BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Statistical Learning in Earth Monitoring
DTSTART:20121023T161500
DTEND:20121023T171500
DTSTAMP:20260506T133034Z
UID:b383768d5347dcf6d1e960452f1a6dfc99d52e863408e36ea3f28a47
CATEGORIES:Conferences - Seminars
DESCRIPTION:Professor Gustavo Camps-Valls\, Dept. of Electronics Engineeri
 ng\, Valencia University\, Spain\nThe last few hundred years\, human activ
 ities have precipitated an environmental crisis on Earth. Quantification o
 f the impact on the Earth's system are matter of current and intense resea
 rch. In the last decade\, advanced statistical methods have been introduce
 d to help monitoring our impact on the vegetation and atmosphere. This sem
 inar will concentrate on two recent efforts conducted in my lab to constra
 in the problem by using remote sensing data and advanced statistical infer
 ence models.\n\nSolar-induced chlorophyll fluorescence (ChF) provides one 
 of the most innovative sources of information to monitor plant activity fr
 om space. Accurate and robust estimation of leaf chlorophyll content is ne
 eded to calculate the ChF yield and to allow extrapolation at the landscap
 e scale. Vegetation indices typically become either too general or specifi
 c to particular scenarios\, and they under-exploit the wealth of the spect
 ral information. Conversely\, physically-based methods such as inverting a
  radiative transfer (RT) model\, require site-specific information for pro
 per model parametrization\, which is not always available. Alternatively\,
  non-parametric models provide generally higher accuracies\, and recent st
 atistical methods like Gaussian processes also provide information about t
 he uncertainty of the estimation\, which may lead to improved quantificati
 on of gross primary productivity (GPP). Applications in airborne data will
  illustrate the capabilities of the methods\, in preparation of the upcomi
 ng ESA Fluorescence Explorer (FLEX) mission.\n\nThe second case study will
  focus on the estimation of temperature\, water vapor\, and ozone concentr
 ation from spaceborne very high spectral resolution infrared sounding inst
 ruments. These sensors can be used to calculate the profiles of such atmos
 pheric parameters with unprecedented accuracy and vertical resolution. Des
 pite the constant advances in sensor design and retrieval techniques\, it 
 is not trivial to invert the full information of the atmospheric state con
 tained by such hyperspectral measurements. Numerical inversion of an RT mo
 del gives accurate results but shows numerical instability and high comput
 ationally burden. Advantageously\, statistical regression methods enable f
 ast retrievals from high volumes of data\, rank the most relevant input va
 riables and observations\, and provide confidence intervals for the predic
 tions. Evidence of the advantages given by such approaches in the upcoming
  Meteosat Third Generation Infrared Sounder (MTG-IRS) data will be given.
LOCATION:GR A3 32 http://plan.epfl.ch/?room=GR%20A3%2032
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR