Need tips on searching satellite imagery for thermally-detectable upwellings generated in waters within the energy "footprint" of existing ocean windfarms

[Ron Huber]

Hello,  I am executive director of Maine NGO Penobscot Bay Watch.  We have a question that we hope Public Laboratory might be able to answer, or at least offer some guidance.

We are looking forward to the coming deployment of floating deepwater windparks off Maine - two are under active review - but are anxious they be sited well away from nearshore currents vital to larval transport of fish and shellfish - and other biological hot spots along the Gulf of Maine coast.  

We are trying to learn to use satellite imagery to spot the water turbulence or upwellings or increased primary productivity that is generated incidental to operating existing ocean windparks. This will be very helpful in helping agencies learn to factor this issue into the application reviews for future ocean windparks like those proposed off Maine and in fact for all future ocean windpark prosposals world wide. However there are so many satellites out there we don't know where to start!

The existence of this "upwelling byproduct" phenomena was first brought to our attention  by Norwegian Meterorological Institute researcher Goran Brostrom's 2008 study "On the influence of large wind farms on the upper ocean circulation"    Brostrom used data from existing ocean windfarms to model the nature and size of upwellings and other eddy effects generated in the water column  beneath the kinetic energy diversion footprint, or "wind shadow", of existing ocean windparks   As his abstract summarizes: "The generated upwelling is sufficiently enough that the local ecosystem will most likely be strongly influenced by the presence of a wind farm".

As we understand it, small-scale features of surface circulation often take the form of eddies, that can be visualized in the ocean color, sea surface temperature and sea surface roughness. Again, there seem to be  so many satellites that we are at a loss of which to look at, and then,  how to isolate the  upwelling imagery - if any- of the various ocean windparks that already exist.

If anyone knows of a way to view at the thermal signatures or boosted chlorophyll signatures of the existing ocean windfarms of the world - including Statoil's Hywind the world's only - so far - floating windmill, please let us know.

Sincerely

Ron Huber
Penobscot Bay Watch
POB 1871
Rockland Maine 04841

e: coast...@gmail.com
tel: 207-691-7485
web: www.penbay.org

[Ned Horning]

Hi Ron,

Have you checked out the GoMOOS site: http://www.gomoos.org? They run one of the flagship ocean data viewing services. If you want to do more of your own processing a good place to start is the NASA ocean color site: http://oceancolor.gsfc.nasa.gov/. Both of the groups that run these services are quite friendly so you might be able to ask them for advice.

Most of the daily ocean data sets have a coarse spatial resolution so depending on the scale of the upwelling it might be difficult to detect.

All the best,

Ned

[Ron Huber]

Thank you Ned!  I've used GOMOOS for its weather services for years, but hadn't considered seeking them out on this - as I should!  Very good people, though I meet them mostly at the GOMOOS display table at the annual Maine Fishermen's Forum.

Taking a look at http://oceancolor.gsfc.nasa.gov/  NASA's Ocean Color Web....wow! 

As you wrote, it comes down to whether the predicted upwellings are of a size that  achieves spatial resolution. I'm wondering which one of them: chlorophyll, sea surface temperature or surface roughness, is of the finest resolution?  And whether that is as important as other parameters when seeking to make them perceptible?

I'll consult with GOMOOS and the NASA Color Web folk, and will report back, though I would be very glad to get input from any and all of the Public Laboratory community, along the way.

Thanks!

Ron

--

Ron Huber
Penobscot Bay Watch
POB 1871
Rockland Maine 04841

e: coast...@gmail.com
tel: 207-691-7485
web: www.penbay.org

blog: penobscotbay.blogspot.com
twitter http://twitter.com/penobscotbay

[Dave Haffner]

Ron,

It's great that you're thinking of satellite data for your kind of study.  My specialty is not ocean color, but I have a good sense of the physical and operational capabilities of the satellite sensors in the VIS, and just looked some stuff up on the IR.  I think it's likely that you'll learn something about your region of interest from these data, but you may need to do some statistical analysis of the results to identify changes in seasonal behavior or bona-fide trends.  Even if you don't have the spatial resolution you believe would be ideal to resolve the ocean processes you have an interest in, you might find a lot of good science can be done by looking for increases whatever resolution you have available. It may be harder to drive your point home without a good looking map that shows a clear change in the spatial pattern in cholorphyll or SST, but if there's a trend you can measure, that's important in and of itself. 

