Authors:
- Ashley Prentice, Forest Carbon Specialist, Colorado State Forest Service
- Ethan Bucholz, Ph.D., Forest Monitoring Program Manager, Colorado State Forest Service
- Tony Vorster, Ph.D., Research Scientist, Natural Resource Ecology Laboratory, Colorado State University
Introduction
Colorado’s diverse forest ecosystems are vital for water, wildlife, recreation, carbon storage and more. Forests absorb carbon through photosynthesis and release it through respiration, decomposition and combustion. When they absorb more carbon than they release, forests act as carbon sinks, reducing the amount of carbon dioxide in the atmosphere. Forests can also be carbon sources if they release more carbon than they absorb, often due to events like wildfires and insect outbreaks. Forests cycle between times of acting as carbon sinks and times of being carbon sources.
Increases in disturbances, current stocking levels and Colorado’s cold, dry climate, along with shifting climatic conditions and drought, can reduce forest productivity and challenge forests’ capacity to be net carbon sinks. Estimates suggest that Colorado’s forest ecosystems, like those in other western states, have been acting as a net carbon source rather than a net sink since 1990 (Domke et al., 2024).
This S&D Byte provides a summary of forest ecosystem carbon results from the Colorado Forest Carbon Inventory published in 2024 and provides some new summarizations of these results. With a focus on the aboveground live carbon pool, we:
- Briefly describe the methodology used for forest ecosystem carbon accounting,
- Highlight key results, and
- Interpret general trends.
There is a separate S&D Byte on Carbon Accounting of Colorado’s Harvested Wood Products.
The results of the 2024 Colorado Forest Carbon Inventory reflect changes measured over the last 20 years, yet these trends are shaped by longer-term processes such as management legacies (e.g., decades of wildfire suppression), a changing climate, ongoing disturbances, the wood products industry and other factors that influence forest carbon dynamics. Because many of Colorado’s forests grow slowly compared to more productive regions of the country (i.e., Pacific Northwest or the American South), even a nearly 20-year window represents only a brief chapter in longer ecological cycles.
The approach: Tracking tree growth, removals and mortality
We used repeat tree measurements to track forest carbon stocks and their changes through growth, removals and mortality. This approach adapts the forest ecosystem carbon inventory framework established by the California Department of Forestry and Fire Protection (Christensen et al., 2019) and incorporates Intergovernmental Panel on Climate Change (IPCC) guidelines.
The core data source for this analysis is the Forest Inventory and Analysis (FIA) program, a nationwide inventory maintained by the U.S. Forest Service (USFS). Since the late 1990s, the USFS has partnered with the Colorado State Forest Service (CSFS) to implement the FIA program, where several CSFS FIA crews conduct field inventory measurements across Colorado, Wyoming and other neighboring states. This program assesses the status and trends of forests across all land ownerships and forest types using a network of permanent, geographically unbiased field plots, spaced approximately one per 6,000 acres (Bechtold & Patterson, 2005).
In the Interior West, approximately 10% of FIA plots are measured annually, completing a full cycle every 10 years. Colorado’s annual inventory began in 2002 and completed its first full cycle in 2011. Plot remeasurement started in 2012, and the third cycle began in 2022. Although the third cycle is ongoing, this report uses data up to 2019, drawing from the first cycle (2002-2011) and eight years of remeasurement data (2012-2019). At the time of this analysis, only data up to 2019 were available.
We applied FIA’s methods for converting tree and plot measurements into estimates of biomass and carbon (Westfall et al., 2023). Carbon stocks were calculated using a full 10-year panel of data (2010-2019). The average annual net carbon flux was calculated from plots measured between 2002-2009 and remeasured between 2012-2019, considering only plots that remained forested during both periods.
