Applications of Climate Regionalization: Statistical Prediction and Patterns of Precipitation Variability in Observations and Global Climate Models

Climate Regionalization;Statistical Predicition;Spatial Analysis;Spatio-temporal;Variability;Africa;Precipitation;Global Climate Models;Climate Change;GCMs;RCPs;GHGs;Earth & Planetary Sciences

Badr, Hamada S.Zaitchik, Benjamin F. ;

Johns Hopkins University

关键词: Climate Regionalization;    Statistical Predicition;    Spatial Analysis;    Spatio-temporal;    Variability;    Africa;    Precipitation;    Global Climate Models;    Climate Change;    GCMs;    RCPs;    GHGs;    Earth & Planetary Sciences;   

Others  :  https://jscholarship.library.jhu.edu/bitstream/handle/1774.2/40268/BADR-DISSERTATION-2016.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: JOHNS HOPKINS DSpace Repository

【 摘 要 】

Rainfall in the tropical Africa is prone to large temporal and spatial variability. The well-documented social impacts of this variability have motivated numerous efforts at seasonal precipitation prediction, many of which employ statistical techniques that forecast precipitation as a function of large-scale climate indices. These statistical models have demonstrated some skill, but nearly all have adopted conventional statistical modeling techniques—most commonly generalized linear models—to associate predictor fields with precipitation anomalies. Here, a new approach is proposed to identify regional patterns of precipitation variability based on objective climate regionalization to understand the physical patterns and drivers of spatio-temporal variability in the regional scale as well as ties to global patterns. Then, advanced statistical models were applied to provide skillful predictions as well as to identify large-scale drivers and capture nonlinear influences that multiple major modes of variability have on each region.This raises questions regarding the spatial extent and regional connectivity of changes inferred from observations, proxies, and/or derived from climate models. Objective regionalization offers a tool for addressing these questions. To demonstrate this potential, Hierarchical Climate Regionalization (HiClimR) package in R was developed and applied to regionalize Africa using observations and Global Climate Model (GCM) historical simulations and future projections. HiClimR was utilized for the regionalization of Africa based on interannual precipitation variability using monthly Climate Hazards Group InfraRed Precipitation with Stations (CHIRPS) data for the period 1981-2014. A number of data processing techniques and clustering algorithms were tested in order to ensure a robust definition of climate regions. The final regionalization results highlight the seasonal and even month-to-month specificity of regional climate associations across the continent, emphasizing the need to consider time of year as well as research question when defining a coherent region for climate analysis. CHIRPS regions were then compared to those of five GCMs for the historic period. Focusing on boreal summer, we note key similarities and differences between CHIRPS and GCMs, including discussion of the coherence of the Sahel in observation and some, but not all, models. This serves as a basis for examining the representation of teleconnections in GCMs relative to observations. It also provides a foundation for interpreting GCM results in cases where the model’s climate regions do not align with observation. Finally, we examine shifts in climate regions under projected 21st century climate change for different GCMs and emissions pathways. We note a projected change in the coherence of the Sahel, in which the western and eastern Sahel become distinct regions with different teleconnections. This pattern is most pronounced in high emissions scenarios.