On the other hand, you may actually have the resolution you want with the available data. The best spatial resolution out there now for these regions of the spectrum is VIIRS.  VIIRS has dazzlingly high resolution for a scientific sensor - 750 m . But VIIRS is a brand new instrument, just launched aboard Suomi NPP this past fall, so for a while the data coverage may be intermittent and the quality of the calibration will improve with time as the on-orbit performance of the instrument is characterized.

The ocean color website at NASA that Ned referred you to is the best place I know to get up to speed with the data from VIIRS ocean products and similar products with lesser spatial resolution that are available from previous satellite missions: SeaWIFS, which is decommissioned, and the two MODIS instruments, one aboard Terra and the other aboard Aqua, both of which are still operational.

You mentioned three products - productivity / Chl, SST, and surface roughness. Here's what I know is available from these four sensors.

SeaWIFS (late 1997-2010), both MODIS (2000 & 2002- present, and VIIRS (just begun) provide Chl measurements

SST is available from MODIS and VIIRS, but not SeaWIFS

I suspect these algorithms take roughness into account, but the derived roughness is available from many instruments. 

All of these sensors I've named have good resolution - better than anything I work with in the UV - but I think you will find that many products will be averaged to a common resolution, say 4 or 10 km, over a day or a week or month. This makes the comparison of the data sets easier.  Of course the data from the newest sensors, like VIIRS, may contain more spatial information at their native resolution than these gridded, averaged products will.

You will see data sets referred to as L2 or L3.   L3 (level 3) data are the gridded averaged products. They are usually screened to give an average of the best quality data. The grids these data are provided on are generally regular, which makes comparison from day to day or year to year much easier. 

The L2 (level 2) data will be the highest resolution data you will find for any given sensor.  L2 data are usually the direct output of the science algorithms that process backscattered or emitted radiances measured by the satellite sensor into something biogeophysical that you can use. The sensors' native spatial resolution is usually retained in the level 2 output, and because of this, the data are reported on an irregular lat/lon grid.  The L2 data are sometimes considered harder to use, but once you get a handle on them they are quite powerful. I think for your study you could have a look at both.  The L3 data may be the best to look at for trends since the averaged data will reduce some of the noise and the regular organization of the grid makes it easier to handle the  data. The L2 data should give you a better looking map owing to it's higher spatial res.

Hope this info is helpful. There's a wealth of data out there from these missions. I hope they are of use to you and your group!  Good luck with everything.  And do contact the folks at the NASA ocean color page -- they're good folks with a lot of experience.

Dave

[Ralph J. Coppola]

I would like to make a suggestion:
 
It would be nice if a e-mails could include PLOTS as the first word in the Subject Field. This would keep the Spam Filters thinking that the messages are junk.
 
 

---

Date: Sun, 12 Feb 2012 16:48:19 -0500
Subject: Re: Need tips on searching satellite imagery for thermally-detectable upwellings generated in waters within the energy "footprint" of existing ocean windfarms
From: daveh...@gmail.com
To: publicla...@googlegroups.com

[Ron Huber]

Thank you Dave! I am very glad for the details on the nature of hte data sets available and related background. We are getting a sense of the satellites that will be useful to our effort. It looks like:

Chlorophyll / Productivity

SeaWIFS   1997-2010 

MODIS  2000 and 2002 to present

VIIRS  2012 - ?

Sea Surface Temperature

MODIS  2000 and 2002 to present

VIIRS  2012 -

I'll be getting to know that ocean color website;  lot's there!

 Thank you for your point that even if the resolution magnification available makes differentiating individual upwellings generated problematic for a media presentation,   "if there's a trend you can measure, that's important in and of itself. "

 

See some notes/links I've aggregated that may add details of whats being looked into about this turbulence or upwelling phenomenon, under "research notes" on the Public Laboratory home page 

Thanks all!