By the numbers
- 3,954 FIA plots used to estimate the forest carbon stock (2010-2019)
- 2,941 FIA plots measured between 2002-2009 and remeasured between 2012-2019 to estimate carbon flux (stock-change)
- 253,485 individual trees measured on the ground by Colorado FIA crews and used to estimate growth, removal and mortality
Results and discussion
Key units
- Carbon stocks: Teragrams of carbon (TgC), equivalent to 1 million metric tons of carbon
- Carbon flux: TgC per year (TgC yr⁻¹), and represent an average annual value from 2002-2019
- Negative flux values represent a loss from that pool; positive flux values represent a growth in that pool
Colorado’s forests store a tremendous amount of carbon (1,552.6Tg C). This amount of carbon is equivalent to annual emissions from more than 1.3 billion cars or the amount of electricity used by nearly 2 billion homes in a year (U.S. Environmental Protection Agency, 2025). While these comparisons use emissions-based metrics, they help illustrate just how much carbon is stored in Colorado’s forests. The carbon accounting report provides a detailed breakdown of carbon stocks across all pools considered (see Vorster et al., 2024 for a complete description).
Here, we present growth, removal and mortality fluxes for aboveground live biomass at the statewide level, and we encourage readers to explore the CSFS Forest Ecosystem Carbon Accounting Dashboard for results at state, CSFS area and county scales. Additionally, we encourage readers to look at the report as the numbers in this analysis are focused on change to ONE carbon pool rather than the cumulative change across ALL carbon pools.
Carbon growth varies across Colorado’s forest types
When looking at carbon gains from tree growth alone, all of Colorado’s forests actively add carbon each year, but not equally (Figure 1 and Figure 2). Statewide, growth from live trees sequesters 5.04 TgC annually, with USFS lands accounting for 68% of that total (3.44 TgC yr⁻¹). Of Colorado’s eight national forest units, the Grand Mesa, Uncompahgre and Gunnison (GMUG) and White River national forests had the highest aboveground live tree growth rates, at 0.71 TgC yr⁻¹ and 0.67 TgC yr⁻¹, respectively.
Both differences in growth rates among forest types and the amount of area each forest type occupies across the state give rise to the patterns by forest type shown in Figure 1. We calculated the average growth rate per acre for the state’s most common forest types to compare how different forest types accumulate carbon more directly (Figure 2). Piñon juniper, our lowest elevation forest type, across the state added an average of 141.7 lbsC ac⁻¹ yr⁻¹. Ponderosa pine forests, our next lowest elevation forest type, added 451.4 lbsC ac⁻¹ yr⁻¹, while Douglas-fir added 573.0 lbsC ac⁻¹ yr⁻¹. Aspen and lodgepole pine forests accrued 794.2 and 681.5 lbsC ac⁻¹ yr⁻¹ respectively, while spruce-fir forests accrued the greatest carbon growth at 885.2 lbsC ac⁻¹ yr⁻¹ (Figure 2).
The carbon growth patterns across forest types reflect the ecology of Colorado’s forests. Growth rates vary across the state depending on elevation, climate, soils, tree species and forest age. The carbon growth rates generally increase with elevation in Colorado (Figure 2), where greater moisture availability supports more productive conditions, while arid environments at lower elevations limit carbon fixation.
The spruce-fir and aspen forest type groups, found in mid- to high-elevation areas, have the highest carbon growth rates across the state (Figure 1) and on a per acre basis (Figure 2). Annual carbon growth rates of spruce-fir forests (1.91 TgC yr⁻¹) and aspen forests (1.26 TgC yr⁻¹) together accounted for 63% of all forest carbon growth. Moisture availability is generally higher in spruce-fir and aspen forests than lower-elevation forests such as ponderosa pine, but shorter growing seasons limit spruce-fir growth, while seasonal leaf shed constrains growth in aspen forests. On a per acre basis, spruce-fir and aspen forests add nearly double the carbon annually through growth compared to ponderosa pine forests. Despite the fact that lodgepole pine and aspen forests occupy similar elevational bands, with aspen being more widely distributed, aspen has higher per-acre growth than lodgepole pine (Figure 2), a likely reflection of the niches each group occupies, with aspen typically occurring on more mesic sites relative to lodgepole pine, a more xeric species.
In contrast, piñon-juniper forests are found in lower elevations between the higher elevation conifer forests of the Rocky Mountains and the drier shrublands and grasslands of the plains and high deserts, where moisture is limited. While growth rates are slow in these forests, piñon juniper forests are the most extensive within the state. Notably, piñon-juniper forests capture more carbon through growth alone than ponderosa pine forests statewide (Figure 1), despite growth rates per acre that are only a third of ponderosa pine forests (Figure 2).
