When Michael Freilich returned to NASA in 2006 to run the Earth Science Division, a National Research Council panel finishing work on a 10-year plan for space-based Earth observation recently had begun warning Washington policymakers that the Bush administration’s focus on human lunar exploration was coming at the expense of the agency’s study of the home planet. The so-called decadal survey the panel ultimately delivered in 2007 prescribed $2 billion in annual funding to pay for no fewer than 15 new missions chosen to ensure U.S. Earth observation leadership through 2020 and beyond. 

In the seven years since Freilich bid adieu to Oregon State University and stepped down as principal investigator for a pair of NASA-funded ocean-wind missions, the Earth Science Division has barely made a dent in that list of 15 missions; although several of the recommended missions are under study, only two — Soil Moisture Active-Passive and IceSat-2 — are well along in development. Both spacecraft are slated to launch in 2015.

While NASA’s Earth Science Division today has the biggest budget of any of the agency’s five science-focused divisions, the $1.7 billion the Obama administration has wrung out of Congress the past couple of years has been squeezed by rising launch costs, two costly launch failures and the ongoing expense of maintaining 20 operating missions.

Freilich spoke recently with the SpaceNews editorial team about some of the challenges facing an Earth Science Division newly tasked with ensuring the continuity of the world’s longest-running Earth observation series and taking over responsibility for some Earth-system measurements originally to be handled by the National Oceanic and Atmospheric Administration (NOAA). 

Have launch costs depressed the launch rate of Earth Science missions?

We are not constraining our portfolio today based on a perceived lack of either launch vehicle opportunities or funding for such launch vehicle opportunities. We don’t have a death spiral of exceedingly costly launches, therefore leading to unreasonable risk intolerance, leading to additional costs for payloads at the unreasonable level. We have identified launch vehicles all the way up through launches through 2018, and we’ve been able to accommodate the four launches that are going to happen from 2014 through the first quarter of 2015 — the Global Precipitation Measurement satellite, Orbiting Carbon Observatory-2, the Soil Moisture Active and Passive mission, and the Stratospheric Aerosol and Gas Experiment 3 to the international space station — plus the additional seven launches between the first quarter of 2015 and 2020. The potential strategic missions we don’t have launch vehicles for yet are the Surface Water and Ocean Topography and Pre-Aerosol, Clouds, and Ocean Ecosystem, both scheduled to launch in the 2020 timeframe.

United Launch Alliance has parts enough for five more Delta 2 rockets. Earth Science has now spoken for three of those. What do you do post-Delta 2?

We’re looking at launchers potentially such as the Falcon 9 version 1.1, potentially such as Antares, and a variety of others that are less mature at the moment.

What about Orbital’s Taurus XL? It’s still available, even though it hasn’t flown since back-to-back failures in 2009 and 2011 destroyed the Orbiting Carbon Observatory and Glory satellites.

If it cleared the technical hurdles, and that is up to the NASA Launch Services Program and the agency, then there would be no basis for not flying on a Taurus XL. But clearly, they have a number of hurdles that will have to be cleared to convince everybody, including themselves, that a return to flight for the Earth Science Division is worthwhile. But I have no giant, inherent bias against them.

Even though you lost two spacecraft to Taurus XL failures?

I actually hadn’t noticed that, but thank you for reminding me.

Would you be the return-to-flight customer for Taurus XL?

I don’t think it’s worthwhile going into hypotheticals right now. It’s my understanding that Taurus XL remains on the NASA Launch Services 2 contract and therefore when a request for launch services goes out Orbital would be free to bid Taurus XL.

Can you save any money on launch costs by flying your instruments as hosted payloads? 

We’re trying some of that in our Venture-class program.

Do you see opportunity to fly NASA Earth Science instruments as hosted payloads on commercial satellites?

Well historically, geostationary orbit has not been the Earth science community’s choice. Look at the decadal survey. They’ve not been particularly enthusiastic about geostationary, which is where the business model exists. You can get high frequency and time measurements from geostationary orbit, but limited geographic range. Our Tropospheric Emissions: Monitoring of Pollution mission, which is a recently selected Venture-class instrument for air quality measurements, is actually going to be going up as a geo-hosted payload at approximately the same time as the European Union’s Sentinel 5 and South Korea’s Geo-Kompsat-2B in 2018. The Korean satellite has a copy of the Geostationary Environment Monitoring Spectrometer pollution sensor on it. So the three of those will be in geostationary orbit and all exchanging data. We will have both the global coverage and the high-frequency measurements. It’s a pathfinder.

