Eyes on Earth Episode 79 – Landsat Global Archive Consolidation

The Landsat Program is all about land change over time. But we can't image the past, or can we?

ADAMSON:

Hello everyone. And welcome to another episode of Eyes on Earth. We're a podcast that focuses on our ever changing planet and on the people here at EROS and across the globe who use remote sensing to monitor and study the health of Earth. I'm your host for today, Tom Adamson. Many of these Eyes on Earth podcasts have emphasized the value of the Landsat archive. Especially noting that it is the depth of the archive that makes it so valuable and useful for any number of land change studies. There is an unsung hero in this. A program that made the Landsat archive considerably deeper than it could have been without it. The Landsat Global Archive Consolidation or L-G-A-C which is as affectionately known around here "LGAC." Joining us today to talk about LGAC are Todd Taylor and Jayson Holter. Todd is the contract task lead for sustaining land imaging partnership support which includes LGAC. Jayson is the task manager for LMOP which is Landsat Missions Operations Project Ground Operations. So first of all, there was a lot of Landsat data from the early Landsats that was not in the archive here at EROS. So, Todd tell us. How did that happen?

TAYLOR:

That happened because early missions of the Landsat satellites were not able to downlink all of the data to EROS ground stations. So, we partnered with a lot of international agencies that had ground receiving stations. The Landsat satellites downlinked to those stations unique data which became part of their archive. Those scenes the data that was downlinked to them was not part of the EROS archive. Eventually there was the LGAC project that was started. To get all that data repatriated to the USGS.

ADAMSON:

Ok. And about when did LGAC begin?

HOLTER:

I think it was around 2009-2010.

TAYLOR:

2010, yes.

ADAMSON:

Ok.

HOLTER:

I think a lot of talks prior to that. But, I think the rubber hit the road really in 2010.

ADAMSON:

And we don't have that going on with the current Landsats that are operating. Landsats 8 & 9 can store data on board?

TAYLOR:

That's right. Landsat 8 & 9 can downlink all the data that it acquires to the USGS at EROS. There is still a portion of LGAC that could happen. If there is a recording error of some kind that we need to go back and get that data from International Cooperators we do have that mechanism in place.

ADAMSON:

Ok. So LGAC could be something of a backstop in case anything was lost with current data?

TAYLOR:

That's correct. Assuming the International Cooperator received that data.

ADAMSON:

So, going back to the older Landsats that had some kind of limitations with how much data they could store. When you went to these ground stations, how were they able to send the data to EROS?

TAYLOR:

The data came back in many different types of formats from HDDTs. Sometimes via FTP. I can let Jayson explain a little bit more about the historical media that this data arrived in.

ADAMSON:

You said HDDT. Maybe Jayson can describe what that is.

HOLTER: 

HDDT is a High-Density Digital Tape. They kinda look like your big old reels on your movie projector. Like you see on the movie projectors. The big tape reels. They're big old tapes. The largest tape is 10,000 feet of tape. And I think the smaller tapes are like 5000 feet or something like that. So, there is well over a mile of tape on these high-density tapes. From a Landsat 5 perspective they could record up to two passes on that one tape.

ADAMSON:

Ok. So, some of the data that came to us from these other ground stations was electronically transferred. And a bunch of it came in shipments of these old media like that. Are we still able to read data off of that old media?

HOLTER:

Yes. USGS has some Honeywell high-definition tape drives. HDT drives. From Martin Marietta to Honeywell. I think those are the two main brands that USGS has inhouse. Over the years we have brought those systems back up from storage and try reading all these sample tapes that Todd and others have gone and worked with the International Cooperators with. To get us sample tapes. And we've restored these various systems and tried recording or reading these tapes. And we've had success with stations in Pakistan, Saudi Arabia, Thailand. I think those were the three key ones that we've had major success with. We are trying to get some Ampex tape drives running to potentially read some high-density tapes from Brazil and some additional high-density tapes that Thailand has. They had a couple different flavors if you will. One that worked with our Honeywell drives and then another one that we are trying to get to work with our Ampex drives.

ADAMSON:

This strikes me like doing something like bringing an old box down from the attic filled with 8 track tapes and we want to try and play them on our smart phone or something like that.

HOLTER:

Exactly.

