Data from: Whitebark pine growth dynamics reflect diverse responses to climate and patterns related to bark beetle-caused mortality
Data files
Jun 29, 2026 version files 670.18 KB
-
CDT_beetle_killed_raw_tree_ring_values_mm.csv
116.42 KB
-
CDT_living_raw_tree_ring_values_mm.csv
95.40 KB
-
CDT_tree_DBH_WPBR.csv
865 B
-
HL_beetle_killed_raw_tree_ring_values_mm.csv
90.82 KB
-
HL_living_raw_tree_ring_values_mm.csv
68.14 KB
-
HL_tree_DBH_WPBR.csv
847 B
-
MM_living_raw_tree_ring_values_mm.csv
58.60 KB
-
MM_tree_DBH_WPBR.csv
431 B
-
PF_beetle_killed_raw_tree_ring_values_mm.csv
122.58 KB
-
PF_living_raw_tree_ring_values_mm.csv
81.08 KB
-
PF_tree_DBH_WPBR.csv
1.60 KB
-
README.md
8.13 KB
-
SW_living_raw_tree_ring_values_mm.csv
25.09 KB
-
SW_tree_DBH_WPBR.csv
155 B
Abstract
In order to conserve high-elevation forest ecosystems, we must understand how cold-adapted tree species respond to climatic shifts and biotic disturbances. We examined growth rates and climatic sensitivities of whitebark pine (Pinus albicaulis) across arid and mesic sites and evaluated their association with mortality caused by mountain pine beetle (Dendroctonus ponderosae). We assessed whether populations were energy- or water-limited, whether recent warming has shifted these limitations, and if shifts corresponded with beetle outbreaks. We also investigated if growth rate and climate sensitivity influenced tree survival. Historically, whitebark pine populations were primarily energy-limited, with growth increasing when warming began around 1940. A transition toward moisture limitation emerged only at the warmest mesic site as warming intensified and droughts became hotter in the 2000s. Contrary to expectations, beetle outbreaks occurred only at cooler arid sites, where trees showed positive growth responses to warmer temperatures, even during drought, while no outbreaks developed at mesic sites where growth declined during hot drought. Trees that survived outbreaks differed in growth from those that were killed, but whether surviving trees grew faster or slower varied among populations. Trees killed by beetles were no more or less sensitive to climate than survivors prior to the outbreak. These results suggest that for now, negative effects on growth due to a warming climate may be limited to warmer sites but that as temperature and growth rates increase, thus increasing water demand, negative effects will become more widespread. Our findings challenge assumptions derived from lower-elevation moisture-limited forests and suggest that there are fundamentally different mechanisms governing beetle outbreaks in energy-limited whitebark pine ecosystems.
Dataset DOI: 10.5061/dryad.63xsj3vgc
Description of the data and file structure
Study sites are abbreviated:
- CDT= Continental Divide Trail
- HL= Herd Lake
- MM= Marshall Mountain
- PF= Poverty Flat
- SW= Schweitzer
Files and variables
File: CDT_beetle_killed_raw_tree_ring_values_mm.csv
Description: This file contains the raw annual growth ring width in millimeters for beetle-killed trees at the CDT study site. The last year of growth has been adjusted for the year the tree died, not the year of sampling. Individual trees are labeled by their number (“01”) and the core (“A,” “B,” etc.) Most trees have two cores (i.e., “01A” and “01B”).
Variables
- annual tree growth in mm
File: CDT_living_raw_tree_ring_values_mm.csv
Description: This file contains the raw annual growth ring width in millimeters for living trees at the CDT study site. The last year of growth is the year of sampling. Individual trees are labeled by their number (“01”) and the core (“A,” “B,” etc.) Most trees have two cores (i.e., “01A” and “01B”).
Variables
- annual tree growth in mm
File: HL_tree_DBH_WPBR.csv
Description: This file contains the diameter at breast height (DBH) in centimeters for each individual living and beetle-killed tree at the Herd Lake study site. The file also contains the white pine blister rust (WPBR) rating for each living tree, assessed using the protocol described in Six and Newcomb 2005 (“A rapid system for rating white pine blister rust incidence, severity, and within-tree distribution in whitebark pine”). The Six and Newcomb (2005) protocol partitions the tree into sections and assigns each section a rating based on visible WPBR symptoms. The ratings are then summed for the whole tree and can range from 0 (no visible infection) to 18 (complete infection).
Variables
- Tree identification number
- DBH in cm
- WPBR rating (0-18)
File: CDT_tree_DBH_WPBR.csv
Description: This file contains the diameter at breast height (DBH) in centimeters for each individual living and beetle-killed tree at the CDT study site. The file also contains the white pine blister rust (WPBR) rating for each living tree, assessed using the protocol described in Six and Newcomb 2005 (“A rapid system for rating white pine blister rust incidence, severity, and within-tree distribution in whitebark pine”). The Six and Newcomb (2005) protocol partitions the tree into sections and assigns each section a rating based on visible WPBR symptoms. The ratings are then summed for the whole tree and can range from 0 (no visible infection) to 18 (complete infection).
