468 
shown in equation (3.1.4) proposed by Goward 
and Huemmrich(1992), since the parameters are 
determined under the very common conditions 
in terms of sun zenith angle, latitude etc., and 
are suitable for global assessment. 
Figure-2 shows the procedure for producing a 
map of annual amount of NPP from monthly 
NDVI and PAR data. 
3.2 Land Use Data 
Land Use Map was downloaded from the 
URL:http://www.geog.umd.edu/landcover/8km- 
map/download.html, which was generated by 
Dept. of Geography, University of Maryland. 
The land uses are classified to 13 classes by the 
decision tree method using NDVI seasonal data 
and other information of BAND-1,2,3,4,5 of 
NOAA AVHRR. 
3.3 Agricultural Production Data 
Agricultural Production data can be down- 
loaded from the URL:http:/apps.fao.org/cgi- 
bin /nph-db.pl?subset= agriculture 
There are some changes on the 
boundary and definition in different years and 
country 
different data sources, therefore analysis is not 
made to such countries. In order to compare 
amount of agriculture production with common 
unit scale, the amount of agricultural production 
is evaluated by converting from typical crops 
production(weight) to calorie base according to 
crop-calorie data: such as (a)rice(3.51Mcalton), 
(b)wheat(3.35Mcal/ton), (c)barley (3.39Mcal/ton), 
(d)corn(3.50Mcal/ton),(e)sorghum (3.36 Mcal/ 
ton ),(Dsoybean(4.28Mcal/ton), (g)cassava (1.50 
Mcal/ton) | (h)potato(0.77Mcal/ton), |. (sweet 
potato (1.23Mcal/ton), (j)oats (3.17 Mcalton), 
(k)millet(3.07Mcal/ton), Urye (3.33 Mcalton). 
4.. OUTPUT 
4.1 NPP Distribution Map 
"NPP Distribution Map "(Fig-3) was generated 
by accumulating monthly NPP for 12 month of 
1989. By overlaying this map with land use map, 
accumulated NPP on points which are classified 
as crop field. There are some countries those 
have no crop field in the land use map such as 
Burkina Faso, Gambia, Jordan, Kuwait and so 
on. This may be caused by the reason that crop 
fields in such countries are not enough large to 
be detected by 8kmx8km ground resolution 
satellite data. 
4.2 Agricultural Biomass 
Agricultural biomass is defined as annual 
production in pixels classified as crop field in the 
land use map. The value was calculated by 
multiplying annual NPP by the area of each 
pixel. 
4.3 Converting Ratio 
“Converting Ratio” indicates how efficient the 
agricultural biomass can be converted to crop 
production. Fig-4 shows “Converting Ratio Map” 
where USA, some countries in Europe, Japan 
and China have high values, and many 
Table-1 Efficiency Index 
  
  
  
  
  
  
  
  
  
  
  
Country E. Indx 
Netherlands 2471 
UK 1934 
Luxembourg 1844 
Japan 1811 
Central Africa 98 
Namibia 94 
Sudan 75 
Botsuwana 68 
  
countries in Africa have low values, Table-1 is 
Efficiency Index for some 
countries with high and low values of it. 
sample of the 
4.4 Agricultural Production Map 
Fig-5 is Agricultural Production Map which was 
generated by the procedure described in Fig-1. 
The value was computed for the pixels of crop 
field in the land use map. 
5. DISCUSSION 
There still have some problems in the algorithm 
described in this paper. The validation is now 
detailed Fig-6 
illustrates the relation between the statistic 
made using information. 
data of agricultural production from “US Census 
Intemational Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998