Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Home » Collaborative Research: Analysis of IPCC-AR5 and CFS model simulated stratosphere-troposphere coupling and its link to Eurasian snow cover variability

Collaborative Research: Analysis of IPCC-AR5 and CFS model simulated stratosphere-troposphere coupling and its link to Eurasian snow cover variability

The goals of the proposed research are to (a) analyze the output from IPCC AR5 class of models and the NCEP Climate Forecast System (CFS) hindcasts to assess the impact of stratosphere-troposphere (ST) coupling and Eurasian snow variability on the winter climate of North America, (b) further quantify atmospheric predictability associated with interannual snow variability using an operational prediction model, and (c) work with the operational community to improve climate forecasts from the intraseasonal to the interannual time scales by incorporating the predictive potential of snow variability into operational practices. The proposed work specifically addresses the goal of CPPA “to evaluate the ability of the IPCC-AR5 class models to simulate and predict ISI climate” and “improve understanding of climate predictability at ISI time scales focusing on stratosphere-troposphere coupling, land memory effects and weather-climate links.”

Skillful climate predictions throughout the extratropics remain a challenge for both statistical and numerical models. For the winter season, ST coupling is now understood to play an important role in winter surface anomalies, especially for those that persist for longer than synoptic time-scales and therefore are important for determining seasonal means. Furthermore, a statistically significant link has been demonstrated between Eurasian snow cover extent and major ST coupling events. Snow cover anomalies often lead ST coupling events by two to three months, making snow cover a potential predictor of winter climate anomalies. Previous analysis has shown that the two regions where snow cover has the highest potential for skillful prediction are East Asia and the eastern United States; the latter will be the focus of this proposal.

Our proposed research will focus on diagnosing model output and performing additional experiments designed to study ST coupling and its link to snow cover variability. Our initial analysis will study archived atmospheric state variables from control (i.e., preindustrial) and time-evolving GHG (i.e., climate of 20th Century and climate projection) GCM experiments from the IPCC AR4 and AR5 class of models. We will then compare the analysis from the IPCC AR4/AR5 models with a similar analysis using existing hindcast output from the CFS model. We will diagnose correlations between simulated snow cover variability and atmospheric temperatures, geopotential heights, winds and energy flux (Eliassen-Palm flux or the three dimensional wave activity flux) and compare with observed co-variability between snow cover and the atmosphere. This analysis will (a) assess the simulation of ST coupling in the IPCC AR5 class of models and in the CFS model compared against the observations, (b) quantify the influence of snow variability on ST coupling and climate variability over North America in the IPCC AR5 models and from the CFS hindcasts by comparing with the observations, and (c) provide an assessment of how much improvement in the simulation of the pathway by which snow anomalies influence climate variability over North America has been accomplished by comparing the analysis from the IPCC AR5 models and the CFS with a similar analysis performed from the IPCC AR4 class of models. Demonstrating improved simulations of ST coupling in the CFS model will enhance credibility of the CFS forecasts.

We will also carry out additional experiments in an effort to assess predictability of atmospheric anomalies associated with the interannual snow variability using an operational prediction model. Experiments will be designed to force the CFS with observed snow cover variability. Model output of atmospheric response to observed snow cover forcing will be analyzed and compared with observations and with archived CFS data where snow cover has not been prescribed. In a second set of GCM experiments we will identify and isolate large ST coupling events. Then we will re-run the model with initialized atmospheric conditions that preceded the stratosphere-troposphere coupling events but with varying amounts of snow cover. This set of experiments will help us determine whether the modeled ST coupling events were altered or modified by changes in snow cover extent.

A reasonable concern is that the CFS model cannot adequately simulate observed snowatmosphere coupling and therefore snow experiments with the CFS will only yield negative results. However, for future model development it is critical to understand and document model errors and deficiencies in order to spur future model improvements, and we are only proposing enough model analysis and experiments to determine the CFS capabilities in regard to this important coupling. Whether realistic snow cover improves seasonal prediction or not, the results will be shared with the operational modeling community. Results from the analysis of archived model data and original GCM experiments will be submitted for publication.

Scroll to Top