Files in this item

FilesDescriptionFormat

application/pdf

application/pdfSHAFIEEJOOD-DISSERTATION-2019.pdf (13MB)
(no description provided)PDF

application/vnd.openxmlformats-officedocument.presentationml.presentation

application/vnd.openxmlformats-officedocument.presentationml.presentationShafiee-Jood_Dissertation_SupplementaryFile.pptx (27MB)
(no description provided)Microsoft PowerPoint 2007

Description

Title:Towards effective use of climate forecasts in agricultural decision making: Bridging the gap between modeling and empirical studies
Author(s):Shafiee Jood, Majid
Director of Research:Cai, Ximing
Doctoral Committee Chair(s):Cai, Ximing; Kumar, Praveen
Doctoral Committee Member(s):Deryugina, Tatyana; Stillwell, Ashlynn
Department / Program:Civil & Environmental Eng
Discipline:Civil Engineering
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):Drought Forecast
Drought Management
Value of Information
SWAT
Behavioral Model
Agent-Based Modeling
End-to-End Forecasting
Reinforcement Learning
Asymmetric Learning
Social Network
Crop Allocation
Crop Insurance
Risk Attitude
Abstract:Weather and climate forecasts offer great potential for managing climate variability and climate risk in agricultural systems. In particular, climate forecasts at the seasonal scale can be considered as a key component of proactive drought management. However, there is not much evidence of sustained use and update of forecasts in the real world. While this, in part, has been associated with the characteristics of forecast information (most importantly, accuracy or skill), findings from field-based, empirical studies have shed light on other possible determinants of forecast adoptions. The primary goal of this dissertation is to use insights from empirical studies to develop refined, more realistic models of forecast valuation and adoption. These models are used to explore and understand the determinants of forecast adoption. The first question addressed in this dissertation is how improvement in forecast accuracy influences the value of forecasts. Using a refined, theoretical model of forecast valuation that incorporates user’s perception about forecast accuracy as a behavioral parameter, it is found that the benefits that users derive from improved forecasts depend on users’ characteristics including risk aversion and wealth level. The second research problem investigated in this work is on the impact of social capital and social network structure on the spatial and temporal dynamics of forecast adoption. An agent-based model is developed that simulates how farmers learn about the value of forecasts based on their own and their neighbors’ experiences. It is shown that the structure of the social network is an important factor in adoption of forecasts especially when farmers’ rate of learning from their own experiences is low. Finally, this dissertation investigates the role of institutional interventions (namely crop insurance and crop price) in the value of improved seasonal forecasts. This is investigated by developing an end-to-end forecast valuation framework that integrates a crop growth simulation model and an economic decision-making model. Focusing on the 2012 drought in U.S. Midwest, it is shown that crop insurance and crop price could significantly reduce the value of improved seasonal forecasts during drought conditions. This Dissertation presents a holistic modeling framework to address the effective use of forecasts for agricultural drought management. The models developed in this dissertation are used to generate hypotheses that can be utilized to design intervention and targeting strategies aiming at increasing forecast adoption and to elicit important insights about the interactions between different factors that influence farmers’ forecast adoption.
Issue Date:2019-09-24
Type:Text
URI:http://hdl.handle.net/2142/106159
Rights Information:Copy Right 2019 Seyed Majid Shafiee Jood
Date Available in IDEALS:2020-03-02
Date Deposited:2019-12


This item appears in the following Collection(s)

Item Statistics