Helping Managers Plan for Climate Change with Remote Sensing at Canyonlands National Park

Sunrise over The Needles. NPS/Emily Ogden

In fall 2019, National Park Service staff in southern Utah noticed something strange about the landscape. Some of it was the wrong color. The region’s dark-green pinyon-juniper communities usually contrast sharply with the red-orange rocks of canyon country. But now, some of the juniper trees were also orange.

In 2019, the death of juniper trees in southern Utah led to questions about whether the same could happen elsewhere on the northern Colorado Plateau. NPS/D. Witwicki.

They later learned the trees were dying—during one of the driest periods in 1,800 years (Gangopadhyay et al. 2022). Another tree-killing drought had occurred in southern Utah parks in 2002–2003. In combination, these two droughts killed trees that had survived for centuries.

This raised some troubling questions. Should park managers expect more of the same in the future? Could pinyon and juniper trees in other areas be affected? Which other vegetation types might change—and when and where? And what, if anything, could be done about it?

If we had a crystal ball, how might our approach to management change?

We know long-term monitoring creates a record of the past. But it also provides a window into the future. Tracking which kinds of vegetation are changing, where they grow, and which aspects of climate are associated with the changes can help us understand what, when, where, and why future changes may occur, based on climate projections.

Linking satellite observations of vegetation condition with climate data over time can reveal how climate influences plant production (growth) and phenology (timing of growth). It can also tell us how sensitive different plant communities are to changes in weather and climate. With information like that, we won’t need a fortuneteller.

To see how this might work, Northern Colorado Plateau Network scientists used satellite imagery to analyze vegetation condition over time at parks on the northern Colorado Plateau, including Canyonlands National Park, and explored the relationships between climate and vegetation, using new tools for assessing climate effects. The results help us understand what to expect over the next few decades, and provide park managers with time and tools to plan for a range of scenarios.

Phenology and Production

Satellite images collected over time show us when plant growth begins in spring (“greenup”)—because the landscape turns green. At the end of the season, it senesces and turns brown (“browndown”). Linking the timing of greenup and browndown to weather provides insight about how weather affects phenology.

We can use those same satellite images to measure plant production. In combination, phenology and production tell us a lot about plant relationships with climate. And because plants are the biological engines of parks, this analysis tells us a lot about the impacts of climate change on park ecosystems.

NCPN alliance groups analyzed for trends in production, phenology, and climate relations at Canyonlands National Park.

At Canyonlands National Park, we used satellite imagery from remote sensing to track phenology and production across 18 vegetation groups in and near the park from 2000 to 2019:

• Alkaline Wet
• Blackbrush
• Disturbed
• Douglas-fir
• Dry Wash
• Geology
• Annual Exotic Grassland
• C3/C4 Grassland
• Juniper
• Pinyon-Juniper
• Ponderosa Pine
• Riparian
• Dry Sagebrush
• Dry Shrubland
• Mixed Montane Shrubland
• Wet Shrubland
• Sparsely Vegetated
• Water

Phenology

By the end of the study period, the growing season was starting 12.6 days earlier on average across all alliance groups, and ending 11.2 days later. The length of growing season was 23.8 days longer on average.

Production

• Multiple indicators suggest annual trends in vegetation production increased in 58% and decreased in 2% of the area analyzed in and near Canyonlands NP between 2000 and 2019.
• Growing-season production increased in the Blackbrush, Douglas-fir, Dry Wash, Geology, Juniper, Pinyon-Juniper, Ponderosa Pine, Dry Shrubland, Mixed Montane Shrubland, and Sparsely Vegetated alliance groups. The largest area of increase occurred in the Pinyon-Juniper and Sparsely Vegetated alliance groups, representing 44 and 7.6% of the study area, respectively.
• No downward growing-season trends were observed. Small areas that decreased in annual production included Blackbrush, Dry Shrubland, Juniper, Dry Wash, Pinyon-Juniper, and Sparsely Vegetated alliance groups, representing only 1.5% of the study area.

Climate Conditions

We used a water-balance model to determine how different aspects of climate and site characteristics affected vegetation phenology and production. To calculate the variables, we used 1-kilometer, gridded, Daymet temperature and precipita­tion data.

