This article originally appeared in the April 8, 2019 issue of SpaceNews magazine.
What a difference a year makes. In late April 2018, NOAA discovered the primary instrument on its newest multibillion-dollar weather satellite was failing to make key observations for hours every day. At the same time, the U.S. Government Accountability Office continued to cite NOAA and the Defense Department’s campaigns to update weather satellite fleets as one of the federal government’s highest risk endeavors.
This year, NOAA has some good news to share. Working with industry and international partners, NOAA has found ways to mitigate problems with Geostationary Operational Environmental Satellite (GOES)-17’s Advanced Baseline Imager (ABI) and obtain 97 percent of the data originally sought. In addition, the congressional watchdog agency dropped weather satellite programs from its high-risk list, citing significant progress by NOAA in updating its fleet. In 2017, NOAA launched the first satellite in its new geostationary fleet, GOES16, plus the first spacecraft in its Joint Polar Satellite System (JPSS). In 2018, it launched GOES-17.
To plan future constellations, the agency is conducting the NOAA Satellite Observing System Architecture (NSOSA) study. What has it revealed so far? And how will commercial satellites and data products feed into future architectures? SpaceNews correspondent Debra Werner spoke with Stephen Volz, NOAA assistant administrator for satellite and information services, to find out.
What’s the latest on the NOAA Satellite Observing System Architecture study?
Since we rolled out of the NSOSA study a year ago, we have looked more deeply into the nuances of what we saw in those analyses and started acting on the findings. I’ll give you some examples. Part of NSOSA was an analysis of the strength and health of our existing constellation so we know when renewal will be required and what aspects of it may be more important than others.
One of the things we saw is there’s a strong need for integrating data from multiple observations, multiple platforms from multiple agencies and multiple nations. We already are doing that, particularly in low Earth orbit where we are using a dozen or more satellites from different agencies operationally. The NSOSA study said to count on that increasing and to be ready for it. We are focused on how to bring in data very efficiently from multiple organizations and satellites with zero latency through Secure Ingest. It is a pilot for how we expect to bring commercial data — and European, Japanese and Indian data — into our systems in the future.
Does NSOSA also reveal priorities for future constellations?
Yes. We looked at our low Earth orbit and geostationary orbit needs and realized that geostationary orbit is something we need to replenish sooner. We will look for opportunities for creativity or disaggregation later. Geostationary orbit is a place where there’s not a lot of immediate innovation in small satellites. On the flip side, there’s a lot of dynamic activity in low Earth orbit. Part of our effort this year and going forward is to experiment and partner with different players in the small satellite and medium satellite regime to explore operations or observations from low Earth orbit to complement the JPSS platform. You don’t jump into a major procurement right away. NSOSA indicated the value of disaggregation; the value of integration of multiple datasets in low Earth orbit.
What are your plans for the Commercial Weather Data Pilot?
We’re conducting phase two of the Commercial Weather Data Pilot. We have three vendors under contract, GeoOptics, PlanetiQ and Spire Global, to provide radio occultation data this fiscal year. We are currently collecting data from one of the three. The periods of performance are a little different for the other two, but we expect data very soon. We are evaluating those data sets for operational use. Can they work for us in our operational mandate, which means it has to be global, it has to be with low latency and it has to be continuous? We expect to make a call by the end of this year.
I noticed funding in the 2020 NOAA budget request to purchase commercial weather data.
We knew that if we did have a positive outcome and we didn’t have a budget, we would have a problem. We wanted to make sure we were prepared for success. We are not guaranteeing it.
How did you resolve ABI problems?
When the GOES ABI anomaly first showed up around the last week of April 2018, it was pretty scary. We thought our entire cooling system was down. Then, a lot of people in NASA, NOAA, industry and our international partners looked at all aspects of how the instrument works, how the satellite works, how the system works and how the information is used by the weather service.
I’ll pinpoint three places where we mitigated the impact of the loss. On the instrument side, we cranked up the power on the cryocooler, which was still working. The passive system, the loop heat pipe, had the fault. That was a significant portion of the recovery. We were operating the instrument at 60 degrees Kelvin. We had a lot of margin in the signal to noise ratio at 60 degrees Kelvin. We adjusted it for each of the different detector channels and found that some could operate as high as the low 80s, lowering the cooling requirement. Then, we looked at other available services or systems. Our Japanese partners have the same instrument on Himawari 8. It observes all the way to the eastern Pacific. It doesn’t get to the continental U.S. or much of Alaska, but it covers Hawaii and the central Pacific very well. We’re developing the ability to have Himawari 8 data available to us and to the weather forecasters.
