Observation impact over the southern polar area during the Concordiasi field campaign

Abstract

The impact of observations on analysis uncertainty and forecast performance was investigated for austral spring 2010 over the southern polar area for four different systems (NRL, GMAO, ECMWF and Météo-France) at the time of the Concordiasi field experiment. The largest multi-model variance in 500 hPa height analyses is found in the southern sub-Antarctic oceanic region, where there are rapidly evolving weather systems, rapid forecast-error growth, and fewer upper-air wind observation data to constrain the analyses. The total impact of all observations on the model forecast was computed using the 24 h forecast sensitivity-to-observations diagnostic. Observation types that contribute most to the reduction of the forecast error are shown to be AMSU, IASI, AIRS, GPS-RO, radiosonde, surface and atmospheric motion vector observations. For sounding data, radiosondes and dropsondes, one can note a large impact on the analysis and forecasts of temperature at low levels and a large impact of wind at high levels. Observing system experiments using the Concordiasi dropsondes show a large impact of the observations over the Antarctic plateau extending to lower latitudes with the forecast range, with the largest impact around 50-70 • S. These experiments indicate there is a potential benefit from using radiance data better over land and sea-ice and from innovative atmospheric motion vectors obtained from a combination of various satellites to fill the current data gaps and improve numerical weather prediction analyses in this region.

Advancing Polar Prediction Capabilities on Daily to Seasonal Time Scales

“…With improved technology and power systems, the barrier is becoming more of a financial one than a logistical one: improved observations of the polar regions are possible, but are they worth the cost? To answer this, observing system experiments (OSEs) are required [see, e.g., Boullot et al (2016)], in which specific observations are withheld (denied) during the data assimilation process, with a particular focus on user requirements for these regions. To carry out these experiments, a sustained observing period is required with significantly enhanced spatial and temporal coverage-a Year of Polar Prediction (see below).…”

Aspects of ECMWF model performance in polar areas

“…The predictive skill when integrated over Southern Hemisphere is largely driven by a latitude band of very high synoptic activity located over the southern oceans near 50∘S. This is an area where analysis increments are large, indicating significant and fast growing model errors (Boullot et al , ). The daily variability of predictive skill is substantial for both poles (not shown) and the seasonal skill cycle is most pronounced for the South Pole, where the errors are lowest during Southern Hemisphere summer.…”

“…Wei et al () and Boullot et al () have identified significant differences between global NWP model analyses in terms of both mean analysis state and so‐called analysis activity, which is the variability of the analysis anomalies calculated from the difference between daily analyses and the analysis climate. Mean state and variability have clear implications for both forecast initialization and verification.…”

“…The apparent difference between analyses means that the consistency between these analyses can no longer be an indication of analysis accuracy as assumed by Jung and Leutbecher (2007) (see section 1). Wei et al (2010) and Boullot et al (2014) have identified significant differences between global NWP model analyses in terms of both mean analysis state and so-called analysis activity, which is the variability of the analysis anomalies calculated from the difference between daily analyses and the analysis climate. Mean state and variability have clear implications for both forecast initialization and verification.…”

Editorial for the Quarterly Journal's special issue on Polar Prediction

“…Consequently, observations from passive and active microwave instruments and infrared spectrometers, and analysis techniques that promise better sensing of the shallow lower polar troposphere, are important . Also here, much can be gained from assessing reanalyses and observational campaigns targeting the poles such as Concordiasi (Boullot et al, 2016).…”