The LUCAS project Land Use Cover Area Frame Survey

1
The LUCAS project Land Use / Cover Area Frame
Survey
Marco FRITZ European Commission Eurostat, E1 -
Agricultural Statistics. Methodology LUCAS
Office address BECH C2/614 5, rue Alphonse
Weicker, L-2721 Luxembourg Tel. [phone]
Fax [phone] e-mail marco.fritz_at_ec.europ
a.eu http//forum.europa.eu.int/irc/dsis/landstat/
info/data/index.htm
2

• What is LUCAS?
• Methodology
• Results
• Conclusions

3
What is LUCAS?

• An area frame survey
• Statistical method based on the visual
  observation of sampled geo-referenced points to
  compute area estimates
• Objectives
• To obtain harmonised data at EU25 level on
• Land use and Land cover
• Environmental features
• To provide a common methodology-nomenclature
  ensuring full comparability of results
• To calculate early estimates of areas (main
  crops) in June/July
• To analyse the interaction agriculture-environment
  -landscape (multi-purpose survey)

4
What is LUCAS? - Calendar of activities

• 2001 - survey carried out in 13 MS
• 2002 - survey carried out in UK and IE and in
  EE-HU-SI
• 2003 - survey carried out in 15 MS Hungary
• 2005 - survey carried out in LT-LV-PL
• 2006 - survey carried out in 11 MS
• PHARE Multi-Country 2000 and 2003 Programmes
  on Sectoral Pilot Projects in Agriculture and
  Environment Statistics

5
What is LUCAS? Coverage 2001
88030 points surveyed
6
What is LUCAS? Coverage 2006
169943 points surveyed
7

• What is LUCAS?
• Methodology
• Results
• Conclusions

8
Methodology Sampling Design

• Base sample
• Uniform square grid of 1km side length over the
  entire territory of the EU Member States
• ETRS89 Lambert Azimuthal Equal Area coordinate
  reference system (ETRS-LAEA)
• Centre point 52North 10 East
• Origin 4.321.000 meters west, 3.210.000 meters
  south of centre point
• 4.025.583 nodes base sample points
• Master sample (even points of the base sample)
• 2km grid about 1 Mio sample points
• Photo-interpretation of all master sample points
  into 7 strata
• arable land
• permanent crops
• grassland,
• wood/shrubland,
• bare land, low or rare vegetation,
• artificial land
• water

9
Methodology Stratification results
10
Methodology Process
Image 2000
1.000.000 points
Photo-Interpretation
Ortho-Photos
Sampling
250.000 points
Survey
Compute statistics
11
Methodology Data collected

• 407 surveyors and 39 supervisors have collected
  the following data
• Land cover
• Land use
• Point and crop photos
• Landscape photos (5 to 6 photos/point in 2006
  1.000.000 photos. Estimated size 1.5 Tb)
• On average
• 13 points surveyed/day
• 400 points/surveyor
• 30 days of work/surveyor

12
Methodology Nomenclature
55 classes
33 classes
13
Methodology Observations on the ground...
Reaching the points
Observing the features in the field
14
(No Transcript)
15
(No Transcript)
16

• What is LUCAS?
• Methodology
• Results
• Conclusions

17
Results Comparison Stratification/Ground survey
CZ
18
Results Accuracy

• Accuracy required by Council Regulations (EEC) N
  959/93 on other crop statistics and (EEC) N
  837/90 on cereals production

19
Results LUCAS 2006 Survey CZ (before QC)
20
Results LUCAS 2006 Survey EU11
21
Results LUCAS 2006 Survey CZ
22

• What is LUCAS?
• Methodology
• Results
• Conclusions

23
Conclusions

• LUCAS is not covering all countries
• Sampling focused on agricultural area but also
  covering all strata
• Stratification available
• Around 1.000.000 photos taken. Photos have to be
  checked and anonymised before transmission
• Follow-up evaluation of landscape classification,
  Grid data combination, comparison with other
  surveys, further use of LUCAS data, technological
  watch. Call for tender for erosion risk
  features
• Photo-interpretation is not an easy process
  without ground information (confusion matrix)

24
Further conclusion

• nor is it observing on the ground!