Not logged inOcean Color Forum

The forum is locked.

The Ocean Color Forum has transitioned over to the Earthdata Forum (https://forum.earthdata.nasa.gov/). The information existing below will be retained for historical reference. Please sign into the Earthdata Forum for active user support.

- By dem1 Date 2006-05-22 11:32
Hi,

Just a little question: I know that MODIS has bowtie effect, but what about SeaWiFS ? OCTS ? CZCS ?

Thanks in advance,

Julien
- By bryanfranz Date 2006-05-22 11:48
Yes, pixel size increases with scan angle for all sensors.  To my knowledge, the first sensor designed to compensate for this is VIIRS.

-- bryan
- By dem1 Date 2006-05-22 12:30
Ok, pixel size increases with scan angle for all sensors, but for MODIS they overlap themselves on track borders every 10 lines strips.
I believe that SeaWiFS work only one line by one line, so track pixels can't overlap, isn't it ?

Julien
- By seanbailey Date 2006-05-22 13:01
Julien,

The bowtie effect is an artifact of the IFOV of the sensor at the larger scan angles. 
It affects SeaWiFS as well.  At nadir, SeaWiFS IFOV is ~1.1km square.  At scan edge
it approaches 3x7km, thus scan edge pixels WILL overlap.

Sean
- By dem1 Date 2006-05-29 09:35
Hi,

In fact I realise we don't speak about the same "bowtie effect", sorry.
The "bowtie effect" I speak is described here:
http://www.sat.dundee.ac.uk/modis-faq.html#geo
"This is the effect whereby the image seems distorted near the edges. One of our clients refers to it as the earthquake effect and this is what it looks most like at the edges. This results from the fact that MODIS scans 10 lines at a time, unlike AVHRR and SeaWiFS which only scan 1 line at a time. What the satellite sees as a pixel, the footprint, increases with distance. Unfortunately the distance to a pixel increases with scan angle mainly due to earth curvature. This means that the pixels near the edge of an image are bigger than the ones in the middle. The satellite is configured to move forward 10km in the time that it takes to scan once. This is so that pixels in the middle of the scan line match up next to each other. At the edges though,the pixels are bigger, up to 6 times wider and 4 times longer. This results in oversampling, i.e. the same bit is imaged twice. The reason that such effects are not obviously seen with LANDSAT MSS data is that the earth curvature effect is smaller as the swath width of the instrument is smaller."
http://www.sat.dundee.ac.uk/modis-bowtie.html

So my question is: in MODIS L2 pixels longitudes and latitude centers are not continuous (they are continuous in each 10 lines), but are there continuous for others sensors (SeaWiFS, ...) ? (Without taking care about 180/-180 limit or poles cases...)

My problem is that I've to compute the coverage of each pixel, and I do this by computing each edge (the mean of the center of the 4 around pixels - I know that it isn't the true coverage of L2 track pixels, which overlap in the true life -), and this doesn't work if lat/lon aren't continuous. For the moment I handle this problem for MODIS by work 10 lines by 10 lines, but I want know if I have to do that for SeaWiFS and others sensors.

Thanks in advance and sorry for the bad explain of the problem,

Julien
- By Anonymous Date 2006-05-30 16:20
Julien,

What you are referring to is scan overlap, which is an artifact of the bowtie effect.  Scan overlap occurs for any sensor that collects multiple scan lines per scan (MODIS and VIIRS).  It causes discontinuities in the pixel-center spacing between the last pixel of one scan and the first pixel of the next, at angles not close to nadir. 

This is not a problem with single-line-per-scan sensors (SeaWiFS and AVHRR), which have fairly uniform pixel-center spacing between scans.

I hope this answers your question.

Fred Patt
- By dem1 Date 2006-05-31 02:34
Fred,

Thank you very much! it's well the "scan overlap" problem I refer.

About OCTS and CZCS, I suppose they are single-line-per-scan sensors, aren't they ?

Thank you,

Julien
- By Anonymous Date 2006-05-31 08:32
OCTS is 10-lines-per-scan.  CZCS is single-line-per-scan.
- By spdelphi Date 2006-12-23 08:47
Can msl12,4 get rid of the 'BOWTIE effect'? Thanks.

Powered by mwForum 2.29.7 © 1999-2015 Markus Wichitill