The U.S. government appears to be striking an appropriate balance between risk and capability with its plan for procuring a dedicated Landsat land imaging satellite that would launch by 2011.

The preliminary strategy, outlined in a Feb. 22 pre-solicitation notice issued by NASA’s Goddard Space Flight Center, emphasizes getting the satellite built and launched quickly, but defers on the question of including a thermal-infrared imaging capability on the sensor. The formal solicitation, to be released later this year, will call on bidders to propose adding a thermal band as an option, according to the notice.

It is not hard to see why NASA and the U.S. Geological Survey, which are jointly managing what is still being called the Landsat Data Continuity Mission, kicked this key programmatic decision down the road. The thermal band would add cost and risk to a satellite in a program that already is facing a data gap, and the community that now uses this data is relatively small. At the same time, however, scientists see tremendous potential in thermal-infrared data, even if that potential has not been fully exploited to date.

Keeping design options open can be a recipe for trouble on space development projects, leading to outcomes that include cost-growth, delays, disaffection among certain groups of users or all of the above. But in the case of the next Landsat mission, this is the correct approach, provided it is managed with discipline.

With the longevity of the existing Landsat satellites uncertain — Landsat 5’s incredible run of 22 years and counting notwithstanding — it is imperative that the new satellite be launched as soon as possible to minimize any gap in the U.S. land observation data record. Given the challenges associated with thermal-infrared sensors, one can make a good case that sacrificing this capability is a price worth paying to keep the satellite on schedule and within budget.

The counter argument is that thermal-band data has been identified by scientists as valuable for detecting and tracking change on the ground. The fact that the data product has a relatively small constituency today may be due to the fact that it is relatively new. Although Landsats 4 and 5 included a thermal-infrared band, it was not until Landsat 7 was launched in 1999 that such data became available at a sharp-enough resolution — 60 meters — for most terrestrial applications.

Further, it is debatable whether the size of a user group is proportional to its importance, especially in the context of a government-provided service, which is what Landsat is.

There can be little doubt that the thermal band will add cost and risk to the Landsat Data Continuity Mission. Traditionally, such sensors have had to be accompanied by complex, power-hungry cryogenic cooling systems.

But there is a promising alternative thermal-band sensor technology that does not require cryogenic cooling. NASA has been working for a decade with so-called microbolometer detectors, which originally were developed by Honeywell under a classified contract with the U.S. Department of Defense. A microbolometer detector-based instrument dubbed the Infrared Spectral Imaging Radiometer flew successfully on the Space Shuttle Discovery in August 1997, and similar technology is featured on a sensor aboard the U.S.-French Calipso cloud-observing satellite awaiting launch from Vandenberg Air Force Base, Calif.

NASA officials are not convinced that a higher-resolution sensor based on this technology could be included on the next Landsat mission without causing major cost growth or schedule slippage — one has to wonder whether the Pentagon could offer some insight here — but clearly some believe it is possible.

This, when coupled with the fact that advancing the state of the art of space technology is an important part of NASA’s mission, strengthens the case for including the thermal-infrared band.

NASA needs to convey the message to prospective Landsat bidders that it takes the thermal-band capability very seriously. If it is possible to indicate a preference for proposals that include a viable thermal-band option, NASA should do so in the request for proposals.

If there are still too many technical question marks when the time comes to start building the satellite, NASA should move ahead — without delay, and without the thermal band. The resulting spacecraft will still have enough of Landsat’s unique characteristics, including compatibility with data from previous missions for those all-important global change studies, to justify its cost.

But dispensing with the thermal-infrared imaging capability should be a last resort. Whether in the form of incentives for industry or work in its own laboratories, NASA should push the development of microbolometer detector technology with every intention of including it aboard the next Landsat mission.