Prof. Yongkang Xue
University of California, USA
Room 1118, New Building, IAP
10am, Sept 12, 2018
SST variability has been shown to have predictive value for land precipitation and significant progress has been achieved in the SST research. Prediction skill for precipitation anomalies in spring and summer months, however, has remained stubbornly low for many years. For instance, during 2015-2016 one of the strongest El Nino events since 1950 was associated with an extraordinary Californian drought, while a 2016-2017 La Nina event has been associated with record rainfall that effectively ended the 5-year Californian drought, contrary to the expected SST - drought/flood relations. While atmospheric internal variability undoubtedly contributes, the sub-seasonal to seasonal (S2S) prediction problem underscores the need to seek explanations beyond SST’s influence alone, to pursue the identification of new mechanisms contributing to droughts/floods and add value for intraseasonal prediction. The GEWEX/GASS Initiative “Impact of initialized land temperature and snowpack on sub-seasonal to seasonal prediction (ILSTSS2S)” puts forward a new approach, which complements SST, snow, and soil moisture research by exploring the possible remote effects of large-scale spring land surface temperature/subsurface temperature (LST/SUBT) anomalies in geographical areas upstream on summer drought or flood- a concept that has been largely ignored by previous extreme hydroclimate events studies.
Preliminary studies have been carried out to explore the relationship between spring LST/SUBT anomalies and summer precipitation anomaly in downstream regions in North America and East Asia (Xue et al., 2016, 2018). Using the worldwide available 2-meter temperature (T2m), which has relatively reliable quality, and maximum covariance analysis method, we have identified that the warm (cold) spring T2m anomaly in western U.S. is associated with the summer wet (dry) conditions in Southern Great Plains and adjacent regions (SGP) and opposite anomaly to the north; it also suggests that warm (cold) spring T2m anomaly in Tibetan Plateau is associated with the summer wet (dry) conditions to the south of the Yangzi River and opposite anomaly to the north. The significant correlation between these two variables is comparable to the well-known SST and precipitation correlations.
To conform the causal relationship between spring LST anomaly and summer precipitation downstream in these two continents, we have conducted several modeling studies using both the General Circulation Model and the regional climate model, to investigate three high signal drought/flood events, which included the 2011 SGP drought, 2015 SGP flood, 2003 East Asian drought to the south of the Yangzi River and flood to the north. The spring T2m temperatures in 2011 and 2015 were very cold and warm respectively, and the Tibetan Plateau spring temperature was very cold in 2003. In these studies, observed T2m anomalies were used as constrain to assess the effect of spring warm (cold) land surface temperature on the summer downstream flood (drought). Meanwhile, for comparison, the SST effects were also tested. Our modeling studies demonstrate that the spring upstream land surface temperature anomalies contributed to the 2011 SGP drought and 2015 SGP flood, as well as 2003 East Asian drought/flood events. The LST contribution is the first order effect and comparable with the SST effect in North America and its effect seems even larger than SST in 2003 East Asian Case. As such, it is essential to include both factors (SST and LST) to produce reliable flood/drought prediction. Some issues/difficulties for this new approach will also be discussed.
The ILSTSS2S project, with participation of more than 30 institutions worldwide, will further examine the potential of the LST/SUBT effect in adding value to S2S prediction, especially extreme drought and flood events in different continents. East Asia is selected as the focus in the first phase because of the high elevation Tibetan Plateau and the exceptional quantity of available observational data, which is jointly supported by the Third Pole Experiment.