3742 IEICE TRANS. COMMUN., VOL.E88–B, NO.9 SEPTEMBER 2005 PAPER On the Inversion of Wind Scatterometer Incomplete Data Sets Maurizio MIGLIACCIO † a) and Maurizio SARTI †† , Nonmembers SUMMARY ERS-1/2 wind scatterometers (WSC) data sets may con- tain cells in which two, instead of three, measurements are available. These data sets are called incomplete. Operational inversion procedures discard such data sets and therefore no wind field is estimated. This is very lim- iting in semi-closed seas such as the Mediterranean Sea. In this paper we propose a new inversion procedure capable to retrieving near surface wind fields from incomplete data sets. This procedure is an enhancement of an algorithm already proposed and tested by the authors. A set of comprehen- sive experiments are presented and discussed. It is shown that the inversion procedure gets to remarkable results even in presence of a large number of doublets. key words: scatterometer, sea wind, wind field reconstruction 1. Introduction Near surface sea wind field knowledge is fundamental. For instance, it is a key element to force circulation models [1], to track tropical cyclones [2] and to detect oil spills [3]. In general, accurate wind field knowledge facilitates rou- tine climatological/meteorological forecasting, supports de- partments of environmental protection to best deal with life- threatening weather situations and accidents due to human activities. Nowdays it is well established that high quality wind field measurements can be obtained by means of remotely sensed data. Although, some very recent innovative tech- niques, based on polarimetric radiometers [4] and SAR mea- surements [5], have been proposed to estimate the near sur- face wind field, the fundamental technique relies on scat- terometer measurements. In fact, at present time, physical and technological aspects suggest that wind field estimated by means of scatterometer measurements are much more re- liable [6]. In addition, scatterometers allow a global cover- age. In this paper, an enhancement of the new inversion pro- cedure described in [7], [8], in order to take into account in- complete data sets, is presented. The ERS-1/2 C-band fan- beam VV-polarized scatterometers, also simply known as wind scatterometers (WSC), share the active measurement instrument (AMI) unit with the Synthetic Aperture Radar Manuscript received September 6, 2004. Manuscript revised March 31, 2005. † The author is with the Dipartimento per le Tecnologie, Uni- versit` a degli Studi di Napoli Parthenope, Via Acton 38, 80133 Napoli, Italy. †† The author is with the Consiglio Nazionale delle Ricerche, IREA, Via Diocleziano 328, 80124 Napoli, Italy. a) E-mail: maurizio.migliaccio@uniparthenope.it DOI: 10.1093/ietcom/e88–b.9.3742 (SAR) [9]. The switching from the SAR to the scatterom- eter makes that in some scatterometer cells the number of available measurements degrade from three (triplets) to two (doublets) [9], [10]. This fact is of particular relevance in the Mediterranean Sea, an almost closed sea in which large islands are present, and operating mode makes the active measurement unit switch from the SAR to the scatterometer when lands are left or reached [9], [10]. Since the presence of doublets is very unlikely over the oceans, space agen- cies inversion procedures do not consider such incomplete data sets and therefore no wind field estimation is provided [11]. Further, wind field retrieval from incomplete data sets is meant to be of low quality. All this has suggested to enhance the new scatterometer pointwise inversion procedure developed by the authors to embody such a feature. In summary, the new inversion procedure presented and tested in this paper is able to deal with incomplete data sets ∗ , i.e. scatterometer data sets in which doublets are present. Such data sets are not processed by the inversion procedures currently used by the European Space Agency (ESA) and national space agencies, e.g. [11]. With respect to the Italian Space Agency (ASI) Processing and Archiv- ing Facility (I-PAF) inversion procedure, it must be noted that the reference procedure [10] is also able to deal with incomplete data sets but the ambiguity removal step is less effective [7]. Actual implementation of [10] at ASI I-PAF is a version which considers only scatterometer data sets with- out doublets. In practice, there is no operational inversion procedure which consider incomplete data sets. On a scientific viewpoint, it must be noted that in liter- ature only [10] addresses the problem of retrieving the wind field from incomplete scatterometer data sets. However in [10] only few experiments, based on simulated data, are pre- sented. In this paper a systematic, severe study of the inver- sion procedure, which includes simulated and real data, is accomplished. In conclusion, in this paper, the new inversion proce- dure, enhanced to process incomplete data sets, is detailed and tested. A set of experiments based on simulated data and real data are presented and discussed. They show that it is possible to retrieve the wind field with remarkable ac- ∗ Note that semantically an incomplete data set is also a data set in which are present singlets (cells with a single measurement) and cells with no measurements. These cases are of no practical interest because theory tells that even in ideal conditions no wind field can be retrieved [12], [13]. Copyright c  2005 The Institute of Electronics, Information and Communication Engineers