The US Southwest has been in an extended drought over the past 20–30 years. However, even during extended drought there can be periods of increased precipitation. This study, which began in 2000 and ended in 2019, needs to be considered in that context. The years 2000–2003 were among the driest between 1980 and 2019. In this study, the initial extremely dry conditions set the stage for subsequent trends in which precipitation increased (from the near-record low in 2000), resulting in increases in vegetation growth relative to conditions at the start of the study. This explains the increase in vegetation growth observed in 58% of the study area.

Taking the long view, an interesting decadal oscillation was evident: peaks in the middle of each decade since 1980 translated to similar peaks in other water-balance variables. If the periodicity of these cycles continues in the future, it could serve as a useful guide for timing restoration efforts that require substantial planning efforts, long-term investments, and multiple years of minimal drought stress for germination and establishment.

A wet spring fills sandstone potholes at Canyonlands National Park. NPS/Amy Washuta

Relationships of Climate and Vegetation

With this information about phenology, production, and climate in hand, we explored which climate variables were the most important drivers of phenology and production, which vegetation types were most sensitive to changes in water availability, and which types were most resistant to drought.

Climate drivers of production and phenology

• Annual growing-season production was most strongly correlated with annual precipitation and soil moisture, suggesting that increases in annual precipitation may have caused some of the increase in production over the study period.
• Two to three years of soil moisture was the best indicator of annual production at the park scale. This was also generally true at the alliance-group scale.
• The Douglas-fir, Mixed Montane Shrubland, and Disturbed alliance groups required the most precipitation to maintain production, whereas Water, Alkaline Wet and Ponderosa Pine required the least precipitation to maintain production.
• The primary determinants of the start of the growing season were growing degree days and precipitation.

Vegetation sensitivity to climate

The way vegetation at Canyonlands National Park responds to various climate drivers reflects the different plant communities and the site conditions where they grow. The same amount of change in precipitation, for example, may cause different responses from different vegetation types, or even in the same vegetation type growing in different soils.

The rate of change in vegetation condition per unit change in a climate variable is known as its climate sensitivity. More-sensitive vegetation types respond more quickly or more strongly to changes in climate conditions. Understanding a vegetation type’s climate sensitivity is critical to understanding its vulnerability to future climate change.

• Vegetation types at Canyonlands National Park, ordered from MOST to LEAST sensitive to precipitation:
  • Alkaline Wet
  • Water
  • Wet Shrubland
  • C3/C4 Grassland
  • Riparian
  • Annual Exotic Grassland
  • Dry Wash
  • Dry Sagebrush
  • Dry Shrubland
  • Sparsely Vegetated
  • Blackbrush
  • Disturbed areas
  • Juniper
  • Ponderosa Pine
  • Pinyon-Juniper
  • Geology
  • Mixed Montane Shrubland
  • Douglas-fir

The Wet Shrubland, C3/C4 Grassland, and Annual Exotic Grassland alliance groups were most sensitive to water deficit and drought.

Vegetation types responded differently to precipitation, according to their sensitivity. Vegetation types in areas not shaded purple tended to be more sensitive to precipitation changes.

Drought tolerance

Understanding drought tolerance helps park managers determine which vegetation types may persist or perish if water stress increases due to climate change. Vegetation that requires less water is more drought-tolerant.

• If the landscape gets drier, then the amount of area suitable for drought-tolerant species will likely increase. Under this scenario, there will likely be less area available for less drought-tolerant species.
• Vegetation types at Canyonlands National Park, ordered from MOST to LEAST drought-tolerant:
  • Water
  • Alkaline Wet
  • Ponderosa Pine
  • Dry Wash
  • Sparsely Vegetated
  • Wet Shrubland
  • Dry Shrubland
  • Riparian
  • Blackbrush
  • Geology
  • Dry Sagebrush
  • Annual Exotic Grassland
  • Pinyon-Juniper
  • C3/C4 Grassland
  • Juniper
  • Disturbed areas
  • Mixed Montane Shrubland
  • Douglas-fir

Vegetation groups in purple areas are less tolerant of drought than other areas.

Two recently dead pinyon pines near a long-term vegetation monitoring plot at Canyonlands National Park. NPS/Amy Washuta

What Lies Ahead for Canyonlands National Park?

