Evaluation of Sub-Selection Methods for Assessing Climate Change Impacts on Low-Flow and Hydrological Drought Conditions
- PDF / 2,255,691 Bytes
- 21 Pages / 439.37 x 666.142 pts Page_size
- 46 Downloads / 190 Views
Evaluation of Sub-Selection Methods for Assessing Climate Change Impacts on Low-Flow and Hydrological Drought Conditions Saeed Golian 1
& Conor Murphy
1
Received: 1 May 2020 / Accepted: 8 November 2020/ # Springer Nature B.V. 2020
Abstract
A challenge for climate impact studies is the identification of a sub-set of climate model projections from the many typically available. Sub-selection has potential benefits, including making large datasets more meaningful and uncovering underlying relationships. We examine the ability of seven sub-selection methods to capture low flow and drought characteristics simulated from a large ensemble of climate models for two catchments. Methods include Multi-Cluster Feature Selection (MCFS), Unsupervised Discriminative Features Selection (UDFS), Diversity-Induced Self-Representation (DISR), Laplacian score (LScore), Structure Preserving Unsupervised Feature Selection (SPUFS), Non-convex Regularized Self-Representation (NRSR) and Katsavounidis–Kuo–Zhang (KKZ). We find that sub-selection methods perform differently in capturing varying aspects of the parent ensemble, i.e. median, lower or upper bounds. They also vary in their effectiveness by catchment, flow metric and season, making it very difficult to identify a best sub-selection method for widespread application. Rather, researchers need to carefully judge sub-selection performance based on the aims of their study, the needs of adaptation decision making and flow metrics of interest, on a catchment by catchment basis. Keywords Climate change . General circulation models (GCMs) . Subs-selection . Uncertainty . Drought
* Saeed Golian [email protected]
1
Irish Climate Analysis and Research UnitS (ICARUS), Department of Geography, Maynooth University, Maynooth, Co. Kildare, Ireland
Golian S., Murphy C.
1 Introduction General Circulation Models (GCMs) are essential for studying changes in the climate system and informing adaptation. However, the unknown trajectory of future emissions, differences in the sensitivity of GCMs to anthropogenic forcing and the chaotic nature of the climate system, mean model projections are subject to much uncertainty (Knutti et al., 2010). Best practice dictates that impact analyses adequately account for uncertainty (Clark et al., 2016). Hence output from large-scale model experiments – comprising simulations from multi-model and/or perturbed physics ensembles – e.g. CMIP (Coupled Model Intercomparison Project; Taylor et al., 2012), CORDEX (Giorgi et al., 2009) and climateprediction.net (Stainforth et al., 2005), represent an invaluable resource. While large ensembles allow better investigation of climate risk, such datasets are assembled on the basis of opportunity; consequently suffering from a lack of model independence and biased representation of constituent uncertainties (Pirtle et al. 2010; Knutti et al., 2010; Masson and Knutti, 2011; Mendlik and Gobiet, 2016). Furthermore, in integrated assessments where additional stressors (e.g. land-use change, socio-economic scenarios) must be con
Data Loading...
