Presentation is loading. Please wait.

Presentation is loading. Please wait.

Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 1 Module.

Published byCecil Robertson Modified over 9 years ago

Similar presentations

Presentation on theme: "Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 1 Module."— Presentation transcript:

1 Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 1 Module 2.1 Monitoring activity data for forests using remote sensing Module developers: Frédéric Achard, European Commission (EC) - Joint Research Centre (JRC) Jukka Miettinen, EC - JRC Brice Mora, Wageningen University Exercise: Using Landsat time series data to derive forest area change estimates Sourcebook (2014) Box 3.2.2 V1, March 2015 Creative Commons License

2 Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 2 Mapping forest area change using multi- temporal segmentation and systematic sampling  The following slides will cover the main steps taken in the JRC-FAO Remote Sensing Survey image processing and interpretation approach which is based on a global systematic sample with a sample size of 10x10 km  Processing steps described in this example include: 1.Downloading and cropping of Landsat data 2.Radiometric calibration and cloud masking 3.Radiometric normalization 4.Image segmentation and classification 5.Visual validation and modification 6.Derivation of a change matrix

3 Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 3 Step 1 : Downloading and cropping of Landsat data  It is possible to assess the availability and download satellite data using e.g. the CEOS COVE tool (http://www.ceoscove.org/index.php/covetool/) or the USGS GLOVIS tool (http://glovis.usgs.gov)http://www.ceoscove.org/index.php/covetool/http://glovis.usgs.gov  Important points in image selection: 1. Overall quality of the image at the sample site location 2. Closeness to the desired acquisition time 3. The most suitable season

4 Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 4  In this exercise, we will use a sample site at N 00 E 102, in the humid tropical peatlands of Indonesia  The change analysis will cover the time period from 1990 to 2000 Step 1 : Downloading and cropping of Landsat data Landsat 5 TM for 1990 Landsat 7 ETM+ for 2000

5 Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 5 Step 2: Radiometric calibration and cloud masking  Sensor specific equations can be used to calibrate satellite data into Top of Atmosphere (ToA) reflectance; for Landsat, the details of this process can be found e.g. in Bodart et al. (2011)  Clouds and cloud shadows can be removed either manually or using automated cloud and cloud shadow detection algorithms Example of segment based automated cloud and cloud shadow removal relying on spectral values and sun illumination direction, used in the JRC-preprocessing chain

6 Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 6 Step 3: Radiometric normalization  For change detection, it is not necessary to convert the data into surface reflectance values  Relative normalization of multi-temporal data can be used to set the radiometric measurements to a common relative scale and ensure spectral comparability  So called Pseudo-Invariant Features (PIF, e.g. unchanged evergreen forest areas) can be used as reference areas  The images are normalized into the same level by comparing the average pixel values at the PIF’s and making necessary histogram shift for one of the images so that the PIF’s averages match in the two images

7 Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 7 Step 4a: Image Segmentation  The segmentation is created using both of images as input  This ensures that features from both images are captured Final image segments overlaid on both of the images

8 Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 8 Step 4b: Image classification  Followed by the segmentation, image classification is run separately in both of the images  A two step process is used: 1) Initial supervised algorithm utilizing signature database, and 2) aggregation of the initial segments into larger polygons (min 1 ha) utilizing a combination of the spectral and thematic information Classification results in the sample site for 1990 (left) and 2000 (right)

9 Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 9 Step 5: Visual validation and modification  In JRC, an in-house software has been produced for visual validation (Simonetti et al. 2011)  It enables a quick comparison of images to the classification result and allows corrections to be made  Furthermore the setup enables easy overlay of the two images, to better evaluate changes

10 Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 10 Step 6: Derivation of a change matrix  A change matrix between the two classifications can be created e.g. using image processing or GIS software  On our sample site Tree Cover was reduced by over 80% between 1990 and 2000, with the majority of the areas being converted into Other Wooded land and Other Land Cover (in this case oil palm plantation) Change matrix of the sample site in km 2 TC = Tree cover, TCM = Tree cover mosaic, OWL = other wooded land, OL = Other land cover, W = Water and C/S = Cloud/Shadow

11 Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 11 Recommended modules as follow up  Module 2.2 to proceed with monitoring activity data for forests remaining forests (incl. forest degradation)  Module 2.7 for exercises on accuracy assessment  Module 2.8 for overview and status of evolving technologies, including e.g. Radar data  Module 3 to learn more about REDD+ assessment and reporting

Download ppt "Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 1 Module."

Similar presentations

A Land Cover Map of Eurasia’s Boreal Ecosystems S. BARTALEV, A. S. BELWARD Institute for Environment and Sustainability, EC Joint Research Centre, Italy.

A Land Cover Map of Eurasia’s Boreal Ecosystems S. BARTALEV, A. S. BELWARD Institute for Environment and Sustainability, EC Joint Research Centre, Italy.
U.S. Department of the Interior U.S. Geological Survey USGS/EROS Data Center Global Land Cover Project – Experiences and Research Interests GLC2000-JRC.

