Agriculture Reference
In-Depth Information
strata can be close to each other and give some extra information on the spatial
correlations between strata.
To reduce the effect of autocorrelation within and between strata, the basic
sampling grid (2 km
18 km)
blocks of 81 points. The set of points with the same relative position in the block is
called a replicate. The replicates are numbered to maximize the distances with the
previous ones. Replicates are then successively chosen until we have obtained the
required sample size for this domain. Points are randomly selected from the
replicate with the highest number. This selection method is combined with a
excluded from the second phase sample, because they are assumed to have little
importance to agriculture.
The LUCAS 2009 survey included a soil module. A topsoil sample was collected
for 10 % of the points. The objective of the soil module was to improve the
availability of harmonized data on soil parameters in Europe.
Recently, EUROSTAT carried out the LUCAS 2012 survey in the European
Union. The LUCAS 2012 covers all 27 EU countries. The fieldwork was carried out
between March and September, 2012. The LUCAS 2012 data can be downloaded
2 km) was divided into square (9
9, that is, 18 km
2.3.3 AGRIT
The Italian AGRIT program (Carfagna and Gallego
2005
; Postiglione et al.
2010
)is
a spatial sampling frame survey that aims at providing conjunctural estimates on
areas, main crop yields, and main land uses. This survey uses spatial sampling
techniques, in particular, point typology.
The method
6
is based on the integration of data collected from ground samples
and remote sensing. The list is formed by a set of points (i.e., the point frame) that
exhaustively covers the land. Each point has an operational dimension area: a 3-m
radius circle centered on the point (covering approximately 30 m
2
). If there are
associations between different land uses, the observation area is extended to approx-
imately 700 m
2
. If there are different land uses, each (up to a maximum of three) is
assigned to the point with a respective proportion according to a pro-rata criterion.
The sampling design is stratified into three phases. The sample of the first phase
uses an aligned spatial systematic selection of geographical units. This list of units
is denoted as the AGRIT sampling frame, and is formed by a regular grid of points
(approximately 1,200,000). The resolution of the grid is 500 m, defined using
the Gauss-Boaga coordinate system,
7
and completely covers the land under
6
The technical details described in this sub-section refer to the 2005 AGRIT program.
7
The Gauss-Boaga projection is a map projection used in Italy that uses a Hayford ellipsoid.
Search WWH ::
Custom Search