Disturbance-related mortality accounts for most carbon loss
Mortality in Colorado’s forests has been the largest driver of aboveground live carbon losses, with 8.48 TgC lost annually from 2002-2019, which is more than annual growth (5.04 TgC yr⁻¹). Disturbances, such as insect outbreaks, most notably mountain pine beetle and spruce beetle, along with wildfire, drought and disease cause tree mortality alone and in combination.
Patterns of mortality reflect the influence of large-scale disturbances across the state, which drive much of the variability observed statewide. During the study period, roughly one-third (34%) of the state’s forests were impacted by disturbance. These disturbance-affected forests acted as a carbon source, losing an average of -2.65 TgC yr⁻¹. Forests not affected by disturbance during this study period were net carbon sinks, gaining an average of 0.79 TgC yr⁻¹.
Insect and disease outbreaks were the primary drivers of disturbance-related carbon losses, responsible for 64% of the total flux across disturbed lands (-1.69 TgC yr⁻¹; Figure 3). While FIA data do not specify insect or disease type, Colorado has experienced two major bark beetle epidemics during this period. The mountain pine beetle outbreak peaked during the first measurement period (2002–2009), having the greatest impact on lodgepole pine forests and a less severe impact on ponderosa pine forests. The spruce beetle epidemic peaked during the second measurement period (2012–2019; Figure 4), mainly impacting spruce-fir forests. Annual mortality rates and associated live aboveground carbon loss were highest in spruce-fir (–3.94 TgC yr⁻¹), lodgepole pine (–0.99 TgC yr⁻¹), and aspen forests (–2.08 TgC yr⁻¹; Figure 1). Aspen was impacted during the study period by a combination of drought, herbivory, insects and disease collectively termed “sudden aspen decline.”
The relative severity and timing of bark beetle disturbances can also be seen in these data. For example, the Rio Grande and San Juan national forests in southwest Colorado show some of the largest mortality-related carbon losses, while the GMUG exhibits a near zero net flux during the study period (See Figure 11 and Appendix Table 7 in the Report), likely reflecting later outbreak timing. On the Rio Grande National Forest, aboveground dead stocks were about half of live stocks, compared to about one-seventh on the GMUG and about a quarter on the San Juan (Appendix Table 6), illustrating both the relative severity and timing of mortality captured by FIA measurements.
Fire was also an important contributor, accounting for 20% of total disturbance-related carbon losses (-0.56 TgC yr⁻¹; Figure 3). This inventory reflects impacts from past fires and fires between 2002 and 2019. Notable fire events during the study period included the Hayman Fire (2002), High Park Fire (2012), West Fork Fire Complex (2013) and Spring Creek Fire (2018), with annual acres burned shown in Figure 3. The analysis does not capture impacts from Colorado’s historic 2020 wildfire season; future updates to the inventory will slowly capture the impacts of these wildfires and their effects on forest carbon.
The USFS manages the largest share (48%) of Colorado’s forested lands, and these lands account for the majority of tree mortality, with nearly 80% of aboveground carbon loss from mortality occurring on USFS lands (-6.68 TgC yr⁻¹). Several northern forests (Arapaho-Roosevelt, Medicine Bow-Rout, and White River national forests), were heavily affected by the mountain pine beetle epidemic, with losses of -0.98, -0.94 and -0.86 TgC yr⁻¹, respectively. In the southwest, the Rio Grande and San Juan national forests also saw high mortality (-1.38 and -1.15 TgC yr⁻¹, respectively), likely reflecting the influence of beetle epidemics in those areas as well.
These disturbances have resulted in significant transfers of carbon from live aboveground trees into dead and down woody material pools and to the atmosphere. The accumulated carbon in the dead and downy woody material pools will slowly decompose and be released to the atmosphere and, to a lesser degree, will be added to soil carbon. Some disturbed forests quickly recover to becoming a carbon sink through tree regeneration and increased growth in surviving trees, while other forests with severe mortality and/or a lack of tree regeneration will remain carbon sources.