Ron

[Jeffrey Warren]

Hi, Ron - I'm not sure I have a lot to add to the wealth of advice posted already -- but I posted a quick response to your excellent summary post (http://publiclaboratory.org/notes/ron-huber/2-12-2012/learning-search-satellite-imagery-thermally-or-productivity-detectable-oce#comment-456) and wanted to make sure you didn't miss it:

I was wondering what kind of measurements of primary production you're looking for -- i.e. could near-infrared aerial photography be used for high-res studies along these lines?

I don't think flying balloons near windmills is a great idea :-0 but i do wonder if NDVI regularly collected from a low-flying aircraft could be used to establish/measure production? Are you talking about productivity of algae, mainly?

Jeff

[Ron Huber]

Thanks for writing Jeff.  Increased algae production would be a detectable factor. But  there should also be a significant water temperature difference in the vicinity of an ocean windpark, at least at certain times of the year. That should be detectable. 

 

We're certainly not committed to any technology type. Having now taken a glance at the infrared aerial photgraphy webpage of public laboratory,  it certainly looks as though that technology could be effective. 

As far as who would carry out such flights, one would think it would be best to piggyback it on some other flight mission.  For example, NOAA's right whale tracking in the Gulf of Maine is done by aircraft.  It shouldn't be too difficult to graft  another mission - infrared photography - onto that. Though of course there are no ocean windmills in the Gulf of Maine. But there ought to be something similar that tracks marine mammals around the UK and waters elsewhere off Europe where ocean windparks already exist.

Ron

[Dave Haffner]

Ron,
You're most welcome! I looks like you have all the right info re: these four instruments. You might also want to have a look at the Giovanni ocean portal at the Goddard DISC:

http://disc.sci.gsfc.nasa.gov/giovanni/overview/index.html

The data here will likely be of lower spatial resolution than you might ideally want (and perhaps can get), but this tool is a great stepping stone that allows you to do useful analysis in a web interface.

Dave

[Ron Huber]

Thank you for the Giovanni link. The  web interface of Giovanni does look user friendly. 

Ron 

[Jeffrey Warren]

This is an old thread, but if you're still interested in measuring algae production and/or sending an infrared camera system along with the NOAA whale monitoring flights, I'd love to help you get set up with one -- it ought to cost about $120. Chris Fastie just posted a great note on getting 2 cameras to trigger 1/100 of a second apart, which should be plenty for getting infrared/visible images: http://publiclaboratory.org/notes/cfastie/4-14-2012/good-shutter-synchronization

These could be mounted out the window with a clamp or something.

Are you in touch with the NOAA whale tracking folks? Could we try a proof of concept in a place where we know there's an algae bloom?

Jeff

[Ron Huber]

That's sounds very interesting.  We have been struggling with satellite data (AquaMODIS); looking at the waters around existing ocean windfarms in the North Sea to see if we can detect a thermal difference. At this point we just want to get to know the rudiments of using the satellite data. But have reached no conclusions yet.  We picked the Alpha Ventus wind farm off Germany which came online late 2009 as a test site See our Bay Blog entry about this 

The aerial infra red imaging system you describe sounds splendid. We would  as soon use it in our Penobscot Bay currents and water quality studies, using our volunteer private aircraft pilots. Would the array you describe be usable on a standard 4seater private aircraft?  

Ron

[gonzoearth]

Hi Ron and All, 
I think this could work out wonderfully. I just put up a research note about using a four seat high-wing Cessna for shooting vertical aerial images. I use these rubber hand clamps to hold the camera(s) in the vertical "geodata" position. This setup is basic but I have found it to just work.
http://publiclaboratory.org/notes/gonzoearth/4-23-2012/light-aircraft-aerial-mapping-setup
Best,
Stewart

[Ron Huber]

interesting. a much more systematic set of aerial (or infrared?)  photos or video than my previous hold it out the window effoerts.  So the vibration is sufficiently damped by light hand pressure on the camera?

