Adaptation: Sensitivity to natural variability, agent assumptions and dynamic climate changes

Stephen H. Schneider, William E. Easterling, Linda O. Mearns

Research output: Contribution to journalArticle

114 Citations (Scopus)

Abstract

The role of adaptation in impact assessment and integrated assessment of climate policy is briefly reviewed. Agriculture in the US is taken as exemplary of this issue. Historic studies in which no adaptation is assumed (so-called 'dumb farmer') versus farmer-agents blessed with perfect foresight (so-called 'clairvoyant farmer') are contrasted, and considered limiting cases as compared to 'realistic farmers.' What kinds of decision rules such realistic farmer-agents would adopt to deal with climate change involves a range of issues. These include degrees of belief the climate is actually changing, knowledge about how it will change, foresight on how technology is changing, estimation of what will happen in competitive granaries and assumptions about what governmental policies will be in various regions and over time. Clearly, a transparent specification of such agent-based decision rules is essential to model adaptation explicitly in any impact assessment. Moreover, open recognition of the limited set of assumptions contained in any one study of adaptation demands that authors clearly note that each individual study can represent only a fraction of plausible outcomes. A set of calculations using the Erosion Productivity Impact Calculator (EPIC) crop model is offered here as an example of explicit decision rules on adaptive behavior on climate impacts. The model is driven by a 2xCO2 regional climate model scenario (from which a 'mock' transient scenario was devised) to calculate yield changes for farmer-agents that practice no adaptation, perfect adaptation and 20-year-lagged adaptation, the latter designed to mimic the masking effects of natural variability on farmers' capacity to see how climate is changing. The results reinforce the expectation that the likely effects of natural variability, which would mask a farmer's capacity to detect climate change, is to place the calculated impacts of climate changes in two regions of the US in between that of perfect and no adaptation. Finally, the use of so-called 'hedonic' methods (in which land prices in different regions with different current average climates are used to derive implicitly farmers' adaptive responses to hypothesized future climate changes) is briefly reviewed. It is noted that this procedure in which space and time are substituted, amounts to 'ergodic economics.' Such cross-sectional analyses are static, and thus neglect the dynamics of both climate and societal evolution. Furthermore, such static methods usually consider only a single measure of change (local mean annual temperature), rather than higher moments like climatic variability, diurnal temperature range, etc. These implicit assumptions in ergodic economics make use of such cross-sectional studies limited for applications to integrated assessments of the actual dynamics of adaptive capacity. While all such methods are appropriate for sensitivity analyses and help to define a plausible range of outcomes, none is by itself likely to define the range of plausible adaptive capacities that might emerge in response to climate change scenarios.

Original languageEnglish (US)
Pages (from-to)203-221
Number of pages19
JournalClimatic Change
Volume45
Issue number1
DOIs
StatePublished - Jun 22 2000

Fingerprint

climate change
climate
climate effect
economics
regional climate
environmental policy
climate modeling
temperature
agriculture
erosion
productivity
crop
method
decision
impact assessment
effect