ADAMSON:

Are there any particular challenges with these tapes?

HOLTER:

Oh, yeah. There's how they've been stored. We are having to bake the tapes. To deal with the oxide. So, we are baking them for periods of time. Up to 24 hours at a low temperature. We are running them through a winder cleaner system. And then running them through exercise drives to fast forward to the end of tape. Rewind to the beginning of the tape. To repack the tapes. And then we attempt to read them on our ingest drives.

ADAMSON:

Ok. So, some of these tapes might not be in the best condition. And then they need to go through this process?

TAYLOR:

In some cases, the technology didn't exist anymore. And we had to procure new technology going back to some old design documents. Having a manufacturer create that and ship it to us to be able to read some of these historic tapes that we have received from the ICs. Many of these tapes were stored and International Cooperators in an area that was a rainforest. And when we did get them back, they had mold and other issues like that which required us to have the tape baking and the cleaning and run them through different processes just so that we would be able to read the data.

ADAMSON:

And this process was pretty successful? You were able to get the data off of most of these?

HOLTER:

The moldy tapes, the more moldy tapes that Todd is talking about are the digital cassette tapes. They are DCTs. They're a different type media than the high-density digital tape. But they look like an oversized VHS tape. If you remember the old VHS tapes in VCRs back in the day. This looks like a VHS tape that's probably about three times the size of a VHS tape. Those are the ones we've been working with from Indonesia that have had various levels of moisture damage from mold. I believe we've had over 300 of these tape media. I believe as of today we're down to about 40 left to go. We started with this media, it was prior to Landsat 8 launching. Then we had some challenges with the old drives and keeping the systems up and running. Over the years we've brought a subject matter expert out of retirement from Ampex. His name is George. George has been out to EROS a few times. And helped our engineer get these old drives running to read these tapes. And then we've also been working with Boe Industries to custom make a winder cleaner to run these moisture damaged tapes through. So that's been a few year process to get the drives running, get people out here. To get the winder cleaners through procurement process and purchased. I think it was last October we started up on the remaining 125 tapes that are the real challenging ones that have various levels of mold on them. We're down to about 40 to go. We're having some great success with this but it has been a very long process over the years.

ADAMSON:

Well with Landsats 8 & 9 having the ability to store data on board and LGAC being something more of a fail safe for that, is there a time when LGAC will come to an end? At least wind down a bit more?

TAYLOR:

We're getting to the point where the remaining scenes to capture from LGAC are the most troublesome. That ends up being a return-on-investment question really. We will keep working the project as long as there is funding and data to get. But at some point it is just going to be cost prohibitive to go after those remaining scenes.

ADAMSON:

Ok. So far it seems to have been worthwhile up until the really challenging ones. And I was kind of wondering, why do we go through all this trouble anyway? What are the benefits of pulling all the Landsat data together into one place?

TAYLOR:

For one thing, it's added 6.6 million images to the archive and it's growing. Many of these are unique scenes. Scenes that did not exist in the USGS archive. But existed in the International Cooperators archive. What this does is it gets those unique scenes available to the public who can use them. A lot of these areas that have collected images are over rainforests. Which clouds are an issue there. So, every pixel every bit of data that we get that is cloud free, is valuable somewhere. Someone will be able to use that for research projects going forward.

ADAMSON:

And here at EROS we process all of the data to a consistent standard. So, researchers around the world are able to have the confidence that the data is all consistent. I gotta repeat that number you said again! 6.6 million images have been added to the Landsat archive through LGAC. Now, let me see if I have this number up to date. There is about 10.3 million scenes that are downloadable and useable in the Landsat archive. So, that is pretty remarkable. Landsat or LGAC added that many more scenes. That is much more than half of the Landsat archive.

TAYLOR:

Yeah, that is correct. It has really been a successful mission when you consider that many of the data is unique to the archive.

ADAMSON:

I'd like to thank Todd Taylor and Jayson Holter, part of the LGAC team at EROS who found a way to image the past. Thank you for joining us for Eyes on Earth. You can find all our shows on the USGS EROS website. You can also follow EROS on Facebook or Twitter to find the latest podcasts. Or just subscribe on Apple or Google podcasts. This podcast is the product of the U.S. Geological Survey Department of Interior.