Harvest removals are a small but important part of Colorado’s forest carbon inventory
Removals in Colorado’s forests, primarily from timber harvesting, accounted for a relatively small portion of carbon flux compared to mortality and growth (Figure 1). Statewide, removals averaged a loss of -0.38 TgC annually, with the majority occurring on USFS (-0.19 TgC yr⁻¹) and private lands (-0.15 TgC yr⁻¹). The Medicine Bow-Routt and Arapahoe-Roosevelt national forests had the highest average annual removals of aboveground live material, at –0.06 TgC yr⁻¹ and –0.05 TgC yr⁻¹, respectively.
Removals were greatest in spruce-fir, ponderosa pine, aspen and lodgepole pine forest types (Figure 1). Higher removals in these forest types reflect their overlap with Colorado’s wildland-urban interface (WUI), where fuels treatments reduce wildfire risk, and the fact that these forest types more often contain merchantable material for wood products. Piñon-juniper forests, Colorado’s most extensive forest type, have lower removals because their size and structure limit removable volume, and markets for this material are limited.
According to the harvested wood products carbon model, an average of 0.14 TgC yr⁻¹of harvested material was transported to mills between 2002 and 2019. These estimates reflect that not all cut material is transported, as some remains on site, related to weak wood products markets and/or project prescriptions, contributing to increases in dead wood and other pools. Some of the differences between the removal flux and the annual harvest estimates is also due to the differences in data sources between the forest ecosystem model (FIA plots) and harvested wood product model (mill surveys).
Removals represent a minor emission in Colorado’s overall forest carbon dynamics, but their role is important to account for because harvested material can provide long-term carbon storage in wood products. For more detail, see Carbon Accounting of Colorado’s Harvested Wood Products.
Conclusion
Colorado continues to hold large amounts of carbon in its forests. However, the live aboveground carbon pool (the focus of this article) shrunk from 2002 to 2019 due to mortality far outpacing growth and removals. This live tree carbon moved to dead and down carbon pools and some was emitted to the atmosphere. The average annual loss of carbon to the atmosphere (-0.87 TgC yr-1) is small (0.06%) relative to the total carbon stored in Colorado’s forests (1,552.60 TgC). For full results and interactive exploration, refer to the full report and the CSFS carbon dashboard.
Forests are known to cycle through periods of serving as a carbon sink and periods of being a carbon source during and after disturbances. This inventory is a snapshot in time, so temporal context is essential for interpreting the results. The state’s carbon stocks were relatively high at the start of the study period in the early 2000s due to a combination of factors. These include a century of fire suppression beginning in the late 19th century that increased forest density and forest extent, along with periods of relatively favorable moisture and temperature conditions in the late 19th and early 20th centuries that supported seedling establishment (e.g., Battaglia et al., 2018; League & Veblen, 2006). From this elevated baseline at the start of the study period, forest carbon stocks declined due to periods of high disturbance, suggesting that current stocks in many areas exceed carbon carrying capacities under the historic range of variability. Yet, forest carbon dynamics are highly variable across the state, which is important to keep in mind when interpreting results. Statewide summaries can mask important local differences, such as variation by region, ownership or disturbance trends. For example, spruce-fir forests functioned as a sink in the northeast area while acting as a source in the southwest area.
We intend to rerun this inventory periodically with updated forest inventory data, which will, in time, continue to describe the state’s forest carbon dynamics. Until then, we can conjecture how the state’s carbon stocks are likely to change. Some areas, such as those recovering from bark beetle outbreaks, may return to net sinks as smaller trees grow and in-fill, replacing killed trees and accumulating carbon through growth faster than carbon is released through decomposition. Other areas, such as some forests that burned at high severity, will continue to act as carbon sources as wood decomposes and tree regeneration is limited by harsh conditions or lack of a seed source. Shifting climatic conditions are also expected to play a major role in shaping Colorado’s forest carbon stocks. Warmer temperatures and drought can stress trees and increase susceptibility to insects and disease, reducing growth and increasing mortality. Along Colorado’s Front Range, increasing mountain pine beetle outbreaks in ponderosa pine forests are expected to drive additional mortality and associated carbon losses. Hotter, drier conditions are also expected to increase fire weather and the likelihood of wildfires, which would increase emissions from forests. Finally, trends in forest carbon will continue to vary across the state depending on forest type, site conditions and disturbance activity.
