This repository contains all the features and instructions necessary to reproduce the agriLOWE ABM model. The model encompasses a spatial-based economy with an ecology of N farmers and 3 different types of land. Technological progress as well as shocks deriving from the interactions with the climate are modelled.
The simulation aims at addressing these two open issues: is it possible to assess the climate change impacts to the agricultural sector in different technologically – developed countries, utilizing a global model? Further, is low technological change in the agricultural sector a bottleneck for development and climate change mitigation?
- Ecology of
N
farmers (no exit & no entry); - 3 types of land: forestry, low-yield land (characterized by low fertility and intrinsic low productivity of the land) and high-yield land (where the fertility of the ground and the related productivity are instead higher);
Production = min {alphaL, gammaS, thetaK}
withalpha
: productivity of labour embodied in the machines;gamma
: productivity of land;theta
: productivity of capital;
- Each farmer produce an homogenous bundle of food and the market is perfectly competitive;
- Demand is increasing at a fix rate over time:
D=D(t-)(1+*d*)
- Labour force is equally distributed among cells :
l(t)=L/(xy-f)
withf
forestry cells - Cost structure is marginalist:
C(it)= w/alpha + p_land/gamma
- Technological progress depends on R&D internal and R&D external
Profit = ((p_food - UC(it))sales(it))) - R&Dint - R&Dext
- Damage function: Nordhaus (2017, Weitzman (2009)
- Carbon uptake: Sterman et al. (2012) and IIASA's Global Biosphere Management Model (GLOBIOM)
Code Language : R, C++ (under development)
Compile : R no need, C++ uses CMake
Inputs: 21 + a spatial matrix
Outputs: 14 time series of interest
- Data: import parameter set;
- Declare
- Initialize
- while t<T do:
- New machines are produced
- Receive food demand
- Order new machineries
- Perform internal R&D
- Produce
- Sell and compute profits
- Obtain new machinery
end
agriLOWE has no system specific requirements. It is designed to minimize the level of constraints.