Application and potentials of RADAR and LiDAR technologies for forest carbon assessment in Pacific Island Countries

Size: px

Start display at page:

Download "Application and potentials of RADAR and LiDAR technologies for forest carbon assessment in Pacific Island Countries"

Steven Parks
5 years ago
Views:

1 Application and potentials of RADAR and LiDAR technologies for forest carbon assessment in Pacific Island Countries June 19th, 2012 PNGFA-JICA Workshop Masamichi HARAGUCHI (Kokusai Kogyo Co., Ltd.) Consultant for PNGFA/JICA&JICS PNGFA JICA & JICS

2 Table of Contents 1. Which data should we use? 2. What can we do with Radar/SAR? 3. How different, LiDAR and Radar? 4. What can we do with LiDAR? 5. What is GeoSAR? 6. Consideration to use LiDAR/GeoSAR 7. Summary PNGFA JICA & JICS 2

3 1. Which data should we use? Remote Sensing Monitoring: Multi Sensor/ Platform High Resolution Satellites (Optical) validation/verification for pilot study/modeling area Identify land features (training data) with detail Medium/Low Resolution Satellites (Optical) national/regional, rapid mapping for benchmark map periodical change analysis for reference level Radar/SAR Setellite annual change analysis/monitoring weather independent analysis/monitoring regular/emergency monitoring Aerial Photos & Airborne LiDAR/Radar validation/verification with interpretation 3D/volume estimation for sampling area Field Survey Inventory field survey for carbon stock estimation Define deforestation/degradation area with GPS. PNGFA JICA & JICS 3

4 1. Which data should we use? Remote Sensing Sensor: Comparison Table Type Satellite/Sensor Advantage Disadvantage Usage Low Resolution Mid Resolution MODIS LANDSAT Free, frequent (daily, seasonal change) Free, archive, wide coverage Difficult to identify vegetation boundary Limitation of interpretation/classify Fire Alert, Seasonal Change Analysis in the past ALOS/ PRISM&AVNIR2 Good panchro resolution, value added service Pan & MS are different sensor National level development High Resolution SPOT5 RapidEye Comparing with past, abundant archive Quick collection, Good MS resolution, RedEdge band Expensive (cmp. to ALOS), No blue band No Pan imagery, few archive imagery, no experiences, subdistributor system National level development /partly update Urgent/Short term development (new tasking) Very High Resolution (VHR) QuickBird/ WorldView GeoEye Possible to interpret tree kinds, village roads Expensive, impossible to cover national level Field survey complement, modeling validation Airborne Satellite Radar/SAR LiDAR Radar/IFSAR ALOS/ PALSAR DSM & DTM, high accuracy Regularly, assured collection, strong for change detection Hilly area or high density area Limitation for using in mountain area Contribution for tree stand volume Regularly change monitoring PNGFA JICA & JICS 4

5 1. Which data should we use? Satellite Positioning Map(Resolution/Swath) TerraSAR-X TamDEM-X COSMO-Skymed 1/2/3/4 GeoEye-1 WorldView-2 WorldView-1 QuickBird CARTOSAT 2 Pleiades1/2 IKONOS KOMPSAT-2 RADARSAT-2 FORMOSAT-2 ALOS/PRISM CARTOSAT 1 SPOT6/7 SPOT5 5.0 Resolution (m) 10.0 RADARSAT-1 SPOT1-4 Optical Satellite RapidEye1/2/3/4/5 ALOS/AVNIR-2 ALOS/PALSAR Radar Satellite PNGFA JICA & JICS 5

6 1. Which data should we use? Difference of Resolution: Satellite Imagery Ref: created by Mekong Map, Lao PNGFA JICA & JICS 6

7 Comparing Optical Satellite & Radar/SAR Satellite Covered by Clouds Penetrate Clouds! Optical Satellite (AVNIR- Radar/SAR Satellite 2) (PALSAR) SAR satellite can monitor in any weather/necessary Ref. JAXA Web time PNGFA JICA & JICS 7

8 Challenges of SAR Analysis: Steep Mountain SAR Satellite SAR: Synthetic Aperture Radar Layover Fore shortening Radar shadow Image of Backscatter Coefficient Local Incident Angle Layover, Radar shadow PNGFA JICA & JICS 8