What progress have you made on ensuring a follow-on for Landsat 8 since the White House assigned that responsibility to Earth Science?

Between now and August  2014, we will be doing system design for a sustained land-imaging system, which will have at least the capabilities of the present Landsat 7 and Landsat 8. Now, it’s the information that’s important, not the individual missions or how the information is gotten. That’s the system-engineering approach that we are infusing in this joint study with the U.S. Geological Survey that we’re kicking off. We will be providing the administration basically with three options for such a system, so that the administration and later Congress can decide where they feel most comfortable. 

Can you go into more detail about those three options?

I’m not saying this is what we’re going to do, but let me give you some hypothetical examples. This might be the approach with the lowest risk to the data record: Make a design that allows us to have continuity with Landsat 7 and Landsat 8, and buy all the parts we need for the next 20 years. We’ll build these missions serially, so there will be no risk of nonrecurring engineering owing to parts obsolescence, but 20 years from now you’re still flying the same design you are today. Think of the Defense Meteorological Satellite Program. Next, there’s a near-term capability approach: Get as much relevant land-imaging capability in the very next mission that we can and put all of our money into that, making it the most capable spacecraft possible and figure out after that one launches what we’re going to do for the next one. That may sound familiar. The technology infusion approach is sort of in the middle. Also, we want to leave the door open for interagency partnerships, international partnerships, and potentially partnerships with the private sector.

Are you wary of Landsat commercialization this time around, given past failures?

We’re not ruling anything out. I don’t intend for us to repeat the mistakes of history, but just because we tried something 15 years ago and it didn’t work does not to my mind say we shouldn’t examine it in the context of an engineering study. 

You said NASA will lead a joint team with the U.S. Geological Survey (USGS) for this Landsat follow-on. Who will be responsible for what?

NASA is the lead and USGS represents the user communities, writ large. We’ve identified a small group of people to scope the study, three or four people from each agency. That group will study what capabilities exist and ideas, in or out of the governmental sphere. What we are fundamentally not doing is first collecting “desirements,” then coming up with a design that satisfies many of those desirements, which have somehow in people’s minds morphed into requirements. Cost is a key element and constraint on the design. It’s on the order of $100 million to $120 million a year, on average, over 20 years.

What are your plans for the climate instruments once manifested for NOAA’s Polar Free Flyer spacecraft? Those were to be launched as part of NOAA’s Joint Polar Satellite System (JPSS) program, but  President Obama’s 2014 budget request moved those over to NASA. Is it NASA’s job to figure out how to fly those instruments?

Actually, what the president’s budget says is that NASA will have responsibility in the post JPSS-1 time frame, in 2021 or something like that, for sustained measurements of three basic quantities: vertical profiles of stratospheric and upper tropospheric ozone, solar irradiance, and Earth radiation budget measurements. Those are lumped under the jargon of “climate sensors.” I think NOAA made that jargon up. It’s the data, the information that’s important, not necessarily how we gather the data, or with which instruments. 

But do you have any ideas yet about how you might do that?

So NOAA’s Polar Free Flyer, which will launch before JPSS-1, has the Total Solar Irradiance Sensor on it. We would then have responsibility for providing TSIS-class measurements for solar irradiance in the post JPSS-1 time frame. The Earth radiation budget measurements are being made by the Clouds and Earth’s Radiant Energy System (CERES) instrument flying on JPSS-1. NOAA’s agreement is that JPSS-2 will continue to provide accommodations for a CERES-like instrument, to the extent that it doesn’t drive their program. We have responsibility for that instrument. Similarly, for the vertical profiles of ozone, NASA is paying for the Ozone Mapping and Profiler Suite (OMPS)-Limb instrument, which is a bolt-on to the OMPS instrument flying on JPSS. OMPS-Limb is relatively low-cost, so I don’t think it’s a giant engineering program to figure out how we do that.