared blocks of 81 points each
2. Pts coordinates are translated in relative position in the block
3. numbering of the replicates is done under the constraint that the distance with the previous one is maximized
4. Replicates are then selected successively in each stratum (starting with replicate 1) until the required sample size by domain is reached. From the replicate with the highest number, points are randomly selected
copyright Eurostat 2011
Quality assurance - Actions put in place:
Different actors/levels of control
Several training steps
Continuous monitoring
Independent data quality check
Standardisation and computerisation of the main phases of data management
copyright Eurostat 2011
copyright Eurostat 2011
External quality check: main results
Main conclusion of the external quality check:
the overall quality of the data is very good since only 4.5% had serious measurement errors;
Main sources of error:• Mistakes in classification of the transect• Mistakes in attribution of land cover and land use;• photos were not always taken in a proper way.