Exploring Climate Patterns Embedded in Global Climate Change Datasets

This paper aims to explore climate patterns embedded in Global Climate Change Datasets generated by the Community Climate System Model (CCSM) for the 4th Assessment Report of the Intergovernmental Report for Climate Change. A Geographic Information Systems (GIS) approach is taken to examine spatiotemporal correlates among climate processes and to draw insights into their relationships and responses to climate change scenarios. With foci on the Southern Oscillation (SO) and precipitation anomalies at mid latitudes, the research explores three CCSM cases from worst-case, average, and conservative CO2 emission scenarios, respectively. The scenarios used are A2: continental economic regions dominate with an economic focus (representing a worst case scenario), A1b: a global economy with and economic focus and balanced energy usage (representing an average scenario), B1: a global economy with an environmental focus (representing a conservative scenario). The model and scenarios used are for the 4th Assessment Report of the Intergovernmental Panel on Climate Change. GIS methods are being developed to reveal the differences in spatiotemporal relationships between SO pressure anomalies and mid-latitude precipitation anomalies.

Central to the GIS methods being developed is a temporal GIS framework with abilities to represent and analyze geographic dynamics. The research expands upon process-based GIS approach to incorporate representational and analytical needs to characterize spatiotemporal behavior and relationships of pressure and precipitation anomalies. Events of pressure and precipitation anomalies will be modeled as GIS data objects with spatial and temporal dimensions. A set of measures will be developed to quantify shape, growth, movement, internal structure, life cycle, and other spatiotemporal characteristics of these data objects of anomalies. In addition to object characterization, these measures serve as the basis for similarity query and correlation analysis. Similarity query facilitates retrieval of anomaly events that exhibit similar spatiotemporal behaviors, such as similar SO anomalies or similar precipitation anomalies in the three test cases. “An anomaly tree” based on similarity measures can suggest the effects of CO2 emission scenarios on pressure and precipitation distributions, respectively. Complementarily, correlation analysis relates precipitation anomalies and SO pressure anomalies to reveal their spatial and temporal correspondence in answering questions like where, when, and how precipitation anomalies echo SO pressure anomalies. With similarity assessment and correlation analysis, much spatiotemporal information can be revealed from the wealth of CCSM Global Climate Change Datasets.