About

History

After more than three decades, the Atmospheric Radiation Measurement (ARM) User Facility is still collecting and providing data needed to answer basic atmospheric science questions and improve earth system models.

ARM’s Uniquely Continuous Atmospheric Observations

In 1989, the U.S. Department of Energy (DOE) Office of Science established ARM. From its home within DOE’s Biological and Environmental Research Program, ARM organized around a task that was both big and new: to establish, over time, a series of instrumented platforms that would provide an observational basis for studying earth systems.

To that end, ARM collects, processes, quality-checks, stores, and distributes (at no cost) continuous atmospheric measurements gathered 24 hours a day in a variety of meteorological regimes, with a strong focus on the United States.

ARM continues to refine its mission in response to DOE priorities and advances in science and technology, while remaining focused on advancing a robust predictive understanding of earth systems. Such predictions are needed to provide solutions to the nation’s energy challenges.

For its entire history, ARM has worked to better the performance of earth system models, which are important for understanding atmospheric conditions affecting the nation’s energy supply. Largely, ARM has done that by improving how such models accurately incorporate the details of radiative energy transfer and the impact of clouds and aerosols.

How ARM Came to Be

By the 1970s, concern was mounting over increasing levels of carbon dioxide in the atmosphere. DOE and the scientific community at large were also troubled by apparent uncertainties within earth system models.

Between 1982 and 1990, federal agencies commissioned intercomparison studies on one major source of these uncertainties: models of radiation in clouds. When heat energy radiates toward Earth from the sun, clouds scatter, absorb, or reflect that energy in ways that are hard to observe and measure.

Some simulations of radiative transfer differed in how clouds and the aerosols that make them influence changes in surface temperatures.

One difficulty was “feedback.” That’s a term for how one initial change (the increased presence of a gas in the atmosphere, for instance) can cause further changes, including the reflective properties of clouds. A “feedback loop” describes how that one initial change gets amplified or diminished over time.

DOE and scientists at its national laboratories determined that the key to improving theoretical models was real-world observations at well-instrumented ground research sites covering a range of atmospheric conditions.

DOE tasked ARM with collecting observations that would inform and improve the representation of cloud radiative feedback in models. In 1992, ARM began operations at its first site, the Southern Great Plains atmospheric observatory in storm-crossed Oklahoma.

Historical Documents & Articles

What ARM Added

Collecting and recording simultaneous measurements of comprehensive atmospheric properties for scientific investigation is still part of ARM’s mission today. The main reason is that direct, continuous observations will help unravel uncertainties in mathematical models designed to predict how clouds and precipitation respond to changes in atmospheric conditions.

In the years after its founding, ARM also developed other ways to acquire data on cloud and radiation properties, including shorter-term field campaigns proposed by scientists to investigate specific scientific questions. Such experiments, large and small, have been conducted or hosted by ARM in a variety of locations. Their archived measurements provide further sources of data, which are available to any earth system researcher.

Beyond fixed-location observatories, ARM collected measurements with mobile observatories (starting in 2005) and airborne assets, including crewed aircraft, uncrewed aerial systems, and instrumented tethered balloons. Collectively, such observational data (coupled with satellite partnerships) provided a unique opportunity to move away from idealized case studies and investigate the wide range of conditions that occur in nature.

Since beginning operations, ARM has enabled scientists to answer fundamental science questions and significantly improve predictive mathematical simulations of radiative heat transfer, aerosol, cloud, and precipitation processes. ARM data allow scientists to use observations to evaluate and test their models.

ARM was the first atmospheric research program to deploy comprehensive suites of sophisticated instrumentation to obtain continuous measurements of cloud, aerosol, and radiative properties. This strategy revolutionized how scientists leveraged continuous observations to develop long-term statistics of detailed cloud properties. Today, ARM is a model and leader for observational programs around the world.

How ARM Grew

In its 30-plus years of data collection, ARM has used its observation sites and strategies to support DOE scientific priorities. It became a national scientific user facility; coordinated hundreds of large and small field campaigns domestically and internationally; established mobile instrument platforms; continuously fine-tuned its data-flow systems; started an aerial instrumentation program; greatly expanded its toolbox of instruments and software; widened and streamlined access to its data; and joined in collaborations with other agencies and ground-based networks to advance data sharing and make data more usable to modelers and satellite users.

ARM took other steps to strengthen the ties between observations and models. It integrated large-eddy simulation (LES) modeling into its data capabilities to bridge the gap between localized high-resolution observations and models on the global scale. Due to computational limitations, global models create simulations only on very coarse scales, using three-dimensional grids spanning tens to hundreds of kilometers. Smaller-scale models use grids as small as 100 meters tall, wide, and deep.

Starting in 2015, ARM further coupled its LES modeling capabilities with its observational data through the LES ARM Symbiotic Simulation and Observation (LASSO) activity. LASSO is an ever-expanding library of high-resolution simulations and data bundles, allowing researchers to test earth system models against curated representations of atmospheric regimes. As of 2025, LASSO’s library of simulations focuses on regimes of shallow, deep convective, and marine clouds.

A range of modeling partnerships includes a library of ARM data cases within DOE’s Energy Exascale Earth System Model (E3SM), with a focus on ARM’s time-tested single-column model. E3SM’s diagnostics package also integrates the ARM Data-Oriented Metrics and Diagnostics Package (ARM-DIAGS), allowing users to compare E3SM simulations with data from multiple ARM sites and field campaigns.

ARM now has three fixed-location observatories in Oklahoma, Alaska, and the eastern Atlantic Ocean. These long-term observatories enable measurements of interannual variability, including the characteristics of extreme events. ARM also has three short-term, mobile observatories to observe a broader set of weather conditions and locations, such as storm systems common in Alabama.

Meanwhile, ARM’s data archive keeps growing. As of mid-2025, more than 18,000 data products are available to users.