Remote sensing for nature monitoring
This report is the result of a review on the possibilities of remote sensing for applications in the nature domain, with emphasis on Natura 2000 habitat monitoring. In recent years, enormous progress has been made in the availability and processing of high-resolution satellite and drone images. This increases the potential application for answering all kinds of policy and nature management questions. We demonstrate that remote sensing can have much added value for the monitoring of habitat distribution and habitat quality across a wide range of nature areas. We also demonstrate that higher spatial resolution of remotely sensed imagery often results in better classification accuracies. Deep learning techniques are also becoming popular since they are able to consider the contextual information and not only the spectral information from the imagery in classifying or identifying objects (from habitats to individual plant species). However, the amount of training data can have a large impact on classification accuracies, much more than for more conventional classification methods. This, then, requires a large investment in the collection of in-situ (field) data as well. Another finding is that including LiDAR and hyperspectral data can significantly improve detailed habitat mapping. In summary, the resource of remote sensing data and techniques should be selected depending on the relevant nature types, research questions and nature targets at a specific local, regional or national scale. It requires more communication between remote sensing researchers and ecologists. If nature goals and remote sensing technologies are brought together at an early stage, many applications will be possible. For the Netherlands, the remote sensing community should focus especially on monitoring the structure and function of habitat types. Also, such large-scale and long-term remote sensing monitoring should become part of a national nature monitoring programme.
Main Authors: | , , , , |
---|---|
Format: | External research report biblioteca |
Language: | English |
Published: |
Wageningen Environmental Research
|
Subjects: | Life Science, |
Online Access: | https://research.wur.nl/en/publications/remote-sensing-for-nature-monitoring |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
dig-wur-nl-wurpubs-617200 |
---|---|
record_format |
koha |
institution |
WUR NL |
collection |
DSpace |
country |
Países bajos |
countrycode |
NL |
component |
Bibliográfico |
access |
En linea |
databasecode |
dig-wur-nl |
tag |
biblioteca |
region |
Europa del Oeste |
libraryname |
WUR Library Netherlands |
language |
English |
topic |
Life Science Life Science |
spellingShingle |
Life Science Life Science Mücher, Sander Los, Stan Kramer, Henk Janssen, John Roerink, Gerbert Remote sensing for nature monitoring |
description |
This report is the result of a review on the possibilities of remote sensing for applications in the nature domain, with emphasis on Natura 2000 habitat monitoring. In recent years, enormous progress has been made in the availability and processing of high-resolution satellite and drone images. This increases the potential application for answering all kinds of policy and nature management questions. We demonstrate that remote sensing can have much added value for the monitoring of habitat distribution and habitat quality across a wide range of nature areas. We also demonstrate that higher spatial resolution of remotely sensed imagery often results in better classification accuracies. Deep learning techniques are also becoming popular since they are able to consider the contextual information and not only the spectral information from the imagery in classifying or identifying objects (from habitats to individual plant species). However, the amount of training data can have a large impact on classification accuracies, much more than for more conventional classification methods. This, then, requires a large investment in the collection of in-situ (field) data as well. Another finding is that including LiDAR and hyperspectral data can significantly improve detailed habitat mapping. In summary, the resource of remote sensing data and techniques should be selected depending on the relevant nature types, research questions and nature targets at a specific local, regional or national scale. It requires more communication between remote sensing researchers and ecologists. If nature goals and remote sensing technologies are brought together at an early stage, many applications will be possible. For the Netherlands, the remote sensing community should focus especially on monitoring the structure and function of habitat types. Also, such large-scale and long-term remote sensing monitoring should become part of a national nature monitoring programme. |
format |
External research report |
topic_facet |
Life Science |
author |
Mücher, Sander Los, Stan Kramer, Henk Janssen, John Roerink, Gerbert |
author_facet |
Mücher, Sander Los, Stan Kramer, Henk Janssen, John Roerink, Gerbert |
author_sort |
Mücher, Sander |
title |
Remote sensing for nature monitoring |
title_short |
