17:15 〜 18:45
[ACG36-P08] Retrieving Foliage Clumping Index from Geostationary Himawari Satellite Observations
キーワード：Clumping Index (CI)、Himawari静止軌道衛星、BRDF フィット、ホットスポットとダークスポットの正規化差分指数（NDHD）
The foliage Clumping Index (CI), accounts for the nonrandom spatial distribution of the foliage elements in the canopy. CI is crucial for accurate Leaf Area Index (LAI) and Gross Primary Production (GPP) estimations, defined as the ratio of effective leaf area index (LAIe) to LAI . Traditional remote sensing methods for CI calculation, mainly relying on low-orbit satellites and the Normalized Difference Hotspot and Darkspot (NDHD) index, often overlook the potential of geostationary satellites. An Enhanced RPV BRDF (E-RPV) model with Himawari satellite land surface reflectance data are proposed in this study, demonstrating greater accuracy than MODIS-based products in CI calculation as for it has been proved as a new opportunity in previous research that geostationary satellite can observe hot-spots well.
As for the method, angle data were sourced from the JAXA Himawari P-Tree system to identify hot-spots, while land surface reflectance data were obtained from Ichi Lab., Chiba University. The E-RPV BRDF model simulated the principal reflection of the Himawari satellite to determine hot-spot and dark-spot reflectance. The NDHD method defines CI differently for coniferous forests and other vegetation classes, based on reflectance values at hot-spots and dark-spots.
Comparisons with in-situ data at three in-situ sites of VALERI net, revealed that Himawari-based CI estimations are more accurate and exhibit more clustered trends for single vegetation types than MODIS-based products. By comparing the average CI obtained from other in-situ data and remote sensing data, it was found that the CI of MODIS fluctuates greatly and is not close to the in-situ data. Despite these promising results, challenges remain, particularly in effectively removing cloud data and how to determine hot-spots under non-observed conditions for model simulation. Future research will also focus on comparing more in-situ and remote sensing-based CI results to further verify the accuracy of the proposed method.
In conclusion, the Himawari satellite presents a significant opportunity for improved CI estimation, especially with effective hot-spot observation. However, issues such as cloud removal, hot-spot restoration under unobserved conditions, and determining accuracy remain to be studied in subsequent studies.
As for the method, angle data were sourced from the JAXA Himawari P-Tree system to identify hot-spots, while land surface reflectance data were obtained from Ichi Lab., Chiba University. The E-RPV BRDF model simulated the principal reflection of the Himawari satellite to determine hot-spot and dark-spot reflectance. The NDHD method defines CI differently for coniferous forests and other vegetation classes, based on reflectance values at hot-spots and dark-spots.
Comparisons with in-situ data at three in-situ sites of VALERI net, revealed that Himawari-based CI estimations are more accurate and exhibit more clustered trends for single vegetation types than MODIS-based products. By comparing the average CI obtained from other in-situ data and remote sensing data, it was found that the CI of MODIS fluctuates greatly and is not close to the in-situ data. Despite these promising results, challenges remain, particularly in effectively removing cloud data and how to determine hot-spots under non-observed conditions for model simulation. Future research will also focus on comparing more in-situ and remote sensing-based CI results to further verify the accuracy of the proposed method.
In conclusion, the Himawari satellite presents a significant opportunity for improved CI estimation, especially with effective hot-spot observation. However, issues such as cloud removal, hot-spot restoration under unobserved conditions, and determining accuracy remain to be studied in subsequent studies.