9 Case Example of PNG: Plantation & Mangrove (Optical) PNGFA JICA & JICS 9

10 Case Example of PNG: Plantation & Mangrove (SAR Dual Pol.) PNGFA JICA & JICS 10

bright/grey) Specular Reflection (look dark/black) Forest

11 Detecting and Estimating Deforestation Area Change of Forest Area caused by logging in Amazon Volume Scattering (look bright/grey) Specular Reflection (look dark/black) Forest Deforest SAR satellite is robust for change detection PNGFA JICA & JICS 11

Applying for Landuse Change: Activity Data (2D)

Harvesting Abundant 2009/12/19 (FBS/HH) Slash

12 Applying for Landuse Change: Activity Data (2D) Crop Calendar Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Season Dry Rainy Dry Crop Calendar Logging Slash & Burn Planting Growing Harvesting Abundant 2009/12/19 (FBS/HH) Slash and Burn Land (SB) 2010/11/06 (FBD/HH) PNGFA JICA & JICS 12

13 Detecting Slash & Burn Shifting Cultivation (LAO case): 1 PALSAR(HH): R:2006/12/11 G:2007/12/14 PALSAR: Slash & Burn Shifting Cultivation Analysis B:2006/12/11 Assuming that purple (R+G) area was burned µ Location Pakxeng Phonxay km Slash_Burn_AOI PALSAR(HH) R: G: B: RGB Red: Band_1 Green: Band_2 Blue: Band_3 PNGFA JICA & JICS 13

Detecting Slash & Burn Shifting Cultivation (LAO case): 2 PALSAR(HH): R:2006/12/11 G:2007/12/14 PALSAR: Slash & Burn Shifting Cultivation Analysis B:2006/12/11 Assuming that purple (R+G)

14 Detecting Slash & Burn Shifting Cultivation (LAO case): 2 PALSAR(HH): R:2006/12/11 G:2007/12/14 PALSAR: Slash & Burn Shifting Cultivation Analysis B:2006/12/11 Assuming that purple (R+G) area was burned µ Location Pakxeng Phonxay km 2007_Slash_Burn Slash_Burn_AOI PALSAR(HH) R: G: B: RGB Red: Band_1 Green: Band_2 Blue: Band_3 PNGFA JICA & JICS 14

15 Detecting Slash & Burn Shifting Cultivation (LAO case): 3 PALSAR(HH): R:2007/12/14 G:2008/12/16 PALSAR: Slash & Burn Shifting Cultivation Analysis B:2007/12/14 Assuming that purple (R+G) area was burned µ Location Pakxeng Phonxay km Slash_Burn_AOI PALSAR(HH) R: G: B: RGB Red: Band_1 Green: Band_2 Blue: Band_3 PNGFA JICA & JICS 15

Detecting Slash & Burn Shifting Cultivation (LAO case): 4 PALSAR(HH): R:2007/12/14 G:2008/12/16 PALSAR: Slash & Burn Shifting Cultivation Analysis B:2007/12/14 Assuming that purple (R+G)

16 Detecting Slash & Burn Shifting Cultivation (LAO case): 4 PALSAR(HH): R:2007/12/14 G:2008/12/16 PALSAR: Slash & Burn Shifting Cultivation Analysis B:2007/12/14 Assuming that purple (R+G) area was burned µ Location Pakxeng Phonxay km 2008_Slash_Burn Slash_Burn_AOI PALSAR(HH) R: G: B: RGB Red: Band_1 Green: Band_2 Blue: Band_3 PNGFA JICA & JICS 16

area change Optical Satellite Inventory Survey Information on carbon stocks

survey for a large number/wide It is not easy to measure tree height accurately

17 What is necessary for REDD+ Monitoring? ACTIVITY DATA x EMISSION FACTOR = CO2 Emission Satellite Monitoring Area and area change Optical Satellite Inventory Survey Information on carbon stocks associated with each land category It is very difficult to implement field survey for a large number/wide It is not easy to measure tree height accurately We would like to some compensating method. Radar/SAR Satellite Airborne LiDAR/SAR LiDAR: Light Detection And Ranging PNGFA JICA & JICS 17

The Role of Urban Development Patterns in Mitigating the Effects of Tsunami Run-up: Final Report

J-RAPID Final Symposium Sendai, Japan The Role of Urban Development Patterns in Mitigating the Effects of Tsunami Run-up: Final Report March 6, 2013 Fumio Yamazaki, Chiba University, Japan and Ronald T.