BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.staging.osgeo.org//foss4g-it-2023//talk//SGWTNN
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-SGWTNN@talks.staging.osgeo.org
DTSTART;TZID=GMT:20230612T163000
DTEND;TZID=GMT:20230612T164500
DESCRIPTION:Glaciers are critical elements in the Earth’s climate system\
 , and can be considered as sensitive indicators of climate change. Glacier
 s store significant amounts of freshwater\, which is essential for animal 
 and human consumption and activities like industry and agriculture. Furthe
 rmore\, glaciers have a significant impact on the hydrological cycle\, and
  their melting also contributes to rising sea levels. Understanding and mo
 nitoring glacier extent changes is critical to informing climate policies\
 , assessing natural hazards and safeguarding global water resources. Nowad
 ays\, remote sensing technology is a proved and widely adopted source of i
 nformation in this sense.\nIn this context\, the proposed study aims to de
 velop a regression model able to predict future changes in glacier extent\
 , using supervised machine learning algorithms applied to open access medi
 um and HR spatial resolution satellite data of the EU Copernicus programme
 . To achieve this objective\, two machine learning models are developed. T
 he first model is a segmentation model that employs a U-Net architecture\,
  along with a final Conditional Random Field (CRF) module\, to digitalize 
 glaciers features from satellite images. The purpose of the segmentation m
 odel is to vastly expand the dataset required by the regression model\, in
  terms of glacier surface values. In fact\, this work presents an addition
 al contribution in the form of a novel dataset consisting of time series o
 f glaciers and snow extent. This dataset is generated using the best-perfo
 rming segmentation model previously trained\, applied to multiple glaciers
 \, spanning a 30-year period and a consistent seasonal interval. To train 
 the segmentation model\, and to create the required ground truth images\, 
 the GLIMS initiative database is used again\, while optical satellite imag
 es are obtained in part from Sentinel-2 data and in part from other public
 ly available datasets such as the "Hindu Kush Himalayas (HKH) glacier mapp
 ing dataset". The latter couples annotated glacier locations\, which were 
 produced by experts\, with multispectral imagery from Landsat 7.\nThe seco
 nd model is a multivariate regression model that seeks to identify the rel
 ationships between Land Surface Temperature (LST) and glacier/snow extent.
  \nIn order to train the models\, two datasets are required. For the regre
 ssion model and specifically LST\, data from the Sentinel-3 SLSTR instrume
 nt\, as well as data from the ESA Climate Change Initiative\, which consol
 idates data from various satellites over the past 25 years\, are utilized.
  Historical data on glacier extent and elevation is obtained from the "Gla
 ciers elevation and mass change data from 1850 to present from the Fluctua
 tions of Glaciers" database by the Copernicus Climate Change Service and d
 atasets provided by the Global Land Ice Measurements from Space (GLIMS) in
 itiative.  Finally\, both models are validated on testing data to assess t
 heir generalization capabilities and their performance on real-world cases
 . A subset of the segmentation dataset is kept aside to extrapolate metric
 s such as the Intersection-Over-Union (IoU)\, which allows to assess the a
 ccuracy of the results obtained and to make comparison with other architec
 tures. For the regression model\, error metrics such as the Root-Mean Squa
 red Error (RMSE) are considered to assess the model performance. The resul
 ts of the study are expected to provide insights that will enhance the mon
 itoring efforts of glacial features and provide useful information about t
 he impact of climate change on glaciers worldwide.
DTSTAMP:20260429T060441Z
LOCATION:Sala Biblioteca @ PoliBa
SUMMARY:Assessing Glacier Extent Changes through Machine Learning Algorithm
 s and Remote Sensing Data - Vanina Fissore\, Lorenza Ranaldi\, Davide Lisi
 \, Piero Boccardo\, Alessandro La Rocca\, Mirko Frigerio\, Daniele Sanmart
 ino
URL:https://talks.staging.osgeo.org/foss4g-it-2023/talk/SGWTNN/
END:VEVENT
END:VCALENDAR