The declassification of historical intelligence satellite imagery has been a boon to scientists, environmentalists and other researchers since it began with President Clinton’s executive order 12951. So, for example, “The declassification of imagery from CORONA and subsequent intelligence satellite programs has inspired a revolution in landscape archaeology in the Near East,” wrote archaeologist Jason Ur.
But last year’s declassification of imagery from the KH-9 HEXAGON intelligence satellite will be slower to generate any such revolutionary impact because the newly declassified images are so hard to access and to use.
The KH-9 imagery was successfully transferred from the National Geospatial-Intelligence Agency to the National Archives. But in order to protect the perishable film it must be maintained in cold storage, and so it was all sent to a National Archives facility in Lenexa, Missouri Kansas. Researchers must make their best guess as to what images they are seeking, and then order the originals to be transferred from cold storage. It’s a slow and cumbersome process.
The larger policy issue is that the archival burden on the National Archives and Records Administration is growing faster than the available resources. The task of curating the nation’s documentary heritage appears to be escalating out of control. Meanwhile, the Archives is literally running out of space. Last month, Archivist of the United States David S. Ferriero announced the closure of three NARA facilities “as part of ongoing budget adjustments.”
Recently, one concerned researcher shared his frustrations about the current procedures for obtaining declassified satellite imagery. Secrecy News forwarded his comments to the National Archives and Records Administration, and a NARA official provided an annotated response, reproduced below.
Researcher: Since the [KH-9 HEXAGON] film is original negative, it was all shipped to Lenexa, Kansas.
NARA: Correct. There is a potential that some of the film was not acetate and as such didn’t require cold storage but we did not have the resources to review each of the 14,685 cans to determine the base format and we erred on the side of caution in determining where to store it.
Researcher: NGA DID make available to NARA under the MOU [Memorandum of Understanding] the imagery, and finding aids, which are image mosaic overlays on maps 1:100,0000. These are completely useless.
NARA: There was no MOU for this particular transfer. Previous transfers had MOUs because there were multiple sets of records which were being distributed between NARA, NGA, and USGS. I think that there is some confusion between the past transfers and this one. For this transfer we were provided with frame metadata. The overlays referenced here do not index KH-9 film, they only index the airborne imagery previously transferred from NGA.
Researcher: There is also a CD-ROM which can be loaded onto a flash drive containing an ASCII file with mission date, pass frame, lat-long footprints, in an Excel format. But there is no way to know if the images are fully cloud-covered or not until the film arrives.
NARA: The CD provided for access as described in the KH-9 reference guide is what was provided to us by NGA. We know we can make it better but it will likely never provide information on cloud cover by image. All of our film, except for that indexed by the overlays, requires looking at it to determine quality and potential cloud cover.
Researcher: One must submit that data to an archivist who then converts the info into Original Negative Can numbers. The researcher then must submit a second request including the ON number and the cold storage numbers to an Archivist, who quality controls it and submits the request to NARA Lenexa.
NARA: As with any other transfer of imagery, there is a process involved in going from whatever index exists to identifying the cans of imagery. In the case of KH-9, once researchers identify imagery from the frame metadata, we have a can locator which converts the information for missions, dates, etc. to an actual can of imagery. This can locator is available for copying by researchers, and is available through the consultant in the research room who can provide the necessary information. It is also available on a hard drive for researchers to use themselves.
There is a need to fill out a pull slip for documentation of use and a Lenexa request form but that is done at the same time and does not require much effort other than writing a can number and barcode.
Researcher: The cans show up a few days later, and an Archivist must then quality control the cans for “supply chain management.”
I have spent a week at College Park just to find this out, and I have yet to actually order a can and see imagery.
NARA: The process for requesting cans from Lenexa is the same for any record stored there. We submit the requests on a daily basis, the Lenexa staff pulls the items and ship them out the next day. They are potentially available two days after the initial request. We do have to take time to document where the cans are every step of the way in order to ensure the security of the holdings but that does not slow the process down significantly.
The biggest issues are those simply related to having records stored offsite–timing of requests, ability of staff pulling the items to find the correct items, and the weather which affects the shipments both during the winter and tornado season. There are sometimes preservation issues identified early before the records are used but that is very rare and they are generally addressed quickly so the researcher does not have to wait.
Researcher: By the next Friday, the researcher can only have the film checked out for 3 business days, Friday, Saturday, and Monday, then the film must be flown back to cold storage.
NARA: All of the research rooms have a 3 business day hold for records. This is simply to ensure that records are looked at in a timely manner and are available for other researchers. There is always the opportunity to extend the period of retention but the researcher needs to communicate a need for that.
The NARA official added a rough estimate of the cost of create a duplicate set of KH-9 imagery to facilitate user access:
“At 14,685 cans, and an estimate of $800 worth of film stock per can, the cost is likely more than 11 million dollars. In addition, we estimate it would take a dedicated employee some 8 years to perform the work (roughly 5 cans/day).”
“Digitization of course avoids the cost of the film stock, but has its own costs and challenges,” the official said. “We have to try and figure out where we focus our limited resources.”
FAS experts believe government shutdowns are science shutdowns: costly and ineffective standoffs that stifle scientific pursuits and do harm.
We always knew that healthy children do better in school. Now we have rigorous empirical research to back it up.
Truly open science requires that the public is not only able to access the products of research, but the knowledge embedded within.
Over the last year we’ve devoted considerable effort to understanding wildfire in the context of U.S. federal policy. Here’s what we learned.