Using the “blue spike” to characterize biomass‐burning sites during Southern African Regional Science Initiative (SAFARI) 2000
2004
[1] During several flights of the ER-2 while participating in the Southern African Regional Science Initiative (SAFARI 2000), the University of Wisconsin–Madison's Scanning High Resolution Interferometer Sounder (S-HIS) obtained spectra containing isolated fires within its field of view (FOV). These fire-laden FOVs contain a spectral feature caused by rotational hot band transitions of CO2 near 2400 cm−1. Because of its location on the blue side of the 4.3 μm band of CO2, this feature is commonly referred to as the “blue spike.” Using this feature, we detected fires on four flights: 24 and 27 August and 6 and 7 September 2000. Fire locations are further verified by the ER-2 pilot's flight logs and elevated brightness temperatures in the thermal detectors of the MODIS Airborne Simulator (MAS) also on board the ER-2. Using line-by-line radiative transfer calculations (Genln2) with corrections for a fire's extreme high temperatures (HiTemp), we model S-HIS spectra for various scenes: background (cool surface and cool atmosphere), smoldering (warm surface and cool atmosphere), hot gas layer (cool surface and warm atmosphere), and fire (hot surface and hot atmosphere) cases. Using the controlled burn in the Timbavati Game Reserve on 7 September 2000 as a test case, we spectrally modeled the blue spike feature seen in the spectra obtained by S-HIS while the ER-2 flew over the fire. For this case, we found that ∼4.12 ± 0.05% of the FOV contained the hot gas layer while ∼0.23 ± 0.05% was actively burning. Originally viewed as a straightforward task of using the blue spike to characterize the fire temperature and size (fraction of S-HIS FOV), our analysis shows that numerous variables, including amount of carbon dioxide, amount of water vapor, and the temperature near the fire, play significant roles in the blue spike's shape and spectral position.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
20
References
1
Citations
NaN
KQI