Keywords
Keywords: Climate change impact assessment, scenario-neutral, bottom-up
Start Date
26-6-2018 3:40 PM
End Date
26-6-2018 5:00 PM
Abstract
Scenario-neutral methods are being adopted more frequently in climate impact assessments, due in part to the uncertainty present in climate model projections. Scenario-neutral methods ‘stress test’ a system against a range of hydroclimate scenarios, which are formed by perturbing statistics of the climate variable time series—climate ‘attributes’. This allows the uncovering of system sensitivities and modes of failure irrespective of climate projections. However, it introduces another form of uncertainty: which climate attributes should a system be stress tested against?
Scenario-neutral studies typically consider changes in simple attributes, the most common being annual averages of hydroclimate variables. The complex interactions between hydroclimate and many systems means that the attributes a system is most sensitive to may not be investigated. Hence, solutions designed in response to uncovered modes of failure may not be appropriate. To address this limitation, we have developed an approach to select climate attributes that describe the most change in system performance. The approach begins by evaluating a comprehensive set of candidate attributes that may affect a system (including means, percentiles and measures of persistence). Variable selection techniques are then used to reduce the number of attributes down to those most critical for system performance. Applied to a case study with two objectives, the approach narrowed ten attributes to two sets of four (i.e. different for each objective) including non-standard attributes like the number of frost days and growing season length. This emphasises the need to strategically evaluate which climate attributes should be used in systems stress testing.
Identifying the climate variables to which water resource systems are most sensitive
Scenario-neutral methods are being adopted more frequently in climate impact assessments, due in part to the uncertainty present in climate model projections. Scenario-neutral methods ‘stress test’ a system against a range of hydroclimate scenarios, which are formed by perturbing statistics of the climate variable time series—climate ‘attributes’. This allows the uncovering of system sensitivities and modes of failure irrespective of climate projections. However, it introduces another form of uncertainty: which climate attributes should a system be stress tested against?
Scenario-neutral studies typically consider changes in simple attributes, the most common being annual averages of hydroclimate variables. The complex interactions between hydroclimate and many systems means that the attributes a system is most sensitive to may not be investigated. Hence, solutions designed in response to uncovered modes of failure may not be appropriate. To address this limitation, we have developed an approach to select climate attributes that describe the most change in system performance. The approach begins by evaluating a comprehensive set of candidate attributes that may affect a system (including means, percentiles and measures of persistence). Variable selection techniques are then used to reduce the number of attributes down to those most critical for system performance. Applied to a case study with two objectives, the approach narrowed ten attributes to two sets of four (i.e. different for each objective) including non-standard attributes like the number of frost days and growing season length. This emphasises the need to strategically evaluate which climate attributes should be used in systems stress testing.
Stream and Session
Stream F: System Identification Approaches for Complex Environmental Systems
F3: Modelling and Decision Making Under Uncertainty