Transcript Biophysical modelling: data pre

INSEA
biophysical modelling:
data pre-processing
By Juraj Balkovič & Rastislav Skalský
SSCRI Bratislava
Workshop at JRC in Ispra, Italy
11th – 12th April, 2005
Outlines:





HRU – delineation
GIS-based prototype for EPIC soil and
topographical inputs
LUCAS Phase I. in EPIC BFM
Crop Rotation Set-Up
Topics for discussion
1k-based delineation of Homogeneous
Response Unit (HRU):
Elevation classes:
1 – 0-300 m lowland
2 – 300-600 m upland
3 – 600-1100 m high mts.
4 – > 1100 m very high mts.
Texture classes:
1 – coarse
2 – medium
3 – medium fine
4 – fine
5 – very fine
6 – no texture
7 – rock
8 – peat
Depth to rock classes:
1 – shallow (< 40 cm)
2 – moderate (40-80 cm)
3 – deep (80-120 cm)
4 – very deep (>120 cm)
HRU
intersect
Depth to Gley horizon:
1 – shallow
2 – moderate
3 – deep
Climate:
?Annual rainfall
Slope classes:
Volume of stones:
1 – without
2 – moderate
3 – stony
Temporary
HRU raster for
EU25:
126 HRUs
It intersects only
elevation, slope
for arable land
and textural
classes
HRU – raster
(1km)
GIS-based prototype for EPIC
soil and topographical inputs

Once HRU-layer is set...The prototype is
designed

ERDAS IMAGINE (GIS)
VISUAL BASIC (Conversion)
MS ACCESS (Database)


NUTS 2 GIS-based
prototype:
Subset in batch
1km data
• Soil
• Topography
• Land Use
Generates raster subsets
for extent of selected
NUTS2 regions
AOI layer
1km subset data
for NUTS2
regions
• Soil
• Topography
• Land Use
1km subset data
for NUTS2
regions
• Soil
• Topography
• Land Use
LandCat specific
Zone statistics (ERDAS IMAGINE Modul)
ASCII outputs
Calculated statistics for
combinations of NUTS2
and Land Categories from
1k subset rasters (soil and
topography)
VISUAL BASIC Script to append
ASCII outputs into final table
ASCII outputs
Calculated statistics for
combinations of NUTS2
and Land Categories from
1k subset rasters (soil and
topography)
Ontology table
MS ACCESS
Filters over RESULT- table (how to reduce the
number of HRUs with certain purpose):
A. Coding by schematic ontology codes >
NUTS2_LC_SOILCLASS
ALTIT_SLOPE_TEXT
e.g. Aggregate by
slope for arable
Aggregate by altitude
CROP ROTATION
ALLOCATION
Redistribute and aggregate results
by simplifying rules
B. Filter by minimum-area criterion >
according to SOILIDFR
LUCAS Phase I. in EPIC BFM
• Breaking Down New Cronos Statistics by LUCAS Data
LUCAS
Rough
Database
Crop Aggregation,
Attribute adjustment,
Filter for Agricultural Land
LUCAS
Pre-processed
processing
100.0
Downscale by altitude
FALLOW
TEMP_PAST
FLORES
VEGETABLE
SUNFLOW
SUGAR
PULSES
POTATO
OIL_REST
RAPE
MAIZETOT
WINTCER
BARLEY_REST
90.0
LUCAS Frequencies
80.0
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
< 300
Altitude
300-600
> 600
LUCAS Phase I. in EPIC BFM
100.0
FALLOW
TEMP_PAST
FLORES
VEGETABLE
SUNFLOW
SUGAR
PULSES
POTATO
OIL_REST
RAPE
MAIZETOT
WINTCER
BARLEY_REST
90.0
LUCAS Frequencies
80.0
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
< 300
Altitude
300-600
> 600
NC Crop Shares
processing
NC Crop shares
broken down to
altitude classes
LUCAS Phase I. in EPIC BFM
Crop Rotation Setup
wi  Pj N i
where PJ denotes NC share of crops included in i-th crop rotation. Ni is number of crops
involved in i-th crop rotation system
Crop Rotation Setup
Original NC data
Crop shares
Broken NC data
Crop shares
CORINE Data
Area of arable land
+ Hetero agric. area
Crop rotation systems for NUTS2 region, for
its HRUs/ aggregated by altitude classes
respectively
Discussion

Digital data



1km soil data
Coverage of climate for delineation (e.g. annual
precipitation 1km from IIASA)
DEM 1km – statistics from 90 x 90 m DEM source
(average slope or dominant slope) – for erosion
simulations

Consistency of GISCO GIS Database and
EUROSTAT Databases in NUTS2 Coding

Fertilization, irrigation and tillage with CAPRIDYNASPAT