Page 269 - Climate Change and Food Systems

chapter 8: the potential impact of climate change-induced sea level riseon the global rice market and food security in asia
rice production and price. An estimated 1.6-2.73 percent reduction in global rice production was predicted, followed by an estimated 7.14-12.77 percent rise in global rice price. The study revealed that many key rice-consuming countries, including Bangladesh, Egypt, Japan, Myanmar and Viet Nam, are expected to be among the hardest hit by the effects of SLR.
The purpose of this study is to investigate
how SLR would affect land-based agricultural production and trade. First, in contrast to previous studies, we take into account not only the impact of SLR-inflicted agricultural extent loss (supply- side impact), but also the changes in food demand due to market price economic growth (demand-side impact). Second, in contrast to
the partial equilibrium approach by Chen et al.,
we use a global computable general equilibrium (CGE) model – the Global Trade Analysis Project (GTAP) model (Hertel, 1997) – in order to account for the substitution effects between rice and other food grains. Third, we use the GTAP Version 7 database (Narayanan and Walmsley, 2008) and the GTAP land-use database (Lee et al., 2009), which adopts the concept of agro-ecological zones (AEZ), as developed by the Food and Agriculture Organization of the United Nations (FAO, 2000) and Fischer et al. (2002). Fourth, for the potential SLR-inflicted agricultural extent loss, we draw on estimates from Dasgupta et al. (2009) regarding agricultural land loss due to a one- metre SLR as the land-input shock to the GTAP model.
The paper proceeds as follows: Section
2 introduces the GTAP model and our modifications in land-use modelling. Section 3 provides the simulation results of SLR impact on region-specific crop production, global food prices, and land-use change. Section 4 concludes the paper.
2. Model and data
The standard GTAP model allows all land-based sectors to compete for land according to relative
land rents4. However, this model does not explicitly identify the suitability and viability of land for growing various crops. For a country with arable land located under diverse climate and terrain conditions, crop suitability of the land may diverge, and thus land-use change between sectors may be subject to temperature, precipitation and soil conditions of the particular locations. Therefore, we modify the GTAP model with the GTAP land-use database of Lee et al. (2009) so that transition of land between alternative uses is subject to crop suitability of land, which is implied by the terrestrial characteristics and weather conditions at the location of the land. We believe this modification describes the adjustment in agricultural land use more effectively and more realistically.
We combined the land-use database and version 7 of the GTAP database (Narayanan and Walmsley, 2008). The version 7 GTAP database
is used as the benchmark equilibrium for global CGE simulations, which present transactions of commodities and services between sectors within and across countries/regions in 2004. Table 1 lists the 29 world countries/regions of our simulation. Table 2 shows the sector disaggregation, covering 17 production activities of the whole economy. Key food crops such as rice, wheat and other cereal grains, as well as processed food sectors, are disaggregated, while activities not directly linked with agricultural land are aggregated into energy, manufacturing and service sectors.
The GTAP land-use database is compiled following the AEZ approach developed jointly by FAO (2000) and Fischer et al. (2002). Arable land is classified into six categories by length of growing period. It is further defined in terms of three climatic zones (boreal, temperate, and tropical), which
refer to the length of time within a year that the temperature is above 5o C and the soil moisture
is sufficient for crop growth (Lee et al., 2009). However, irrigation is not accounted for in the AEZ methodology; the GTAP land-use model considers only the rainfed conditions of land resources.
Land rents refer to the market value of the output that
4
equalizes the added cost of an additional unit of land.
249