BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.osgeo.org//DT7LLU
BEGIN:VTIMEZONE
TZID:GMT
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:GMT
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T020000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:BST
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-foss4g-it-2023-DT7LLU@talks.osgeo.org
DTSTART;TZID=GMT:20230613T170000
DTEND;TZID=GMT:20230613T171500
DESCRIPTION:In the past\, the scarcity of hyperspectral Earth Observation (
 EO) data hindered the development of operational applications based on suc
 h technology. Considering the current increasing availability of this kind
  of data (e.g.\, PRISMA\, EnMap)\, that it is expected to further grow in 
 the future (e.g.\, Copernicus CHIME\, PRISMA Second Generation)\, it is im
 portant to evaluate the potential retained by hyperspectral remote sensing
  for EO applications that could provide operational services in the next f
 ew years. Within this context\, this work was conceived to perform a preli
 minary investigation of the capabilities of the PRISMA hyperspectral senso
 r for burned area (BA) mapping in an operational context (e.g.\, civil pro
 tection applications).\n \nOne of the most common approaches used for BA m
 apping via EO data is based on the Differenced Normalized Burn Ratio (dNBR
 ) index\, which detects the fire-induces alterations to vegetation and soi
 ls by taking advantage of the spectral information acquired in the Near In
 fraRed (NIR: 0.7-1.2 µm) and Short-Wave InfraRed (SWIR: 1.2-2.5 µm) band
 s of two images: one acquired before the fire event\, one after [1]. Multi
 spectral imagery commonly used for performing BA mapping for operational a
 pplications (e.g.\, Sentinel 2\, Landsat) have specific NIR and SWIR bands
  that can be used for dNBR computation [2]. Hyperspectral images\, instead
 \, allow for several bands combinations of data acquired in the NIR and SW
 IR spectral regions\, thereby generating numerous (and\, in some cases\, s
 lightly) different definitions of dNBR maps. Amongst these bands’ combin
 ations\, the more reliable ones shall be identified (i.e.\, the ones capab
 le of producing BA maps more accurate). At the same time – since the dNB
 R is also sensible to non-fire induced spectral alterations [1] – the le
 ss reliable ones shall be avoided.\n\nThe aim of this study was to set up 
 an experiment in which it was prototyped an automatic methodology of opera
 tional BA mapping based on PRISMA Level2D products (i.e.\, orthorectified\
 , surface reflectance imagery\; GSD: 30 m). The wildfire that occurred in 
 Pantelleria Island (Italy) on 17/08/2022 was used as a case study. For thi
 s event\, there were available two PRISMA images acquired on 06/08/2022 (p
 re-event) and 16/07/2022 (post-event). An ancillary shapefile produced by 
 the Copernicus Emergency Management Service (EMS) and representing the ext
 ent of the BA on 19/08/2022 (ca. 28 ha) was used as a reference layer to v
 alidate the analysis results. \n\nThe methodology that was set up – conc
 eptually similar to the one developed by [2] – produced more than 7600 d
 NBR maps (obtained from the combinations of the PRISMA NIR and SWIR spectr
 al bands)\, from which the pixels corresponding to the BA were mapped by u
 sing the Otsu approach for automatic threshold selection. The analysis was
  carried out over the whole Pantelleria Island territory\, where water bod
 ies\, clouds and clouds’ shadows were masked out (as well as poor qualit
 y PRISMA bands). Then\, the accuracy of the classification was quantified 
 (as a percentage) by means of the Dice Coefficient (DC) [3]\, which was ca
 lculated by using the Copernicus EMS reference BA layer. According to the 
 DC\, the best bands combination for mapping the BA of the Pantelleria 2022
  wildfire corresponds to the 0.903 (NIR) and 2.253 µm (SWIR) wavelengths.
  The DC associated with this BA map was 89.4%.\n\nIn an operational contex
 t\, ancillary information (i.e.\, BA reference layers) are often not avail
 able to identify the most reliable bands for BA mapping. Therefore\, an im
 age-based selection criterion useful to achieve this objective shall be us
 ed. Indeed\, for every NIR/SWIR bands combination used during the analysis
 \, the spectral separability [3] of the pixels classified as BA – from t
 he neighbouring ones classified as not BA – was computed. Then\, the ban
 ds combination characterized by the highest separability value was used fo
 r identifying the best dNBR map to use for BA mapping. For this specific e
 xercise\, this combination corresponds to the 1.038 µm (NIR) and 2.245 µ
 m (SWIR) wavelengths. The DC associated with this BA map was 88.8%. This v
 alue is very similar to the one identified via the ancillary reference BA 
 layer.\n\nThe details of the methodology will be presented at the conferen
 ce\, where the analysis results will be also thoroughly discussed.\n\nRefe
 rences:\n\n[1] van Gerrevink M.J. & Veraverbeke S. (2021). Evaluating the 
 Hyperspectral Sensitivity of the Differenced Normalized Burn Ratio for Ass
 essing Fire Severity. Remote Sensing. 13(22):4611.\n\n[2] Pulvirenti L. et
  al. (2023). Near real-time generation of a country-level burned area data
 base for Italy from Sentinel-2 data and active fire detections. Remote Sen
 sing Applications: Society and Environment. 29.\n\n[3] Roteta E. et al. (2
 019). Development of a Sentinel‐2 burned area algorithm: Generation of a
  small fire database for sub‐Saharan Africa. Remote Sensing of Environme
 nt. 222\, 1–17.
DTSTAMP:20260316T150856Z
LOCATION:Sala Videoconferenza @ PoliBa
SUMMARY:A Preliminary Investigation of the PRISMA Hyperspectral Sensor Pote
 ntial for Burned Area Mapping in an Operational Context - Luca Cenci\, Luc
 a Pulvirenti\, Giuseppe Squicciarino
URL:https://talks.osgeo.org/foss4g-it-2023/talk/DT7LLU/
END:VEVENT
END:VCALENDAR