Ron

-- Post to this group at publicla...@googlegroups.com. To unsubscribe, email publiclaborato...@googlegroups.com. Options at https://groups.google.com/d/forum/publiclaboratory?hl=en

[gonzoearth]

Yes, it can be tiring if it is a long flight -but it really helps steady the camera with this basic setup. You do pull it back inside while not shooting as well.

It wants to move back and forth on the x-axis because of the linear configuration with only one clamp area holding it in place. If it was "h" or "H" shaped this would help, but we are using this basic "l" shape here.

Stewart

Ron

-- Post to this group at publiclaboratory@googlegroups.com. To unsubscribe, email publiclaboratory+unsubscribe@googlegroups.com. Options at https://groups.google.com/d/forum/publiclaboratory?hl=en

[Dave Haffner]

Ron,  The images on your blog have got me curious - what Aqua MODIS  temperature product are you looking at? It  would be interesting to validate/compare the MODIS product with thermal infrared data collected from aircraft.
-dave h.

[Ron Huber]

Dave,

Here's how I obtained these  SST images (and chlorophyll images for the same purpose)

I used this browser http://oceancolor.gsfc.nasa.gov/cgi/browse.pl?sen=am

Sea Surface Temperatures

I selected Aquamodis. Clicked on SST in the upper row of buttons. Selected Night. Created global image list. Selected the Januaries of each year 2004 to 2012.  Saved those off as pngs (Jan 2008 - Jan 2012 images here only).Cropped them to only the southern North Sea location of interest. 

Magnified/zoomed that by 1,000% and by 2,000%          

Chlorophyll. Used the same browser and sequence except selected for chlorophyll instead of SST

  February 2002 to February 2012 

Folder A contains the original large sized png images  from NASA's ocean color web site. 

Folder B contains cropped to North Sea area

Folder C contains cropped to southeast North Sea

I am of course always open to other better (and probably simpler) methods for extracting the desired information out of these great Aquamodis and related satellite images!

Ron 

[Randy Sargent]

One of the MODIS Ocean Chlorophyll products is also explorable here:
http://timemachine.gigapan.org/wiki/Ocean_Chlorophyll.  

(We're close to beta test on open-source tools to let anyone build these zoomable and time-explorable timelapses and we're looking for testers)

-- Randy

[Ron Huber]

Randy

thank you! This looks very helpful for our purposes.

Ron

[Dave Haffner]

Hi Ron,

I took a quick look at the Monthly nighttime 4 micron SST and Chlorophyll A data from MODIS Aqua using Giovanni in the 1x1 degree region 6-7 E, 53.75-54.75 N.  It's a first cut analysis, but there's a hint of a drop in surface temperature and Chl A after Alpha Ventus went operational in April 2010.   Other things are making the temperature go up and down too though, so it would be useful to "de-seasonalize"  the data by removing the mean annual cycle from a period before Alpha Ventus was built.  Then you may see a trend more clearly. Thankfully, Aqua MODIS has been operating since 2002, you can look at data from several years prior to when Ventus opened to characterize a baseline.

I don't know how precise the Giovanni spatial sub-setting is. If they can only go as small as 1x1 degree regions, it might be worth the extra work of getting familiar with the 4 or 9 km resolution MODIS data. There may be a trend signal in the data at those scales that might get washed out when you average over 1 x 1 degree (roughly 60 x 100 km at this latitude).

Dave

[Jeffrey Warren]

Hi Ron, I just came across this article on temperatures under wind farms, though it seems like the study is for onshore turbines: http://www.nasa.gov/topics/earth/features/tx-wind-farm.html Is that a similar warming effect?

Are there any flights coming up in the gulf of Maine? I'd like to help you get a dual camera rig set up if you're interested... it'd be a very interesting experiment to get some really hi-res infrared imagery, like 10-20 cm/px. 

I just uploaded more documentation on synchronized camera triggering and dual camera mounting for the recently shipped dual infrared camera kits. The synchronized triggering is especially useful.

http://publiclaboratory.org/wiki/dual-camera-kit-guide

http://publiclaboratory.org/wiki/dual-camera-kit-electronics

Chris Fastie and I have been working on these guides and if anyone's interested in helping out, the "interpreting infrared images" section needs the most input, but there are loose ends throughout.

Jeff