U.S. Department of the Interior U.S. Geological Survey USGS/EROS Data Center Global Land Cover Project – Experiences and Research Interests GLC2000-JRC.
Xiangming Xiao Department of Botany and Microbiology, College of Arts and Sciences Center for Spatial Analysis, College of Atmospheric.

Xiangming Xiao Department of Botany and Microbiology, College of Arts and Sciences Center for Spatial Analysis, College of Atmospheric.
Toward Near Real Time Forest Fire Monitoring in Thailand Honda Kiyoshi and Veerachai Tanpipat Space Technology Applications and Research, School of Advanced.

Toward Near Real Time Forest Fire Monitoring in Thailand Honda Kiyoshi and Veerachai Tanpipat Space Technology Applications and Research, School of Advanced.
Algorithm Performance Evaluation Burnt surface area statistics compared to inventories/fire surveys

Algorithm Performance Evaluation Burnt surface area statistics compared to inventories/fire surveys
Major Operations of Digital Image Processing (DIP) Image Quality Assessment Radiometric Correction Geometric Correction Image Classification Introduction.

Major Operations of Digital Image Processing (DIP) Image Quality Assessment Radiometric Correction Geometric Correction Image Classification Introduction.
USGS Core Data Stream Report Session 7: Baseline Global Observation Scenario SDCG-7 Sydney, Australia March 4 th – 6 th 2015.

USGS Core Data Stream Report Session 7: Baseline Global Observation Scenario SDCG-7 Sydney, Australia March 4 th – 6 th 2015.
ReCover for REDD and sustainable forest management EU ReCover project: Remote sensing services to support REDD and sustainable forest management in Fiji.

ReCover for REDD and sustainable forest management EU ReCover project: Remote sensing services to support REDD and sustainable forest management in Fiji.
Module 2.2 Monitoring activity data for forests remaining forests (incl. forest degradation) REDD+ Sourcebook training materials by GOFC-GOLD, Wageningen.

Module 2.2 Monitoring activity data for forests remaining forests (incl. forest degradation) REDD+ Sourcebook training materials by GOFC-GOLD, Wageningen.
Module 2.2 Monitoring activity data for forests remaining forests (incl. forest degradation) REDD+ training materials by GOFC-GOLD, Wageningen University,

Module 2.2 Monitoring activity data for forests remaining forests (incl. forest degradation) REDD+ training materials by GOFC-GOLD, Wageningen University,
Module 2.2 Monitoring activity data for forests remaining forests (incl. forest degradation) REDD+ training materials by GOFC-GOLD, Wageningen University,

Module 2.2 Monitoring activity data for forests remaining forests (incl. forest degradation) REDD+ training materials by GOFC-GOLD, Wageningen University,
Module 2.2 Monitoring activity data for forests remaining forests (incl. forest degradation) REDD+ Sourcebook training materials by GOFC-GOLD, Wageningen.

Module 2.2 Monitoring activity data for forests remaining forests (incl. forest degradation) REDD+ Sourcebook training materials by GOFC-GOLD, Wageningen.
Module 2.5 Estimation of carbon emissions from deforestation and forest degradation REDD+ training materials by GOFC-GOLD, Wageningen University, World.

Module 2.5 Estimation of carbon emissions from deforestation and forest degradation REDD+ training materials by GOFC-GOLD, Wageningen University, World.
Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 1 Module.

Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 1 Module.
Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 1 Module.

Module 2.1 Monitoring activity data for forests using remote sensing REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 1 Module.
Module 2.8 Overview and status of evolving technologies REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 1 Module 2.8 Overview.

Module 2.8 Overview and status of evolving technologies REDD+ training materials by GOFC-GOLD, Wageningen University, World Bank FCPF 1 Module 2.8 Overview.
Module 2.1 Monitoring activity data for forests using remote sensing Module developers: Frédéric Achard, European Commission (EC)– Joint Research Centre.

Module 2.1 Monitoring activity data for forests using remote sensing Module developers: Frédéric Achard, European Commission (EC)– Joint Research Centre.
Validation of the GLC2000 products Philippe Mayaux.

Validation of the GLC2000 products Philippe Mayaux.
Module 1.3 Assessing and analyzing drivers of deforestation and forest degradation REDD+ training materials by GOFC-GOLD, Wageningen University, World.

Module 1.3 Assessing and analyzing drivers of deforestation and forest degradation REDD+ training materials by GOFC-GOLD, Wageningen University, World.
Module 1.3 Assessing and analyzing drivers of deforestation and forest degradation REDD+ training materials by GOFC-GOLD, Wageningen University, World.

Module 1.3 Assessing and analyzing drivers of deforestation and forest degradation REDD+ training materials by GOFC-GOLD, Wageningen University, World.

Similar presentations

Ads by Google