All Science Journal Classification (ASJC) codes

  • Global and Planetary Change
  • Atmospheric Science

Cite this

@article{95acef9177b34958bc3f28422d8a69ae,
title = "Adaptation: Sensitivity to natural variability, agent assumptions and dynamic climate changes",
abstract = "The role of adaptation in impact assessment and integrated assessment of climate policy is briefly reviewed. Agriculture in the US is taken as exemplary of this issue. Historic studies in which no adaptation is assumed (so-called 'dumb farmer') versus farmer-agents blessed with perfect foresight (so-called 'clairvoyant farmer') are contrasted, and considered limiting cases as compared to 'realistic farmers.' What kinds of decision rules such realistic farmer-agents would adopt to deal with climate change involves a range of issues. These include degrees of belief the climate is actually changing, knowledge about how it will change, foresight on how technology is changing, estimation of what will happen in competitive granaries and assumptions about what governmental policies will be in various regions and over time. Clearly, a transparent specification of such agent-based decision rules is essential to model adaptation explicitly in any impact assessment. Moreover, open recognition of the limited set of assumptions contained in any one study of adaptation demands that authors clearly note that each individual study can represent only a fraction of plausible outcomes. A set of calculations using the Erosion Productivity Impact Calculator (EPIC) crop model is offered here as an example of explicit decision rules on adaptive behavior on climate impacts. The model is driven by a 2xCO2 regional climate model scenario (from which a 'mock' transient scenario was devised) to calculate yield changes for farmer-agents that practice no adaptation, perfect adaptation and 20-year-lagged adaptation, the latter designed to mimic the masking effects of natural variability on farmers' capacity to see how climate is changing. The results reinforce the expectation that the likely effects of natural variability, which would mask a farmer's capacity to detect climate change, is to place the calculated impacts of climate changes in two regions of the US in between that of perfect and no adaptation. Finally, the use of so-called 'hedonic' methods (in which land prices in different regions with different current average climates are used to derive implicitly farmers' adaptive responses to hypothesized future climate changes) is briefly reviewed. It is noted that this procedure in which space and time are substituted, amounts to 'ergodic economics.' Such cross-sectional analyses are static, and thus neglect the dynamics of both climate and societal evolution. Furthermore, such static methods usually consider only a single measure of change (local mean annual temperature), rather than higher moments like climatic variability, diurnal temperature range, etc. These implicit assumptions in ergodic economics make use of such cross-sectional studies limited for applications to integrated assessments of the actual dynamics of adaptive capacity. While all such methods are appropriate for sensitivity analyses and help to define a plausible range of outcomes, none is by itself likely to define the range of plausible adaptive capacities that might emerge in response to climate change scenarios.",
author = "Schneider, {Stephen H.} and Easterling, {William E.} and Mearns, {Linda O.}",
year = "2000",
month = "6",
day = "22",
doi = "10.1023/A:1005657421149",
language = "English (US)",
volume = "45",
pages = "203--221",
journal = "Climatic Change",
issn = "0165-0009",
publisher = "Springer Netherlands",
number = "1",

}

Adaptation : Sensitivity to natural variability, agent assumptions and dynamic climate changes. / Schneider, Stephen H.; Easterling, William E.; Mearns, Linda O.

In: Climatic Change, Vol. 45, No. 1, 22.06.2000, p. 203-221.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Adaptation

T2 - Sensitivity to natural variability, agent assumptions and dynamic climate changes

AU - Schneider, Stephen H.

AU - Easterling, William E.

AU - Mearns, Linda O.

PY - 2000/6/22

Y1 - 2000/6/22

N2 - The role of adaptation in impact assessment and integrated assessment of climate policy is briefly reviewed. Agriculture in the US is taken as exemplary of this issue. Historic studies in which no adaptation is assumed (so-called 'dumb farmer') versus farmer-agents blessed with perfect foresight (so-called 'clairvoyant farmer') are contrasted, and considered limiting cases as compared to 'realistic farmers.' What kinds of decision rules such realistic farmer-agents would adopt to deal with climate change involves a range of issues. These include degrees of belief the climate is actually changing, knowledge about how it will change, foresight on how technology is changing, estimation of what will happen in competitive granaries and assumptions about what governmental policies will be in various regions and over time. Clearly, a transparent specification of such agent-based decision rules is essential to model adaptation explicitly in any impact assessment. Moreover, open recognition of the limited set of assumptions contained in any one study of adaptation demands that authors clearly note that each individual study can represent only a fraction of plausible outcomes. A set of calculations using the Erosion Productivity Impact Calculator (EPIC) crop model is offered here as an example of explicit decision rules on adaptive behavior on climate impacts. The model is driven by a 2xCO2 regional climate model scenario (from which a 'mock' transient scenario was devised) to calculate yield changes for farmer-agents that practice no adaptation, perfect adaptation and 20-year-lagged adaptation, the latter designed to mimic the masking effects of natural variability on farmers' capacity to see how climate is changing. The results reinforce the expectation that the likely effects of natural variability, which would mask a farmer's capacity to detect climate change, is to place the calculated impacts of climate changes in two regions of the US in between that of perfect and no adaptation. Finally, the use of so-called 'hedonic' methods (in which land prices in different regions with different current average climates are used to derive implicitly farmers' adaptive responses to hypothesized future climate changes) is briefly reviewed. It is noted that this procedure in which space and time are substituted, amounts to 'ergodic economics.' Such cross-sectional analyses are static, and thus neglect the dynamics of both climate and societal evolution. Furthermore, such static methods usually consider only a single measure of change (local mean annual temperature), rather than higher moments like climatic variability, diurnal temperature range, etc. These implicit assumptions in ergodic economics make use of such cross-sectional studies limited for applications to integrated assessments of the actual dynamics of adaptive capacity. While all such methods are appropriate for sensitivity analyses and help to define a plausible range of outcomes, none is by itself likely to define the range of plausible adaptive capacities that might emerge in response to climate change scenarios.

