CSDMS 2016 annual meeting poster DylanWard
A dynamical-statistical approach to forecasting regional glacier response to projected warming: an example from the Cordillera Real, Bolivia.
Zack Guido, Institute of the Environment, University of Arizona, Tucson Arizona, United States. zguido@email.arizona.edu
Abstract:
Alpine glaciers in many mountain ranges around the globe have been retreating in recent decades partially as a consequence of increasing temperatures. Small glaciers are particularly sensitive to climate changes and will likely disappear sooner than larger glaciers and levy more immediate impacts. In regions where the mass balance of glaciers is correlated with air temperature, estimating changes is more feasible because temperature projections are more confidently and robustly projected by global climate models (GCMs) than other variables. One major challenge facing the prediction of responses of small alpine glaciers to climate change is that their responses are highly variable, due in part to local aspect and microclimate effects, and in part because of their subdecadal response times to years with anomalous precipitation or temperature. For example, in the Cordillera Real, Bolivia, satellite imagery analyzed over 1985-2005 shows a systematic increase in ice loss with glacier size, but the scatter around this trend is marked, and makes prediction of glacier response based on climate model projections highly uncertain.
Here, we test a modeling approach to evaluate the collective response of these glaciers to a particular temperature timeseries by treating the glaciers of a given massif as an ensemble, and quantifying the variability around the temperature response due to local effects (e.g., aspect, precipitation variability), initial state of transience, and internal dynamics. Our approach couples a surface energy-mass balance model and dynamical flow model to simulate a group of glaciers on a real or synthetic topography, allowing glaciers to have different sizes and response times as simulated by the dynamical model. We compare a control run (glaciers in steady state at start of run) to a run with a starting condition forced with a randomly perturbed climate, so that the state of each glacier at the beginning of the study period is variable and dependent on its local climatic factors and internal dynamics. We then impose a temperature change drawn from reanalysis data during the 1985-2005 period covered by the satellite imagery, and evaluate the scatter of individual glacier response around that trend. We compare this variability to that evaluated from the satellite imagery to quantify the contributions due to internal dynamics and/or the transient state of glaciers prior to the study period. We then evaluate the magnitude of local climatic effects needed to explain any remaining variability.
This approach is in the early experimental stages, and we look forward to discussions and feedback from the CSDMS community.
* Please acknowledge the original contributors when you are using this material. If there are any copyright issues, please let us know and we will respond as soon as possible.
