RR = DB 2 
NPP itself is not really standing for the crop 
production, but representing the amount of 
biomass production. Crop Production is driven 
using NPP by considering “Conversion Ratio” 
which gives a ratio on how many agricultural 
products can be converted from agricultural 
NPP amount at agricultural areas. 
2.2 Agricultural Production 
In this paper crop production is assumed 
according to following rules. 
(a)The biomass produced in the agricultural 
field is converted to crops according to a 
“converting ratio”. (b) The “conversion ratios” 
are different depending on the country 
economic levels. (c) The "converting ratio" 
from NPP to agricultural production is same 
value in one country. 
The procedure to assume agricultural 
production is as follows and shown in Fig-1: 
(1)NPP of each pixel is assumed using Remote 
Sensing data . 
(2NPP on Agricultural fields which can be 
extracted from global land use map is 
computed. 
(3)Biomass on agricultural fields which is 
called "agricultural NPP amount" in this 
paper is computed by multiplied by area of 
crop field, and totaled country base. 
(4)Crop production is estimated from statistics 
for each country. 
(5)A "converting ratio" which provides a ratio 
converting agricultural NPP amount to crop 
production. The "conversion ratio is same 
value in one country. 
(6)An agricultural production map is generated 
by the “converting ratio” multiplied by 
agricultural NPP amount 
Land se | 
estimation map | 
  
Ps National | 
| Boundary Map 
NPP ni on 
Crop Area | 
: 1 A | 
| NPP x Area | | 
| 
I FS Loam ad 
Agricultural National Biomass 
Statistics amount on Crop Area 
Efficiency Index 
for Each Country 
ben males ascet 
Agricultural Productivity 
. Estimation Map 
  
  
Fig-1 Procedure for Agricultural 
Productivity Estimation 
.3. Data 
3.1 NPP 
NPP can be computed by using following 
equation(3-1-1) where APAR (Absorbed Photo- 
synthetically Active Radiation) can be driven by 
using equation(3-1-2) with PAR data and 
NDVI(Goward and Huemmrich, 1992, Price and 
Goward, 1995). 
NDVI and PAR data can be driven respectively 
from 8km composite NOAA/AVHRR NDVI data 
and 1 degree monthly PAR data generated from 
NPP = € [APAR dt (3.1.1) 
APAR = farrx PAR... (3.1.2) 
fra =a+b-NDVI... (3.1.3) 
NPP : Net Primary Production [gDM/mZ/timej 
e : Efficiency Index [g/MJj 
APAR:Absorbed PAR [MJ/m2} 
farar =—0.008+1.075-NDVI (3.1.4) 
Nimbus-7 Total Ozone Mapping 
Spectrometer(TOMS) data. 
The e (Efficiency Index) is always set to 
1.5[g/MJ] proposed by Ruimy(1994), and it is 
suitable for global analysis without precise 
information of land cover. In this study, data 
resolution of 8km is not enough to set parameter 
considering the land cover/usr type. 
Referring the parameters of "a" and "b" in 
equation(3.1.3), there are some discussions to 
determine them. Here, we applied parameters 
10 day composite NDVI (Skm) | PA A deu) | 
i doy ox 9o m) | PAR data A aer 
  
ET ERR EI. UT e PUR 
| ont ly composi te NDVI (8kn) 
CRT EXT ER EY 
fans = -0.008+ 1.075- NDVI 
pue B LL p 
pen) | [mons aw | 
TT TRIO NPPLeent fe 
  
| Monthly NPP (8km) 
| MBP= y NPP.. 
— he 
L Annual NPP (8km) ^ s 
  
Fig-2 Procedure for NPP estimation using 
NDVI and PAR data 
Intemational Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998 467