Monday marks the culmination of a remarkable year for weather, weather forecasting and the great American collaboration we call the weather enterprise. It also marks the beginning of a new journey, requiring even greater cooperation for years to come.

It was Nov. 19, 2016, when the revolutionary Geostationary Operational Environmental Satellite-16 (GOES-16), the first satellite in the National Oceanic and Atmospheric Administration’s GOES-R series, launched from Cape Canaveral, Fla. Thousands of us who helped in some way to develop the satellite watched it take flight into the evening sky, but no one could know the historic 2017 hurricane season that would follow.

On Monday, GOES-16 was declared operational, after spending much of its beta period watching over a constant stream of hurricanes, each one seemingly more devastating than the last.

Meanwhile, NOAA’s Joint Polar Satellite System-1 (JPSS-1), America’s newest polar-orbiting weather satellite, is now circling the globe 14 times per day after successfully launching in the dark of night from Vandenberg Air Force Base, Calif., in the early hours of Nov. 18 — almost exactly one year after the GOES-16 launch.

Together, these next-generation geostationary and polar-orbiting satellites will deliver the biggest one-two punch ever for weather forecasting.

From its constant perch over the Western Hemisphere, GOES-16 is already helping meteorologists issue more accurate and earlier warnings for severe thunderstorms, tornadoes, flooding and hurricanes because of its greater resolution and faster refresh than previous GOES satellites.

JPSS-1 collects data globally over land and sea to increase the accuracy of computer weather-model forecasts three to seven days in advance, especially important for extreme weather such as hurricanes and winter storms.

Even as we celebrate the arrival of our nation’s most advanced weather satellites ever, we must move now to plan and develop their replacements. That process starts in earnest with the imminent release of two landmark publications: the National Research Council’s 2017-2027 Decadal Survey for Earth Science and Applications from Space, and NOAA’s Satellite Observing System Architecture study. While these documents will serve as important guides for prioritizing weather and environmental observations, successful implementation will ultimately depend on connecting observations to the specific forecast needs of people and industries.

The road to what comes next is filled with more opportunities than ever before: Rapid advancements in sensor and ground processing technology will continue to increase the amount of data collected and the speed at which it’s delivered; new forecast models and more powerful supercomputers will assimilate more data and produce more frequent forecasts at finer resolutions; and machine learning will extract new forms of weather intelligence for an expanding range of uses.

There are plenty of challenges ahead, as well: Budget uncertainties will continue to threaten gaps in the core satellite programs comprising the backbone of our global observing system; commercial weather satellites carry the potential to augment and enhance traditional government systems, but market viability and data-sharing policies remain open questions; and merging more data from an increasingly diverse number of sources will require innovative approaches to new ground systems and data-processing techniques.

Weather forecasting has made tremendous strides despite what has sometimes been a technology-first approach, where a sensor capability is developed in the hope of finding a mission and market. As we look toward the future, we must shift to a mission-first mentality, where end-user needs drive technology development. What are the data and information requirements of a world where 1 in 4 vehicles on the road could be self-driving, where autonomous drones may be just another way to hitch a ride, and where the Internet of Things promises to infuse weather intelligence into virtually every device we use at home and away?

Evolving our weather infrastructure to meet the demands of these and other emerging markets is only possible if we double down on the joint effort among government, industry and academia that got us to where we are today.

That means greater engagement between government and industry to identify weather requirements and drive research and development investment; partnerships between established companies and new space start-ups to get innovative technologies to space faster and cheaper; and regular dialogue with a broad set of end users, ranging from insurance companies to autonomous vehicle makers, to understand the forecast needs of the future.

Just as we couldn’t know a year ago the series of destructive hurricanes that would arrive by summer, we don’t know exactly what weather lies in the coming year and beyond. But we do know that a well-coordinated weather enterprise, acting with clear missions and goals in mind, gives us the best shot of predicting it with the accuracy, detail and lead time desired by a world becoming more affected by weather every day.

Eric Webster is vice president and general manager of Harris Corp.’s Environmental Solutions business, which built the GOES-R series main instrument, the Advanced Baseline Imager, the GOES-R ground system and the JPSS series Cross-track Infrared Sounder.