Trends from the last two decades show that Colorado’s forests are, overall, in a period of declining carbon stocks. Continued monitoring will reveal whether, where and how quickly forests recover, but recent trends and anticipated future conditions suggest that forests may remain a net carbon source in many areas in the foreseeable future. Ultimately, at a statewide scale, current conditions in Colorado’s forests and the nature of ongoing disturbances constrain the ability of the forest sector to offset emissions from other sectors (e.g., transportation, electricity, building and industrial energy use) for the foreseeable future.
Forest carbon is a valuable indicator to track over time, but it is only one dimension of forest health. Carbon-rich forests are not necessarily resilient forests. Efforts are underway to identify project-level practices that help align forests’ carbon levels with long-term resilience, alongside the benefits we already manage for (e.g., wildfire risk, wildlife habitat, watershed health, forest products). In some cases, this may mean reducing overly high live-tree stocks to lower disturbance risk and support forest persistence. In other cases, it may involve planting seedlings to increase forest cover and support recovery of carbon stocks. Putting it all together, Colorado’s forest carbon story is one of substantial existing stocks and highly variable, disturbance-driven change, underscoring the importance of continued, project-level monitoring to align carbon outcomes with the broader goal of sustaining resilient forests and the benefits they provide.
More information
Download the full report and explore the forest ecosystem dashboard.
References
Battaglia, M. A., Gannon, B., Brown, P. M., Fornwalt, P. J., Cheng, A. S., & Huckaby, L. S. (2018). Changes in forest structure since 1860 in ponderosa pine dominated forests in the Colorado and Wyoming Front Range, USA. Forest Ecology and Management, 422, 147–160.
Bechtold, W. A., & Patterson, P. L. (2005). The enhanced forest inventory and analysis program – national sampling design and estimation procedures. Gen. Tech. Rep. SRS-80. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern Research Station. 85 p.
Christensen, G. A., Gray, A. N., Kuegler, O., Tase, N. A., Rosenberg, M., Loeffler, D., Anderson, N., Stockmann, K., & Morgan, T. A. (2019). AB 1504 California Forest Ecosystem and Harvested Wood Product Carbon Inventory: 2017 Reporting Period. U.S. Forest Service agreement no. 18-CO-11052021-214, 17-CO-11261979-086, California Department of Forestry and Fire Protection agreement no. 8CA04056 and 8CA03714 and the University of Montana. California Department of Forestry and Fire Protection and California Board of Forestry and Fire Protection.
Domke, G. M., Walters, B. F., Smith, J. E., Greenfield, E. J., Giebink, C. L., Ogle, S. M., Steller, J., Rewcastle, K., Knott, J. A., Coulston, J. W., Heilman, K. A., & Lang, A. K. (2024). Greenhouse gas emissions and removals from forest land, woodlands, urban trees, and harvested wood products in the United States, 1990–2022. Resource Bulletin. WO-102. Washington, DC: U.S. Department of Agriculture, Forest Service, Washington Office. 13 p.
League, K., & Veblen, T. (2006). Climatic variability and episodic Pinus ponderosa establishment along the forest-grassland ecotones of Colorado. Forest Ecology and Management, 228(1–3), 98–107.
U.S. Environmental Protection Agency. (2025, February 24). Greenhouse gas equivalencies calculator.
Westfall, J. A., Coulston, J. W., Gray, A. N., Shaw, J. D., Radtke, P. J., Walker, D. M., Weiskittel, A. R., MacFarlane, D. W., Affleck, D. L. R., Zhao, D., Temesgen, H., Poudel, K. P., Frank, J. M., Prisley, S. P., Wang, Y., Meador, A. J. S., Auty, D., & Domke, G. M. (2023). A national-scale tree volume, biomass, and carbon modeling system for the United States. U.S. Department of Agriculture, Forest Service.