The third part was to look at the products used by weather forecasters and modelers to determine which are affected and which are not affected. Often a product might use pieces of several instruments or several channels. It might use four GOES channels and only one was affected by the thermal issue. We looked at how to mitigate the loss of that one channel.
What was the problem with ABI?
We believe particulates got into the fluid, which clogged the filters which shut down the loop heat pipe. We’re in the process of fixing that for GOES-T and GOES-U. We think GOES-17 will have a long life in the current condition. It’s providing excellent service to our users. The weather service forecasters, in Alaska and Hawaii in particular, have been thrilled by the GOES West performance they’re getting.
Will the anomaly shorten the lifetime of the instrument?
It will. We still expect to exceed the 10- year lifetime requirement even with the elevated power draw on the cryocooler. We’ve used margin that was built into the system to generate the maximum level of performance we think is prudent given the expectations for the mission. Everything else on GOES-17 is working just fine.
What impact will the ABI anomaly have on future GOES satellites?
Fixing ABI led to a delay in the launch of GOES-T. We should be deciding in the next couple of months the new target launch date for GOES-T. We are deciding what recovery, repair or refurbishment we want to do on GOES-T ABI, as well. It will delay the launch of GOES-T from one and a half to two years.
What’s next in terms of updating NOAA’s fleet?
With JPSS-1 (now NOAA-20) and Suomi NPP, we have two polar satellites with comparable performance. With GOES-16 and GOES-17, we have two next-generation geostationary satellites with comparable performance. That is the benchmark for future JPSS and GOES satellites. They will carry us into the early 2030s. We also know it takes a while to develop new capabilities. And as I mentioned, we recognize — particularly in geostationary orbit, which requires two functioning satellites and one spare for a robust constellation — that it’s time to think about what we’re going to fly next.
In addition to continuing to deliver all of the JPSS and GOES satellites on cost and on schedule, NSOSA has indicated strategically attractive places for innovation and development. We will explore low Earth orbit constellation innovation to complement JPSS and the next step in geostationary. That’s on the platform side. The greatest growth that we expect to see is on the information technology side and the ground system side as we start integrating our new observations with other available observations to develop merged and integrated data products.
We already have a heavily disaggregated satellite system. We are not currently exploiting the satellites to their fullest extent to meet NOAA and U.S. needs. We are also looking at how we work in the cloud for data processing and data integration. And we are looking at how we can merge individual satellite observations into products. We are looking at combining low Earth and geostationary orbit observations to draw on the benefits of both to give us merged products. These are not numerical weather forecast products, but they are forecast products. With a low Earth orbit satellite, you have high resolution of floods and fires but the latency is 90 minutes and you might not be on the same ground track for a day.
Geostationary satellites, on the other hand, don’t give you high resolution but they look at floods and fires every five to 10 minutes. We have developed combined products using GOES ABI and NOAA’s Visible Infrared Imaging Radiometer Suite to show how we can use the resolution of one with the timeliness of the other in merging products. We’re able to provide flood maps in near real time to emergency managers on their timescale. Not just once a day, but once an hour.
Do you have other plans for commercial partnerships?
When we talk about innovation in low Earth orbit, that’s a place where I expect to be working in different ways with the commercial sector. Last fall, we issued a Request for Information about flying commercial satellites to low Earth orbit on the JPSS-2 launch vehicle. We would provide the launch at no cost and in exchange we would get the data for free for testing and operations. They get a free ride to space. And then together we have a science collaboration on exploitation. That’s one of the places we’ve reached out to the commercial sector.
If you look at the National Integrated Drought Information System Reauthorization Act, the new reauthorization of the weather budget made in January this year, it provided NOAA with other transactional authority, which gives us a different way to do cooperative research and development with commercial sectors. We’re looking to see how we can use that as we explore new possibilities in low Earth orbit. Our traditional approach would require specification, requests for proposals and probably five years before we go from concept to any hardware. Those are not exactly commercial partnerships but it is co-investment.