AB - The role of adaptation in impact assessment and integrated assessment of climate policy is briefly reviewed. Agriculture in the US is taken as exemplary of this issue. Historic studies in which no adaptation is assumed (so-called 'dumb farmer') versus farmer-agents blessed with perfect foresight (so-called 'clairvoyant farmer') are contrasted, and considered limiting cases as compared to 'realistic farmers.' What kinds of decision rules such realistic farmer-agents would adopt to deal with climate change involves a range of issues. These include degrees of belief the climate is actually changing, knowledge about how it will change, foresight on how technology is changing, estimation of what will happen in competitive granaries and assumptions about what governmental policies will be in various regions and over time. Clearly, a transparent specification of such agent-based decision rules is essential to model adaptation explicitly in any impact assessment. Moreover, open recognition of the limited set of assumptions contained in any one study of adaptation demands that authors clearly note that each individual study can represent only a fraction of plausible outcomes. A set of calculations using the Erosion Productivity Impact Calculator (EPIC) crop model is offered here as an example of explicit decision rules on adaptive behavior on climate impacts. The model is driven by a 2xCO2 regional climate model scenario (from which a 'mock' transient scenario was devised) to calculate yield changes for farmer-agents that practice no adaptation, perfect adaptation and 20-year-lagged adaptation, the latter designed to mimic the masking effects of natural variability on farmers' capacity to see how climate is changing. The results reinforce the expectation that the likely effects of natural variability, which would mask a farmer's capacity to detect climate change, is to place the calculated impacts of climate changes in two regions of the US in between that of perfect and no adaptation. Finally, the use of so-called 'hedonic' methods (in which land prices in different regions with different current average climates are used to derive implicitly farmers' adaptive responses to hypothesized future climate changes) is briefly reviewed. It is noted that this procedure in which space and time are substituted, amounts to 'ergodic economics.' Such cross-sectional analyses are static, and thus neglect the dynamics of both climate and societal evolution. Furthermore, such static methods usually consider only a single measure of change (local mean annual temperature), rather than higher moments like climatic variability, diurnal temperature range, etc. These implicit assumptions in ergodic economics make use of such cross-sectional studies limited for applications to integrated assessments of the actual dynamics of adaptive capacity. While all such methods are appropriate for sensitivity analyses and help to define a plausible range of outcomes, none is by itself likely to define the range of plausible adaptive capacities that might emerge in response to climate change scenarios.

UR - http://www.scopus.com/inward/record.url?scp=0034036823&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034036823&partnerID=8YFLogxK

U2 - 10.1023/A:1005657421149

DO - 10.1023/A:1005657421149

M3 - Article

AN - SCOPUS:0034036823

VL - 45

SP - 203

EP - 221

JO - Climatic Change

JF - Climatic Change

SN - 0165-0009

IS - 1

ER -