20110728 IGARSS 2011 Vancouver Canada Relationships between PALSAR
2011/07/28: IGARSS 2011, Vancouver, Canada. Relationships between PALSAR backscattering data and forest above ground biomass in Japan ○ Takeshi Motohka (Japan Aerospace Exploration Agency) Masanobu Shimada (Japan Aerospace Exploration Agency) Osamu Isoguchi (Remote Sensing Technology Center of Japan) Masae I. Ishihara (Japan Wildlife Research Center) Satoshi N. Suzuki (Japan Wildlife Research Center)
Outline - Background - Data - In situ biomass data - PALSAR yearly mosaic data - Results - Relationship between biomass and PALSAR data - Mapping forest biomass - Summary and future works
Background Forest biomass is a key parameter to assess ü Emission of greenhouse gasses (CO 2, CH 4, etc. ) ü Accumulated carbon in forests ü Biodiversity A large-scale, time-series, globally consistent biomass monitoring is important for various projects such as REDD+.
Biomass monitoring by PALSAR ALOS-2 PALSAR-2 ALOS PALSAR Phased Array type L-band Synthetic Aperture Radar - Microwave backscatter shows high correlation with forest tree biomass especially for longer wavelength (i. e. L-band, P-band, …). - Spatial (10 – 100 m) and temporal (46 days) resolution meet the REDD+ or FCT methodologies. - Well calibrated global datasets for 5 years (2006~) - ALOS mission was ended in 2011, but next ALOS-2 (PALSAR-2) will be launched in 2013.
Many studies have revealed the relationships between forest biomass and PALSAR data at various regions and forest types. Australia Africa Lucas et al. , 2011 Indonesia Mitchard et al. , 2009 Englhart et al. , 2011
Purpose of the study • Target = Japan • Investigating the relationships between PALSAR backscattering data and above ground biomass of Japanese forests • Testing the retrieval of forest biomass using the obtained empirical relationships and PALSAR yearly mosaic data
In situ Biomass Data “Monitoring Site 1000” project by the Ministry of Environment of Japan (since 2003) • Network of long-term research sites for biodiversity assessment • 49 tree census sites • Located at various forest types • Only natural forests were selected in the study (not including artificial forests) Website: http: //www. biodic. go. jp/moni 1000/
Deciduous broadleaf forest (Tomakomai, Hokkaido) Evergreen coniferous forest (Otanomousu-daira, Nagano) Deciduous broadleaf forest (Chichibu, Saitama) Evergreen broadleaf forest (Yona, Okinawa)
Processing of tree census data Tree diameter of breast height [cm] (DBH) measured all trees in about 1 ha plot except for DBH < about 5 cm. Allometric equations for each specie Dry weight [kg] Ʃ (dry weight) / stand-size Biomass [t/ha]
Statistics of the forest stands (n=44) DBH: Diameter at Breast Height
PALSAR yearly mosaic R: HH G: HV B: HH/HV Year: 2007, 2009 Mode: Fine beam dual (HH, HV) Mosaicking period: Jun. - Sep. Pixel sampling: 10 m Orbit: Ascending
Generation of PALSAR yearly mosaics - Long-strip processing - Ortho-rectification - Slope-correction - Mosaicking Shimada & Otaki (2011); Shimada (2011) in “IEEE JSTAR special issue on Kyoto and Carbon Initiative” Converting DN to gamma naught γ 0 [d. B] = 10 log〈DN 2〉- 83 15 x 15 pixels averaging
PALSAR γ 0 vs. biomass HV RMSE: 0. 703 [d. B] Saturation level: - dy/dx = 0. 01 … 91 [t/ha] - dy/dx = 0. 005 … 182 [t/ha]
PALSAR γ 0 vs. biomass HH RMSE: 1. 053 [d. B] Saturation level: - dy/dx = 0. 01 … 68 [t/ha] - dy/dx = 0. 005 … 136 [t/ha]
Effect of slope-correction RMSE: 1. 053 [d. B] RMSE: 2. 312 [d. B] Red ●: After correction Green +: Before correction RMSE 0. 703 [d. B] RMSE 2. 073 [d. B]
Precipitation vs. PALSAR gamma naught rainfall (over 10 mm during 3 days before obs. ) HH HV RMSE: HH: 0. 670 d. B HV: 0. 402 d. B Mean bias between rainy and non-rainy data: HH: +0. 177 d. B HV: +0. 044 d. B
Inversion of forest biomass RMSE: 106. 23 t/ha %RMSE: 39. 3 % (=RMSE/mean) PALSAR HV γ 0 (15 x 15 pix average) Biomass map Water, Lay-over, Shadowing mask Urban area mask
Above ground biomass (t/ha) 0 100 200 300 400~
Above ground biomass (t/ha) Cropland 0 Cropland Wetland 100 200 300 400~
Above ground biomass (t/ha) 0 1 km (c) Google Earth 100 200 300 400~
Summary • We examined the relationships between PALSAR backscattering data and forest biomass in Japan. ü HV polarization was better to use (low RMSE and high saturation level). ü Slope correction was very important to reduce the error especially in mountainous regions. ü More data points were needed to investigate the difference among vegetation types. Airborne Li. DAR can be good solution of this.
Summary • Simple inversion of forest biomass ü Spatial pattern seems to be good. ü Problems: accuracy and saturation ü Possible solution: • Additional SAR analysis (multi-temporal data, full-polarimetric data, etc…) • Data fusion with ALOS/PRISM DSM and ALOS/AVNIR-2 (10 -m res. VIS&NIR) data • Data fusion with ICESAT/GLAS data • More in-situ data points and more evaluation
- Slides: 23