Remote mapping of soil erosion risk in Iceland
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
Title |
| |
Title of Series | ||
Number of Parts | 351 | |
Author | ||
License | CC Attribution 3.0 Unported: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor. | |
Identifiers | 10.5446/68911 (DOI) | |
Publisher | ||
Release Date | ||
Language | ||
Production Year | 2022 |
Content Metadata
Subject Area | ||
Genre | ||
Abstract |
| |
Keywords |
FOSS4G Firenze 2022331 / 351
1
7
13
22
25
31
33
36
39
41
43
44
46
52
53
55
58
59
60
76
80
93
98
104
108
127
128
133
135
141
142
143
150
151
168
173
176
178
190
196
200
201
202
204
211
219
225
226
236
242
251
258
263
270
284
285
292
00:00
Multiplication signComputer animation
00:17
Slide ruleIntegrated development environmentConservation lawDecision theoryInsertion lossOrganic computingPoint cloudDirected setPressure volume diagramDegree (graph theory)Computer-generated imageryGeometryFunction (mathematics)Kernel (computing)Musical ensembleBasis <Mathematik>Image resolutionPrice indexPixelSoftwareoutputRepresentation (politics)Latent heatMatter waveSet (mathematics)PreprocessorDigital signalEndliche ModelltheorieStatistical dispersionStandard deviationPredictionAreaProcess (computing)Observational studyCategory of beingDialectPoint cloudWave packetMatter waveMusical ensembleOrder (biology)Support vector machineLevel (video gaming)PredictabilityAreaSet theoryMedical imagingInsertion lossLimit (category theory)Covering spacePrice indexValidity (statistics)ResultantAuditory maskingFinite-state machineFocus (optics)Roundness (object)Set (mathematics)Flow separationComputer animation
Transcript: English(auto-generated)
00:00
This is Daniel from Fletan. We are a startup company based on Iceland dedicated to studying soil erosion risk through remote sensing methods. These are our five team members. Soil erosion has become a global degradation threat in recent years. It leads to desertification
00:25
and loss of biodiversity and in particular for the Arctic regions it can also lead to the release of organic carbon that has been stored under the permafrost for millions of years and that is one of the foreseen main contributors to climate change in the future. Also the Arctic comes
00:47
with its own set of challenges because it's a happily accessible place. It's frequently covered in snow and clouds so remote sensing becomes difficult and the methodologies that
01:00
have proven to be successful in other regions of the world cannot be straightforwardly generalized to the Arctic because of its peculiar soils and the fact that the vegetation covers are not not a good indicator. The methodology that we are following so far uses a support vector
01:23
machine SVM classifier. We split the data that we have from 80-20 into training and validation sets. The ground truth data has been provided by Land Grazland, the soil conservation agency
01:40
of Iceland. So we have about a 10,000 km2 of land for which we have both the classification that has been done by this agency and on the other hand we have Sentinel-2 images that can be compared and we can train on that. These images they need to be pre-processed
02:06
first, they need to be geo-referenced and atmospherically corrected. They can in 12 available wavelength bands that we can combine in order to create indices and we also need to develop new masks adapted for the Arctic region
02:24
such as for things like legend and permafrost. We also use digital elevation models in order to take into account the fact that soil erosion happens more quickly in steep slopes and that really applies for the Arctic very importantly as well. Now with this the results that we
02:47
are obtaining, they show quite a large overall accuracy so we are very confident on our results. Also certain features that can be seen or you can recognize rivers and hilltops and
03:05
we can already extract a few conclusions such as the fact that loss of our ability it's a common limitation of these methods and it is being a limitation in our case so that needs to be addressed and it needs to be improved.
03:22
We may be able to expand what we are doing in order to cover not only the region for which we have the ground truth data but we can extend it and provide a soil erosion map for all of Iceland and also taking into account that these methods can be applied, that we have
03:43
yearly data, we may be able to provide prediction for the future and so far in that way indicate which areas need a more specific focus in the future. And with that I thank you for your attention.