The objective of this SEED Statement of Need (SON) was to provide credible and relevant climate information at moderate time scales two to 20 years into the future. Such information addressed decision needs of Department of Defense (DoD) installation managers within the 20-year timeframe.

Specific research needs are listed below. Proposers clearly stated which aspect of the research need will be addressed during the scope of the SEED effort and what criteria was met to propose a subsequent, multi-year project.

1. Characterizing uncertainty: Approaches are needed that can better describe and quantify contributions to climate uncertainty over the two- to 20-year timeframe. Research is required to determine both those components of the climate system that provide meaningful information for decision-making and those components that are the primary sources of uncertainty. Moreover, research should aim to ascertain the limits of predictability and enable a priori estimates of the confidence and reliability of moderate time scale climate projections.
2. Modes of climate variability: Certain large spatial scale climate patterns, such as the El Niño Southern Oscillation (ENSO) and its teleconnections at longer time scales, will have their own particular responses to climate change with respect to changes in the intensity, frequency, or duration of these patterns. Within the next 20 years, ENSO and other large-scale patterns may undergo substantial changes that can strengthen or weaken regional patterns, such as seasonal or extreme precipitation or sea-level responses, for which a predictive understanding of general trends over moderate time scales of two to 20 years may be important. Modeling and statistical approaches are needed to inform resiliency planning based on what is known about the variability of, and potential changes in, climate patterns and particularly changes in the intensity, frequency, or duration of extreme events. Such information in turn can better constrain potential future climate impacts over moderate time scales.
3. Regional trends with strong directionality: Some aspects of climate change (for example, temperature) have strong directionality but uncertain magnitude and variability, whereas for 2 other aspects (such as precipitation) even projecting directionality is problematic. Research is needed to understand how observed associated regional trends are mechanistically linked to identify causality and to develop reliable guidance on likely future conditions. Approaches are needed that can incorporate what is known about regional trends in climate-related variables and their interactions, and with what confidence, that can better constrain potential future conditions over moderate time scales. Moreover, of additional interest is research that addresses combinations of climate-related variables and their timing (such as increases in summer precipitation when evapotranspiration is higher) that may affect the directionality of environmental variables (for example, soil moisture) that result from these interactions.
4. Using “recent” historical signals to update the non-stationarity of the climate signal: Research is needed to (a) determine if changes in the base state of climate—either in the mean or variance or both—can be discerned and defensibly extracted from the “noise” of natural variability and (b) whether a scientific basis exists for extrapolating potential changes in the climate base state forward in a credible manner over an appropriate time period. Diagnostic approaches are needed to address the preceding through the quantitative analyses of the relatively short observation record and climate reanalyses, as well as by analyzing the statistics from ensembles of sufficiently long model simulations of preindustrial and current climate.
5. Frameworks for using climate information in the two- to 20-year timeframe: Given that advances can be made for providing defensible climate change information over moderate time scales, frameworks and protocols are needed to accompany such information to ensure the information and its use and limitations are placed into an appropriate context that end users can access and understand.

Proposals submitted in response to this SON addressed one or more of the research needs listed above. Proposals explicitly discuss the methods, outcomes, and metrics for success needed to demonstrate the viability of a particular line of inquiry upon which a successful SEED project is eligible to propose additional work.

Information related to climate change often is provided as a long-term response of the climate system to different levels of anthropogenic forcing. In addition, emission scenario-dependent projections via General Circulation Models (GCM) generally are provided as time slices and usually for mid-twenty-first century and end of century. Moreover, because of natural variation in the climate system and its predominance in characterizing the uncertainty of climate projections even up to mid-century, projections show little divergence up until that time.

Although understanding the long-term implications of climate change is critical, scientific studies and synthesis reports such as the Third National Climate Assessment have documented that climate is already changing and is not just a long-term issue. As a result, decision-makers, including those within the DoD, are now aware they must prepare for and respond to climate change. They require information about how it affects their decisions not just mid- and end of century, but within the more near-term timeframes in which they typically operate.

The focus of this SON is on the two- to 20-year time horizon. The two-year lower bound avoids overlap with those efforts attempting to extend weather prediction capabilities into at least the seasonal realm in which an underlying climate trend signal is not a focus. The 20-year upper bound 3 reflects the decision horizon for those that are not interested in long-term infrastructure planning or facility siting issues, but for which infrastructure maintenance and operational decisions may still be significantly affected. These decisions explicitly involve infrastructure with a planned performance of less than 20 years, changes in flood risk that may change operational performance or response action relative to extant infrastructure, and or narrowing of operational opportunities. This upper bound still falls well short of the time when divergence in driving emission or concentration pathway scenarios becomes important.

Some research efforts are underway in this space and proposals in response to this SON should not duplicate those efforts. For example, Coupled Model Intercomparison Project, Phase 5 (CMIP5) experiments include research related to decadal prediction using the GCMs with a focus on better representing model initial conditions that may persist over time. The National Earth System Prediction Capability project is focused currently on research questions in the near-term (i.e., less than two years). This SON is intended to explore other approaches that may inform decisions in the two- to 20-year timeframe.

Complementary SERDP-Funded Projects: SERDP is supporting several projects relating to potential non-stationary changes in the hydrologic cycle and coastal water level extremes that in part are investigating moderate time scales.

To meet the objectives of this SEED SON, proposals did not exceed $200,000 in total cost and approximately one year in duration. Work performed under the SEED SON investigated innovative approaches that entail high technical risk and/or have minimal supporting data. At the conclusion of the project, sufficient data and analysis is available to provide risk reduction and/or a proof-of-concept. SEED projects are eligible for follow-on funding if they result in a successful initial project.