The Advances in Targeted Observations for Tropical Cyclone Prediction Based on Conditional Nonlinear Optimal Perturbation (CNOP) Method - Data Assimilation for Atmospheric, Oceanic and Hydrologic Applications - page 568

filter (EnKF, Hamill and Snyder 2002 ) and the ensemble transform Kalman filter
(ETKF, Bishop et al. 2001 ). The strategies mentioned have been tested in field
experiments such as the Fronts and Atlantic Storm-Track Experiment (FASTEX;
Snyder 1996 ; Joly et al. 1997 ), the North Pacific Experiment (NORPEX; Langland
et al. 1999a ), the Winter Storm Reconnaissance Programs (WSR; Szunyogh
et al. 2000 , 2002 ), the Dropwindsonde Observations for Typhoon Surveillance near
the Taiwan region (DOTSTAR; Wu et al. 2005 ), the Atlantic THORPEX Regional
Campaign (ATReC; Rabier et al. 2008 ), etc. Forecasts are generally improved
by assimilation of targeted observations ( Gelaro et al. 1999 ; Langland 2005 ; Wu
et al. 2005 ; Buizza et al. 2007 ; Rabier et al. 2008 ).
The strategies mentioned above are generally linear methods. They are
constrained by linear approximations. To study the effect of nonlinearity, Mu
et al. ( 2003 ) proposed a novel approach of conditional nonlinear optimal
perturbation (CNOP). The CNOP is an extension of the linear singular vector (SV)
method in to the nonlinear regime, and it has been applied to some research fields
such as El Nino-Southern Oscillation (ENSO) predictability ( Mu et al. 2007 ;
Duan and Mu 2009 ; Duan and Luo 2010 ; Duan and Luo 2010 ; Peng et al. 2011 ),
the nonlinear behavior of baroclinic unstable flows ( Riviere et al. 2008 ), ensemble
forecasting ( Mu and Jiang 2008 ), and the transitions between multiple equilibria
states of the ecosystem ( Sun and Mu 2009 ).
Recently, Mu et al. ( 2009 ) suggested that the CNOP can be used to identify
the sensitive areas for tropical cyclone predictions in targeted observations since
the forecasts benefit more from reductions of CNOP-type initial errors than from
reductions of SV-type initial errors. Then Zhou and Mu ( 2011 , 2012a ,b)used
the CNOP to identify the sensitive areas, and studied the properties of the CNOP
sensitive areas with respect to variations of the horizontal resolution, the verification
area design and the optimization time period. Furthermore, Chen and Mu ( 2012 )
carried out sensitivity analysis by studying the impact of initial errors introduced
into the CNOP sensitive areas on the forecasts. Moreover, Qin ( 2010a , b )and
Qin and Mu ( 2011a , b ) performed the observing system simulation experiments
(OSSEs) to assess whether the sensitive areas identified by CNOP can be considered
as dropping sites in realtime targeting. The observation system experiments (OSEs)
using the DOTSTAR Data have also been carried out by Chen ( 2011 ) to demonstrate
the utility of the CNOP method.
This chapter will summarize the above works about the approach of the CNOP to
targeted observations for tropical cyclone predictions. The SV sensitive areas have
also been studied correspondingly for comparison in some works. The structure
of this paper is as follows. Section 24.2 provides an introduction to CNOP and
SV, Sect. 24.3 introduces the properties of the CNOP sensitive areas. Section 24.4
describes the serviceability of the CNOP sensitive areas by OSSEs and OSEs.
A brief summary and discussion are given in the final section.