Generating Global NPP Map for Estimating Agricultural Productivity
Atsushi Hirakoba, Ryosuke Shibasaki and Shiro Ochi
Murai-shibasaki lab., Institute of Industrial Science,
University of Tokyo
7-22-1 Roppongi Minato-ku Tokyo 106, Japan
Tel: (81)-3-3402-6231 Fax: (81)-3-3479-2762
E-mail:hirakoba@shunji.iis.u-tokyo.ac.jp
Abstract The world population are increasing year by year and are estimated to be 8300 million in 2025. Nowadays, there are a lot of arguments whether this Earth has the capacity to supply sufficient food to these drastically increasing population. We need objective materials to debate this kind of global problem and promote consensus for concrete actions.
In this study, we are going to estimate the world net primary productivity (NPP) using NDVI derived from NOAA AVHRR(8km) data and climate data (PAR) with "efficiency" approach (Monteith, 1972, 1977). This study will be followed by the estimation of agricultural productivity map.
Introduction
The world population are increasing year by year and are estimated to be 8300 million in 2025. This expansion of human activity brings pressure on global environment such as land degradation, resource exhaustion, global climate changes, and so on. One of them is due to food supply.
Thomas Rebert malthus, an English Economist, pointed out that "we may be perfectly certain, that the ratio of their increase in a limited territory must be of a totally different nature from the ratio of the ratio of the increase of population. A thousand millions are just as easily doubled every twenty-five years by the power of population as a thousand. But the food to support the increase from the greater number will by no means be obtained with the same facility." In his famous work "An Essay on Population". The relation between population increase and cereal population increase and cereal
production from 1961-1995 wee shown in Figure 1. In spite of population increase, from 3100 million to 5700 million, serious food deficit has not occurred. That is because food production has increased more drastically.
Figure 1: Relation between Population & Production 1961-1995 (Faostat, 1995)
Most of researches about food productivity are based on concept that same situation continue statistically but they don't consider clearly the limitation of water resources, land degradation, and so on. The final goal of this study is to estimate global food productivity using satellite data. In can be used to plan land use considering ground condition such as water resource limitation, land degradation and so on.
Methodology - Production Efficiency Approach-
In this study, we estimate NPP, which is the energy of biomass saved in dry organic. (Total amount of energy that plants produce minus the energy plants use during cellular respiration.) Accumulation of NPP - approximate of total biomass-in arable land must have strong correlation to crop production.
NPP estimation is based on "Production Efficiency Concept". This concept is first introduction by Monteith(1972). Monteith observed that plant production is correlated with the amount of photosynthetically action radiation (PAR) absorbed or intercepted by green foliage (APAR). Subsequently Monteith(1977) showed that plant production is approximately related to APPAR. This relation is summarized in equation below.
P=eňfAPAR . PARdt
Where
PAR is incoming photosynthetically active radiation (MJ/m
2)
f
APAR is the fraction of PAR absorbed by canopy
f
APAR *PAR is APAR, canopy-absorbed incident solar radiation (MJ/m
2)
e is conversion efficiency of APAR into organic dry matter (g/MJ)
Furthermore, several researchers showed the fraction of incident PAR absorbed by the canopy (fAPAR) cab be linearly related to the NDVI (Goward and Huemmrich (1992), Prince and Goward (1995)).
fAPAR = a + b * NDVI
where, NDVI is normalized difference vegetation index.
| NDVI =
| NIR-R
----------------
NIR + R
|
NIR: surface reflectance's in the near-infrared wavelengths
R: surface reflectance's in the red wavelengths
The ratio has limit of -1 and +1. The higher NDVI is, the denser vegetation is.
Conversion efficiency of absorbed photosynthetically active radiation into primary production, e, is the variable introducing the greatest incertitude on the estimates. This is regarded as an empirical constant when it was first introduced (Monteith, 1972). Without climatic or other limitations, the upper limit was about 3.5 g/MJ. Hunt (1994) showed that the upper limit of e decreased to 2.0 g/MJ because of increased maintenance respiration with increasing woody biomass. e is constrained by environmental parameters, mainly air temperature, mainly air temperature, vapor pressure deficit, and soil moisture. Ruimy (1994) showed the variation of e according to vegetation. Production thus may be predicted from fAPAR -NDVI relation and the efficiency, e.
