A data acquisition framework for runoff prediction in ungauged basins - Runoff Prediction in Ungauged Basins: Synthesis across Processes, Places and Scales - page 42

Figure 3.9. Example of a global
data set that will be needed for a
hyper-resolution hydrological
model. The data set consists of
elevation, stream networks,
catchment boundaries, drainage
directions, and ancillary data layers
such as flow accumulations,
distances, and river topology at
various resolutions from
approximately 90 m to 10 km and
is based on data from NASA s

Shuttle Radar Topography Mission.
From Wood et al.( 2011 ).

'

three resolutions (1 km, 8 km, and 1 degree). For model-
ling purposes, hydrologically relevant parameters (evap-
oration resistance, leaf area index etc.) need to be
associated with the assigned land use classes. The Global
Land Cover Characterization (GLCC) has been more
recently developed through a joint effort of the USGS,
the University of Nebraska-Lincoln (UNL), and the Euro-
pean Joint Research Center (JRC). This data set was also
compiled from AVHRR data at a resolution of 1 km (or
30 arc-seconds), but more land cover classes have been
identified ( Figure 3.10 ). Another very recent global land
cover data set, released in September 2008, is the Glob-
Cover project of the ESA. This data set, compiled from
ENVISAT MERIS (MEdium Resolution Imaging Spec-
trometer) images between January 2005 and June 2006,
has a spatial resolution of 300 metres. Assessing the
accuracy of satellite-derived land cover data is a challenge
as a range of different assessment methods are used in the
scientific community (Foody, 2002 ) and the data sets are
not always consistent (Giri et al., 2005 ; Mayaux et al.,
2006 ).
Besides global data sets, continental-scale land cover
maps have been developed, especially for the USA and
Europe. Two examples of maps for Europe are the COR-
INE (Coordination of Information on the Environment)
and PELCOM (Pan-European Land Cover Monitoring
and Mapping project; Mücher et al., 2000 ) databases.
Regional, basin-wide, and local land cover and land use
maps have also been developed, but their availability
varies widely across countries and the globe.

3.5.3 Soils and geology
Whereas land cover data can be estimated relatively easily
from remote sensing, this is much more difficult for soil
properties. The FAO-UNESCO digital soil map of the
world, compiled between 1971 and 1981, has been used
in many global analyses (e.g., Nijssen et al., 2001 ; Hurk-
mans et al., 2008 ). It has a spatial resolution of 5 arc-
minutes, and is compiled from over 600 national soil maps
and over 11 000 other maps that were provided by national
soil organisations (Reynolds et al., 2000 ). This map has
been extended to the FAO Soil Database system (SDB),
where for each mapping unit in the soil map, parameters
have been assigned to the topsoil (0

-

30 cm) and the subsoil

(30 100 cm). Parameters include soil texture classes (per-

centages of sand, clay and silt), porosity, bulk density and
organic carbon fragments (Reynolds et al., 2000 ). A very
recent expansion of the FAO soil map of the world is the
Harmonized World Soil Database (HWSD; Nachtergaele
et al., 2009 ). This data set is a joint effort of FAO, IIASA
(International Institute for Applied Systems Analysis),
ISRIC World Soil Information, Institute of Soil Sciences,
Chinese Academy of Sciences and the Joint Research
Center. It is basically a high-resolution (30 arc-second)
soil map of the world, with each pixel containing data
including organic carbon, pH, soil depth, water storage
capacity, sand, silt and clay contents, USDA texture,
exchangeable nutrients, sodicity, salinity, lime and gypsum
fractions, and other properties. This data is available for
two layers: 0

-

100 cm. Additional infor-
mation about HWSD is provided by Nachtergaele et al.

- 30 cm and 30 -