Potential vegetation transitions

Though our science-based findings are more reliable than the musings of a fortuneteller, they have limits. We can’t know exactly how vegetation change will play out at Canyonlands National Park. But change is coming. The cause of juniper mortality seen by park managers in southern Utah in 2018 was acute drought stress (Kannenberg 2021)—a surprising finding, given the relative insensitivity of juniper to interannual precipitation changes. And even under a range of scenarios, projections trend toward more aridity in the future (see figure).

Under best-case and worst-case scenarios, Canyonlands National Park is projected to become more arid in the future.

In the figure, “average annual soil moisture” represents the amount of moisture stored in the top meter of soil at a single Pinyon-Juniper polygon at the park's center. The “pivot point” represents how much drought stress a vegetation type can tolerate and still persist over time. Soil moisture at levels below the pivot point restricts plant growth and survival. Combined with climate projections of increased aridity, the vegetation sensitivity to soil moisture reported in this study can help guide the time and place for management actions at Canyonlands National Park.

Under the warm-and-wet (best-case) scenario, annual soil moisture will only achieve the historical soil moisture pivot point occasionally—meaning that even in the best case, soil moisture will not be sufficient to meet vegetation needs in this polygon. This will likely result in transition to a different vegetation assemblage due to drought stress.

Under a worst-case (warm-and-dry) scenario, soil moisture will be lower than the best-case scenario in most years. This will result in more intense and prolonged drought stress that will likely make transitions happen sooner. It may also result in more dry-adapted vegetation composition than in the best-case scenario. Either way, the time to start planning for change is now.

Informed planning

The success of restoration projects improves when growing conditions are more likely to promote establishment of desirable plant species.

Managers can use information from studies like this one in a powerful planning framework. Climate Smart Conservation (CSC) is a roadmap for managing natural resources as the climate changes. Ultimately, the CSC process helps managers decide whether to resist, accept, or direct vegetation change and identify management options for each approach. One of its key inputs is a resource’s sensitivity and vulnerability to climate change, which this study provides. The report associated with this web article offers a step-by-step application of this kind of information for Climate Smart Conservation planning.

Information in the report can also be used to identify optimum restoration windows. Restoration projects are an indispensable resource-management tool. They are also expensive and time-consuming. If park managers can anticipate how different vegetation types are likely to respond to projected climate conditions, then they can adjust the species mix and timing of restoration projects to optimize seed germination and establishment of desirable species most likely to survive new climate conditions.

Other potential questions—and answers

The information provided in this research can also help managers anticipate and plan for cascading ecological effects, such as effects on pollinators, or the movement of wildlife (migrations) across park lands. It can also be leveraged to answer many more questions, such as:

• Where are the most- and least-sensitive vegetation types in the park, and when might they experience chronic conditions that force change?
• Are forests likely to experience drought stress or transition sooner than shrublands?
• Which soil types most effectively buffer drought stress, and where do they occur in the park?
• After disturbances, such as wildfire or extreme drought, are climate conditions likely to support historical vegetation types?
• Are animal species of concern dependent on climate-sensitive vegetation types?

Vegetation provides valuable ecological services and adds to park viewsheds in immeasurable ways. But plants are sensitive to weather and climate—and when those conditions change, they respond. This report provides a foundation for understanding where, when, what, and why change may occur. Park managers can use that information to manage park resources through continuous change.

---

Information in this article was summarized from Landscape phenology, vegetation condition, and relations with climate at Canyonlands National Park, 2000–2019, by D. Thoma (2023).

Other cited material includes:

Gangopadhyay, S., C.A. Woodhouse, G.J. McCabe, C.C. Routson, and D.M. Meko. 2022. Tree rings reveal unmatched 2nd century drought in the Colorado River Basin. Geophysical Research Letters, June 9. https://doi.org/10.1029/2022GL098781.

Kannenberg, S.A., A.W. Driscoll, D. Malesky, and W.R.L. Anderegg. 2021. Rapid and surprising dieback of Utah juniper in the southwestern USA due to acute drought stress. Forest Ecology and Management 480 (January 15):118639. https://doi.org/10.1016/j.foreco.2020.118639.