For the past four decades, scientists at Colorado State University’s Natural Resource Ecology Laboratory (NREL) have helped shape how we understand and manage natural systems. The CENTURY Model, a pioneering terrestrial ecosystem simulation grew from a regional project in the 1980s into a globally recognized tool for ecological forecasting. The evolution of CENTURY, and its daily version DayCent, has transformed our understanding of soil dynamics and has become an important tool for ecosystem science, carbon assessment, and conservation planning.

Origin and Evolution

Developed by Dr. Bill Parton and colleagues in the early 1980s at NREL, the CENTURY Model was originally designed to evaluate how agriculture in the Great Plains affected soil carbon and plant production. “We started with wheat systems and grasslands,” Parton recalls. “What we found was striking–around fifty percent of the soil carbon was lost under agriculture within the first 50 years.”

Building the model required massive groundwork. Parton’s team compiled soil data from nearly 12,000 sites and plant production data from about 100 locations. Originally built with annual time steps, the model soon adopted monthly intervals to better capture seasonal dynamics.

As climate science gained urgency in the 1990s, the model’s scope expanded as well. “We went from just doing agriculture and grasslands to modeling systems across the world,” says Parton. CENTURY set itself apart by emphasizing soil nutrient availability, instead of temperature and radiation, like many other models at the time.

From CENTURY to DayCent and Beyond

The launch of DayCent in the late 1990s marked a significant advancement. While CENTURY worked on monthly time steps, DayCent operated daily – an essential shift for modeling greenhouse gases like nitrous oxide, which spike dramatically after events like rainfall or spring thaw.

“You can get half your annual nitrous oxide emissions in just a few weeks,” Parton explains. “You couldn’t capture that with a monthly model.”

DayCent’s comprehensive handling of biogeochemical processes and flexibility to be parameterized for multiple ecosystem types made it well-suited for national and global applications. NREL scientists like Dr. Keith Paustian adapted it into Comet-Farm, a platform now used by the USDA’s Natural Resources Conservation Service to help farmers estimate the climate impacts of different practices. Dr. Stephen Ogle used DayCent to support national greenhouse gas inventories.

Further innovations followed, including integrating eddy covariance data to represent photosynthesis, simulating deep soil carbon, and explicitly modeling microbial activity. Yet one guiding principle stayed constant: keep the model as simple as possible without sacrificing accuracy.

Models in Action

What began as a theoretical research tool has become a practical engine for land use management. “The main goal of using all these models is to figure out how we can reduce greenhouse gases from ecosystems,” says Parton.

One major use is helping land managers reduce nitrous oxide emissions, a greenhouse gas hundreds of times more potent than carbon dioxide. “If you add more fertilizer, you get more nitrous oxide,” he says. “The models help identify the sweet spot – where you still get strong yields without excessive emissions.”

DayCent’s ability to assess different management strategies, like slow-release fertilizers or optimized application timing, makes it a powerful decision-support tool. And with models now integrated into economic frameworks, researchers can weigh environmental and financial trade-offs. Projects like the Center for Advanced Bioenergy and Bioproducts Innovation (CABBI) use model outputs to identify farming practices that are both sustainable and economically feasible.

In a full-circle moment, the original vision of supporting land managers directly is being realized. Tools like Grass-Cast combine ecosystem models with satellite data to help ranchers forecast forage availability across the growing season, delivering science directly to the people who need it the most.

At an even broader scale, land use practices modeled by CENTURY and DayCent simulations are part of larger global climate models, enabling a more complete understanding of the carbon cycle, land use choices, and ecosystem feedbacks.

Looking Ahead

Despite the impressive evolution of the CENTURY and DayCent models over four decades, Bill Parton’s vision for their future remains focused on continuous improvement. “The key is just always doing a better job at what we’re currently doing,” he explains. Current efforts focus on reducing uncertainty in predictions, especially for soil carbon and nitrous oxide emissions, which remain difficult to measure precisely.

“We care about the applications,” he adds, “but I more enjoy working on improving the models than applying them necessarily.” That commitment to curiosity and scientific rigor has been a hallmark of the models’ development.

Yet Parton flags a growing concern: the increasing role of private companies in ecosystem modeling, particularly around carbon markets. “The biggest problem we have now is the commercialization of the greenhouse gas field,” he says. Proprietary interests sometimes prevent the publication of model parameters, which can undermine transparency and reproducibility, the foundations of the scientific process.

Despite these tensions, Parton remains proud of what the CENTURY and DayCent models have accomplished. “Developing ecosystem models that are used and applied by many people has been particularly rewarding,” he reflects, “especially since these types of models didn’t exist when I started.”

He credits the success to a combination of hard work, good timing, and strong collaboration, especially among NREL colleagues who have remained involved in the model’s journey for decades. “It’s luck and hard work,” he says, simply.

Forty years in, the CENTURY Model remains a cornerstone of ecosystem science, serving as a vital tool for researchers and land managers alike. Its ability to evolve, while staying rooted in sound science, has kept it relevant in a changing world, where ecosystem modeling is more critical than ever.

Beyond the Model: The Voices Who Shaped CENTURY

The enduring impact of the CENTURY and DayCent models stems not only from their scientific foundations but from the dedicated individuals who built and expanded them over decades.

Melannie Hartman, who joined NREL as a computer programmer in 1993, was tasked with developing DayCent alongside Bill Parton. “In my initial year at NREL, both Bill Parton and Dennis Ojima expressed interest in developing a daily version of CENTURY, and they asked me to work on it,” she recalls. “I worked with Bill at the computer, programming while he was describing the updates that were needed for the SOM model.” Together, they built DayCent piece by piece, incorporating daily soil models, soil temperature models, and nitrogen trace gas components based on Stephen Del Grosso’s PhD dissertation.

“The model worked well, but I still never thought DayCent would be more than a model NREL would use for research. I didn’t have the vision Bill had,” Hartman admits. Her 32-year career at NREL has centered around DayCent’s evolution, including developing GrassCast in 2016 and recently improving the model for bioenergy crops.

Dr. Stephen Del Grosso developed the trace gas components that made daily resolution essential. “The majority of nitrous oxide emissions often occur in pulses lasting a few days following rainfall or snowmelt events, which cannot be captured when using monthly weather,” he explains. His validation work helped establish DayCent’s credibility for the EPA’s National GHG Inventory, making the U.S. “the first country to use a process-based model to calculate emissions from agricultural soils.”

Dr. Stephen Ogle developed frameworks for using DayCent in national greenhouse gas reporting, while Ames Fowler continues pushing the model’s capabilities forward. “We’ve simulated over 70 climate-smart agricultural practices across the U.S. to inform biofuel and conservation policy,” Fowler explains. He also maintains the collaborative tradition through training sessions that draw scientists from corporate and academic sectors worldwide.

The collaborative spirit established by the model’s originators continues to define its community. As Fowler notes, “Just as the DayCent community has grown by ‘giving it all away,’ we now face challenges in standardization and fair sharing of data.” Despite these challenges, efforts to grow the community align with initiatives to standardize data, create calibration datasets, develop shared tools, and compile contributions from users worldwide.

These personal accounts illustrate how CENTURY and DayCent have become more than scientific tools—they’ve created a community dedicated to providing tools to help land managers navigate climate challenges with collaborative, evidence-based science.