The Climate Forecast System (CFS), the fully coupled ocean–land–atmosphere dynamical seasonal prediction system, which became operational at NCEP in August 2004, is described and evaluated in this paper. The CFS provides important advances in operational seasonal prediction on a number of fronts. For the first time in the history of U.S. operational seasonal prediction, a dynamical modeling system has demonstrated a level of skill in forecasting U.S. surface temperature and precipitation that is comparable to the skill of the statistical methods used by the NCEP Climate Prediction Center (CPC). This represents a significant improvement over the previous dynamical modeling system used at NCEP. Furthermore, the skill provided by the CFS spatially and temporally complements the skill provided by the statistical tools. The availability of a dynamical modeling tool with demonstrated skill should result in overall improvement in the operational seasonal forecasts produced by CPC. The atmospheric component of the CFS is a lower-resolution version of the Global Forecast System (GFS) that was the operational global weather prediction model at NCEP during 2003. The ocean component is the GFDL Modular Ocean Model version 3 (MOM3). There are several important improvements inherent in the new CFS relative to the previous dynamical forecast system. These include (i) the atmosphere–ocean coupling spans almost all of the globe (as opposed to the tropical Pacific only); (ii) the CFS is a fully coupled modeling system with no flux correction (as opposed to the previous uncoupled “tier-2” system, which employed multiple bias and flux corrections); and (iii) a set of fully coupled retrospective forecasts covering a 24-yr period (1981–2004), with 15 forecasts per calendar month out to nine months into the future, have been produced with the CFS. These 24 years of fully coupled retrospective forecasts are of paramount importance to the proper calibration (bias correction) of subsequent operational seasonal forecasts. They provide a meaningful a priori estimate of model skill that is critical in determining the utility of the real-time dynamical forecast in the operational framework. The retrospective dataset also provides a wealth of information for researchers to study interactive atmosphere–land–ocean processes.

No abstract

Land‐atmosphere feedback, by which precipitation‐induced soil moisture anomalies affect subsequent precipitation, may be an important element of Earth's climate system, but its very existence has never been demonstrated conclusively at regional to continental scales. Evidence for the feedback is sought in a 50‐yearobservational precipitation dataset covering the United States. The precipitation variance and autocorrelation fields are characterized by features that agree (in structure, though not in magnitude) with those produced by an atmospheric general circulation model (AGCM). Because the model‐generated features are known to result from land‐atmosphere feedback alone, the observed features are suggestive of the existence of feedback in nature.

A B S T R A C TA local quasi-geostrophic energetics analysis indicates that within the jet core, low-frequency (LF) eddies behave baroclinically essentially the same as high-frequency (HF) eddies. They both have a westward tilting vertical structure and both grow baroclinically by transporting heat poleward and by converting eddy potential energy to kinetic energy. However, the difference in the horizontal orientations of HF and LF eddies has several important implications to their amplitude and peak locations, as well as their interaction with stationary waves. The barotropic decay of meridionally elongated HF eddies tends to terminate the growth of HF eddies beyond the jet exit region. The barotropic growth of the zonally elongated LF eddies not only ensure a continuous growth of LF eddies in the jet exit region, but also results in a new baroclinic growth of LF eddies farther downstream due to the presence of the west-east temperature contrast associated with stationary waves. The continuous growth of LF eddies due to both barotropic and baroclinic processes in the jet exit region is consistent with the facts that LF eddies reach maximum variability farther downstream of the two major jet streams and that the LF variability is much stronger than HF eddies.The results of energetics analysis are confirmed by the feedback analysis, showing that HF eddies, being dominated by meridional orientations, mainly act to maintain (damp) stationary waves by locally enhancing (reducing) north-south gradient of the height (temperature) field near the jet core regions. The zonally elongated LF eddies, on the other hand, act to primarily reduce the zonal gradient associated with stationary waves both barotropically and baroclinically.

The National Weather Service intends to begin routinely issuing long-lead forecasts of 3-month mean U. S. temperature and precipitation by the beginning of 1995. The ability to produce useful forecasts for certain seasons and regions at projection times of up to 1 yr is attributed to advances in data observing and processing, computercapability, and physical understanding-particularly, for tropical ocean-atmosphere phenomena. Because much of the skill of the forecasts comes from anomalies of tropical SST related to ENSO, we highlight here long-lead forecasts of the tropical Pacific SST itself, which have higher skill than the U.S forecasts that are made largely on their basis. The performance of five ENSO prediction systems is examined: Two are dynamical [the Cane-Zebiak simple coupled model of Lamont-Doherty Earth Observatory and the nonsimple coupled model of the National Centersfor Environmental Prediction (NCEP)]; one is a hybrid coupled model (the Scripps Institution for Oceanogra-phy-Max Planck Institute for Meteorology system with a full ocean general circulation model and a statistical atmosphere); and two are statistical (canonical correlation analysis and constructed analogs, used at the Climate Prediction Center of NCEP). With increasing physical understanding, dynamically based forecasts have the potential to become more skillful than purely statistical ones. Currently , however, the two approaches deliver roughly equally skillful forecasts, and the simplest model performs about as well as the more comprehensive models. At a lead time of 6 months (defined here as the time between the end of the latest observed period and the beginning of the predictand period), the SST forecasts have an overall correlation skill in the 0.60s for 1982-93, which easily outperforms persistence and is regarded as useful. Skill for extra-tropical surface climate is this high only in limited regions for certain seasons. Both types of forecasts are not much better than local higher-order autoregressive controls. However, continual progress is being made in understanding relations among global oceanic and atmospheric climate-scale anomaly fields. It is important that more real-time forecasts be made before we rush to judgement. Performance in the real-time setting is the "Climate Prediction Center NCEP/NWS/NOAA ultimate test of the utility of a long-lead forecast. The National Weather Service's plan to implement new operational long-lead seasonal forecast products demonstrates its effectiveness in identifying and transferring "cutting edge" technologies from theory to applications. This could not have been accomplished without close ties with, and the active cooperation of, the academic and research communities.