The primary objective of this project is to use signals in the historical climate record to make credible climate projections for the next 2-20 years. The basis for extrapolation is improvements to traditional “climate normals” (historical averages of local weather conditions), which are commonly applied in planning, design, and management situations where weather and climate are relevant. Traditional climate normals are based on 30-years of data and assume a stationary climate. In contrast, this project will focus on the development and application of new (alternative) climate normals that explicitly address the general non-stationarity of the climate system. This research works to address the need to utilize recent historical signals to update the non-stationarity of the climate signal.
The alternative normals will be derived using methods that exploit the statistical properties (e.g., short-term persistence, trends, cyclical behavior) of historical time series. A suite of complementary methods will be considered, ranging from the use of shorter averaging periods to data-adaptive techniques that accommodate non-stationary processes. In addition to the traditional 30-year averages, the methods will include N-year normals (averages for less than 30 years), optimal climate normals (which determine the “best” number of years for averages), hinge fit normals (which account for discontinuities in historical trends), empirical mode decomposition normals (which account for more complex cyclical behavior), and global climate normals (which tie local normals to global temperature). The methods are designed to improve skill without the need to understand the physical basis of ongoing changes, although some methods (e.g., global climate normals) have explanatory capabilities. For each method, the project will quantify the extent to which the normals help planners anticipate climatic conditions over the next 2-20 years. This work is organized around the following hypotheses:
The emphasis of this work is on the delivery of climatic information that is decision-relevant in a Department of Defense (DoD) context and that supports climate impacts and adaption goals articulated in the 2014 Department of Defense Climate Change Adaptation Roadmap. In that regard, the project will address a representative suite of 10 climate variables, including monthly mean climate conditions, 1% highest and lowest daily temperatures, annual average extreme minimum temperature, heating and cooling degree days, and 1% highest wet bulb globe temperature. The effort will focus on 10 DoD facilities that represent a spectrum of climate conditions across the conterminous United States, including the five largest bases in the country. Success will justify a follow-on effort that includes a larger suite of climatic variables and the global range of DoD installations.