the high 
(MTPE) 
rmation. 
yrmation 
riculture 
ation of 
1y other 
tatistical 
o derive 
stigated. 
n types, 
current 
ble. In 
retations 
can be made (e.g. variations in soil moisture). The techniques developed for this project can be 
applied to new data sets as they become available. 
In order to carry this out, the central areas of the soil FAO units are cut from the areas defined. The 
resulting extent is determined from the unit size: statistically small samples are eliminated. Currently 
10511 units remain large enough of the potential 34000 globally. These define the "type soils extent" 
of the unit related to that area. It is against these individually numbered types that the other 
characteristics are evaluated. This then gives the best fit identification of each pathfinder pixel 
location against soil type. At the same time a probability distribution of the goodness of fit is 
generated. 
Previous CERL Global Soils Research 
CERL has developed an extensive global soils interpretation capability by combining the UN and 
USDA background information in a Geographical Information System (GIS). This allows important 
data developed by these groups to be extracted and remanipulated with the rest of the extensive global 
digital data set. 
— 21 —