Remote Sensing Can Gather Key Permafrost Data

by Duane Nichols on July 30, 2013

Technical Description by J.W. (Bill) Rymer, July 28, 2013

There has been very intense recent interest in increased permafrost exposure and resulting release of greenhouse gases. Based on a very short search and my own experience, I will cite some examples of Remote Sensing (RS) and then draw a few conclusions regarding permafrost investigations using RS.

The technology for Remote Sensing has been growing rapidly since the mid 1960’s when agricultural remote sensing research at Purdue University led to the early NASA LANDSAT family of satellites.[1] Many schools and researchers are engaged in remote sensing world-wide. Sensors include multi-spectral scanners, hyper spectral scanning and ground penetrating radar among others. The Laboratory for Agricultural Remote Sensing (LARS) at Purdue remains very active and involves many researchers and disciplines.[2]

LANDSAT has been used to map global methane distribution. See ref. [3] which shows on pp. 16 a global / macro scale map of CH4 concentration. While this was just a macro scale, the sensors involved had 200 meter or better resolution using 2005 data. No doubt the sensors and available data products for the newer Landsat8 are much better. Other satellites such as IKONOS (commercial earth observation satellite launched in 1999) have additional sensors and varied applications. IKONOS collects publically available (for sale) imagery at 1 and 4 meter resolution according to the relevant WIKI page.

Students at Kansas University have applied inexpensive versions of ground penetrating radar to map the depth of glaciers from a light plane rather than expensive, tedious and inherently misleading small-sample drilling. It appears very likely that the same technology could be used to map permafrost both under land and under water. It should be feasible to map not only location but depth. It seems very analogous (investigation of permafrost vs investigation of glaciers) in that overview information is needed, yet far more localized than “pictures of the earth” in scale.

At least 50 non-military sensors are currently in orbit around the earth gathering a wide variety of data. These thematic mappers take images based on many frequencies/wavelengths. LANDSAT8 was launched 2/11/13 and free data distribution was to begin 5/31/13. I have not obtained specifications as to what sensors and resolutions are available on LANDSAT8.

Australians [4] have developed reflected laser technology for allowing methane detection over large areas. It would seem that this opens possible future scanning of permafrost areas looking for methane leaks or vent holes. Such technology would likely be more applicable to low altitude platforms (aircraft) rather than satellites due to the range limitations of lasers.

The Univ. of Bremen in Germany [5] has already utilized European ENVISAT data from as early 2004 to map CH4 concentrations on a global scale.

An excellent overview of hyper-spectral imaging is at the Wiki location [6] . This site mentions use of the tools for chemical imaging, mining, etc. Of great interest are tools such as applications within MATLAB that were already developed for analyzing and displaying such data.

In a 2008 example, the University of Alaska has a Permafrost Laboratory and published a paper titled “Changing Permafrost Landscapes in North Eurasia; Some Remote Sensing Observations and Challenges.” [7] This paper involves researchers from the US, Germany, Russia, Canada and others. Clearly work is already being done in this area. Slide 9 in this paper alludes to large numbers of “key parameters” that can be measured using RS. However, most if not all are macro effects rather than narrow specifics like methane leakage spots. Slide 25 shows a network of Lake-ice Methane Monitoring sites which looks to me like an admission that RS has not yet been fully applied. Regardless, I think this is a must-read paper for those interested in permafrost issues.

Other publications of interest include “Remote Sensing in Northern Hydrology: Measuring Environmental Change” 160 pp. in hardback, edited by Duqua and Pietroniro. My impression is that the book deals more with gross large area features than with specific artifacts like methane vent holes.

It should be noted that analysis of data already gathered by LANDSAT, IKONOS, ENVISAT and other platforms has by no stretch of the imagination been exhausted. Instead, analysis is driven by specific funded projects and objectives within NASA, private corporations, Universities and numerous research organizations. Thus the “data products” (maps, readily usable visual materials, etc.) as results of processing sensor data are by no means complete or exhaustive. It would be foolish to assume that the all the data has been “used up.” Instead the data products and published results are merely samples of what can be derived from the raw data. Certainly all responsible organizations involved retain the raw data for future research.

From the above examples and extensive remote sensing technology in existence, I conclude that:

1. Remote Sensing on regions or relatively large geographic areas (less than global) can be used to map permafrost and similar subsurface features. This has probably been done.

2. Effective permafrost studies should be done on much smaller samples (e.g. miles x miles rather than earth-sized) compared to most of the highly publicized “view of the earth” data products. Higher resolution is required than that aimed at “global pictures.” Investigation of sensor specifications (and metadata regarding what has already been collected) should reveal whether the resolution is adequate or whether other sensors should be launched. Inexpensive scanning platforms (such as light planes) seem attractive.

3. It is very likely that the raw data already exists to do detailed (versus global) permafrost mapping and investigations. Since my search has been short and cursory, I would be very surprised if some detailed permafrost investigations have not already been done, especially in areas like Alaska.

4. Methane vents and leaks should be detectable using existing technology. Methods could be tested and proven using existing data (if this has not already been done.)

5. A key area for further investigation would be finding out whether spectrometers and similar devices used to detect specific elements and compounds on distant astronomic targets have been used to scan permafrost areas. If they have, then detection of methane vents should be a relatively simple data processing chore.

6. An early effort should focus on finding out to what extent permafrost has already been targeted by funded research and data analysis, what tools have been used, and the extent of existing RS applications to permafrost.

J.W. (Bill) Rymer, BSEE, West Va. Univ. ’66, MSEE, Purdue ’67. Mr. Rymer managed the first Real-time Telemetry Processing System (RTPS) in the Department of Defense. He maintained a specialty in spectrum analysis and vibration for most of his career. He currently works part-time for Spiral Technology Corp. and the Office of the Secretary of Defense on Network Enhanced Telemetry. The longstanding family farm is in Lewis County.


[1] See Brief History of LARS by David Landgrebe, Dec. 1986 at

[2] See .

[3] See the following document:

[4] See

[5] See


[7] “Changing Permafrost Landscapes in North Eurasia: Some Remote Sensing Observations and Challenges By Guido Grosse, Geophysical Institute, University of Alaska Fairbanks at

{ 1 comment… read it below or add one }

Bill Rymer August 3, 2013 at 11:17 pm

As a post script: The Kansas University work I mentioned is contained in an article in IEEE Spectrum at:

The article is fairly good as a tutorial on application of synthetic aperture radar (SAR) to earth surface sensing. Also describes the context and overview rather well. Secondly, a post Doctoral student at Stanford named Lin Liu is working on permafrost mapping using SAR.


Leave a Comment

Previous post:

Next post: