Title: MODIS Aqua PSF Point Spread Function
1MODIS Aqua PSF (Point Spread Function)
- Gerhard Meister, OBPG
- NASA Goddard Space Flight Center
- (Futuretech Corp.)
- January 4, 2006
-
-
2Background
- Ocean color remote sensing imposes very stringent
requirements on radiometric accuracy of bands
relative to each other. - Stray light effects are very large for bright
clouds next to dark ocean water. - Theoretical solution correction with
point-spread function (PSF). - Problems PSF not well characterized, light
source intensity sometimes unknown,
computationally very expensive.
3PSF definition
Lm (i0,j0) Sij PSF(i- i0,j- j0)LT(i,j) Lm
Measured radiance LT True radiance PSF Point
Spread Function i line index j column index
PSF kernel 25x25 i i0-12, i0 -11,..., i011,
i012 j j0-12, j0 -11,..., j011, j012
4Available from SBRS
- LSF (Line Spread Functions) smaller slit, 1
pixel from either side of slit, dynamic range
1-10-3, scan and (simulated) track direction - NFR (Near Field Response) measurements scans of
1x10 slit (slit in track direction), up to 100
pixels to either side of peak, dynamic range
1-10-7 - NFR modeling scatter model, certain shape
parameters, no absolute scale for PSF
5Creation of PSF
Assume that PSF of adjacent pixel in scan
direction is given by 0.125/0.75 (theoretical
value from Geolocation ATBD) of center value,
adjacent pixel in track direction is 5 (from
LSF)
Adjust center value and scale parameter from
SBRS model (Harvey-Shack) so that NFR
measure- ments are reproduced
6NFR measurements and model Band 11
7NFR measurements and model All bands
8Paper by Qiu et al. on Terra PSF
9(No Transcript)
10Aqua and Terra Band 11 with Lcloud/Ltypical20
11Aqua all bands with Lcloud/Ltypical20 (PSF
size512x512)
12Aqua all bands with Lcloud/Ltypical from 10-40
13Same as before, but 10x10 pixel cloud
1410x10 pixel cloud, band 11 only, different PSF
sizes
15100x100 pixel cloud, band 11 only
16Summary
- Created Aqua PSF based on SBRS model and NFR
measurements - For huge clouds, TOA radiances 50 pixels away
from cloud will be contaminated up to 1 (even
after correcting with 101x101 PSF) - For 100x100 clouds, radiances will be up to 0.5
too high after correction with 25x25 PSF (0.2
for 51x51) - For small clouds, radiances can be adequately
corrected with 25x25 PSF as close as 2 pixels to
the cloud
17Outlook
- Still need to check the model in track direction
- Refinement needed for adjacent pixel
- Validate with real data (nLw should look
better, but more noise is expected) - If successful, implement into SeaDAS as an option
- Work with MCST on Terra PSF derivation/validation
18Backup slides