Remote sensing for nature monitoring |
title_full |
Remote sensing for nature monitoring |
title_fullStr |
Remote sensing for nature monitoring |
title_full_unstemmed |
Remote sensing for nature monitoring |
title_sort |
remote sensing for nature monitoring |
publisher |
Wageningen Environmental Research |
url |
https://research.wur.nl/en/publications/remote-sensing-for-nature-monitoring |
work_keys_str_mv |
AT muchersander remotesensingfornaturemonitoring AT losstan remotesensingfornaturemonitoring AT kramerhenk remotesensingfornaturemonitoring AT janssenjohn remotesensingfornaturemonitoring AT roerinkgerbert remotesensingfornaturemonitoring |
_version_ |
1819141493725593600 |
spelling |
dig-wur-nl-wurpubs-6172002024-12-03 Mücher, Sander Los, Stan Kramer, Henk Janssen, John Roerink, Gerbert External research report Remote sensing for nature monitoring 2023 This report is the result of a review on the possibilities of remote sensing for applications in the nature domain, with emphasis on Natura 2000 habitat monitoring. In recent years, enormous progress has been made in the availability and processing of high-resolution satellite and drone images. This increases the potential application for answering all kinds of policy and nature management questions. We demonstrate that remote sensing can have much added value for the monitoring of habitat distribution and habitat quality across a wide range of nature areas. We also demonstrate that higher spatial resolution of remotely sensed imagery often results in better classification accuracies. Deep learning techniques are also becoming popular since they are able to consider the contextual information and not only the spectral information from the imagery in classifying or identifying objects (from habitats to individual plant species). However, the amount of training data can have a large impact on classification accuracies, much more than for more conventional classification methods. This, then, requires a large investment in the collection of in-situ (field) data as well. Another finding is that including LiDAR and hyperspectral data can significantly improve detailed habitat mapping. In summary, the resource of remote sensing data and techniques should be selected depending on the relevant nature types, research questions and nature targets at a specific local, regional or national scale. It requires more communication between remote sensing researchers and ecologists. If nature goals and remote sensing technologies are brought together at an early stage, many applications will be possible. For the Netherlands, the remote sensing community should focus especially on monitoring the structure and function of habitat types. Also, such large-scale and long-term remote sensing monitoring should become part of a national nature monitoring programme. Dit rapport is een verkenning naar de mogelijkheden van remote sensing voor natuur monitoring, met nadruk op Natura 2000 gebieden. In de beschikbaarheid en processing van hoge resolutie satelliet- en drone beelden is de afgelopen jaren een enorme vooruitgang geboekt, waardoor de toepassingsmogelijkheden voor allerlei beleids- en natuurbeheervraagstukken toenemen. We tonen aan dat remote sensing veel toegevoegde waarde kan hebben voor zowel de monitoring in de verspreiding van habitats als de habitatkwaliteit over een breed scala aan natuurgebieden. We tonen ook aan dat een hogere ruimtelijke resolutie van remote sensing beelden vaak resulteert in een betere classificatie nauwkeurigheid. Deep learning-technieken worden nu ook populair omdat ze ook de contextuele informatie in aanmerking nemen en niet alleen de spectrale informatie van de beelden voor het classificeren en/of identificeren van objecten (van habitat typen tot individuele plantensoorten). De hoeveelheid trainingsgegevens kan echter een grote invloed hebben op de classificatienauwkeurigheid, veel meer dan voor meer conventionele classificatie technieken. Het betekent dus ook een grote investering in het verzamelen van in-situ (veld)data. Een andere bevinding is dat het opnemen van LiDAR en hyperspectrale gegevens de gedetailleerde habitatkartering en monitoring aanzienlijk kan verbeteren. Samengevat, de bron van remote sensing data en technieken dient gekozen te worden afhankelijk van de relevante natuurtypen, onderzoeksvragen en natuurdoelen op een specifieke lokale, regionale of landelijke schaal. Het vereist meer communicatie tussen teledetectie onderzoekers en ecologen. Als de natuurdoelen en remote sensing-mogelijkheden in een vroeg stadium bij elkaar worden gebracht, zijn er veel toepassingen mogelijk. Voor Nederland zou de remote sensing gemeenschap zich vooral moeten richten op het monitoren van structuur en functie van habitattypen. En een grootschalig en langdurig remote sensing monitoring programma zou onderdeel moeten worden van een nationaal natuurmonitoringprogramma. en Wageningen Environmental Research application/pdf https://research.wur.nl/en/publications/remote-sensing-for-nature-monitoring 10.18174/633129 https://edepot.wur.nl/633129 Life Science (c) open_access_other Wageningen University & Research |