Variables
- Tree identification number
- DBH in cm
- WPBR rating (0-18)
File: SW_living_raw_tree_ring_values_mm.csv
Description: This file contains the raw annual growth ring width in millimeters for living trees at the Schweitzer study site. The last year of growth is the year of sampling. Individual trees are labeled by their number (“01”) and the core (“A,” “B,” etc.) Most trees have two cores (i.e., “01A” and “01B”).
Variables
- annual tree growth in mm
File: MM_tree_DBH_WPBR.csv
Description: This file contains the diameter at breast height (DBH) in centimeters for each individual living tree at the Marshall Mountain study site. The file also contains the white pine blister rust (WPBR) rating for each living tree, assessed using the protocol described in Six and Newcomb 2005 (“A rapid system for rating white pine blister rust incidence, severity, and within-tree distribution in whitebark pine”). The Six and Newcomb (2005) protocol partitions the tree into sections and assigns each section a rating based on visible WPBR symptoms. The ratings are then summed for the whole tree and can range from 0 (no visible infection) to 18 (complete infection).
Variables
- Tree identification number
- DBH in cm
- WPBR rating (0-18)
File: HL_beetle_killed_raw_tree_ring_values_mm.csv
Description: This file contains the raw annual growth ring width in millimeters for beetle-killed trees at the Herd Lake study site. The last year of growth has been adjusted for the year the tree died, not the year of sampling. Individual trees are labeled by their number (“01”) and the core (“A,” “B,” etc.) Most trees have two cores (i.e., “01A” and “01B”).
Variables
- annual tree growth in mm
File: HL_living_raw_tree_ring_values_mm.csv
Description: This file contains the raw annual growth ring width in millimeters for living trees at the Herd Lake study site. The last year of growth is the year of sampling. Individual trees are labeled by their number (“01”) and the core (“A,” “B,” etc.) Most trees have two cores (i.e., “01A” and “01B”).
Variables
- annual tree growth in mm
File: SW_tree_DBH_WPBR.csv
Description: This file contains the diameter at breast height (DBH) in centimeters for each individual living tree at the Schweitzer study site. The file also contains the white pine blister rust (WPBR) rating for each living tree, assessed using the protocol described in Six and Newcomb 2005 (“A rapid system for rating white pine blister rust incidence, severity, and within-tree distribution in whitebark pine”). The Six and Newcomb (2005) protocol partitions the tree into sections and assigns each section a rating based on visible WPBR symptoms. The ratings are then summed for the whole tree and can range from 0 (no visible infection) to 18 (complete infection).
Variables
- Tree identification number
- DBH in cm
- WPBR rating (0-18)
File: MM_living_raw_tree_ring_values_mm.csv
Description: This file contains the raw annual growth ring width in millimeters for living trees at the Marshall Mountain study site. The last year of growth is the year of sampling. Individual trees are labeled by their number (“01”) and the core (“A,” “B,” etc.) Most trees have two cores (i.e., “01A” and “01B”).
Variables
- annual tree growth in mm
File: PF_living_raw_tree_ring_values_mm.csv
Description: This file contains the raw annual growth ring width in millimeters for living trees at the Poverty Flat study site. The last year of growth is the year of sampling. Individual trees are labeled by their number (“01”) and the core (“A,” “B,” etc.) Most trees have two cores (i.e., “01A” and “01B”).
Variables
- annual tree growth in mm
File: PF_tree_DBH_WPBR.csv
Description: This file contains the diameter at breast height (DBH) in centimeters for each individual living and beetle-killed tree at the Poverty Flat study site. The file also contains the white pine blister rust (WPBR) rating for each living tree, assessed using the protocol described in Six and Newcomb 2005 (“A rapid system for rating white pine blister rust incidence, severity, and within-tree distribution in whitebark pine”). The Six and Newcomb (2005) protocol partitions the tree into sections and assigns each section a rating based on visible WPBR symptoms. The ratings are then summed for the whole tree and can range from 0 (no visible infection) to 18 (complete infection).
Variables
- Tree identification number
- DBH in cm
- WPBR rating (0-18)
File: PF_beetle_killed_raw_tree_ring_values_mm.csv
Description: This file contains the raw annual growth ring width in millimeters for beetle-killed trees at the Poverty Flat study site. The last year of growth has been adjusted for the year the tree died, not the year of sampling. Individual trees are labeled by their number (“01”) and the core (“A,” “B,” etc.) Most trees have two cores (i.e., “01A” and “01B”).
Variables
- annual tree growth in mm
Code/software
Excel or other spreadsheet software
We used data previously collected from within variable-length belt transects established at Poverty Flat in late July 2017-2018 (Six et al., 2021), and established new transects using the same approach at the CDT site in late July 2022, the Herd Lake site in early September 2022, and at the Marshall Mountain and Schweitzer sites in mid-August 2023. The transects were approximately 10 meters (m) wide and extended through forested areas. Large living whitebark pines remaining after the outbreak (arid sites) or after many decades of blister rust infection (mesic sites) were dispersed across the landscape among large numbers of dead whitebark pine and meadows. Because of the dispersed and patchy distribution of living trees and forested areas, we started a new randomized transect each time a non-whitebark pine forest type, meadow, or steep slope was encountered. Transects extended until a target number of trees was met (35-40) in each category (live, beetle-killed) except at Schweitzer, where very few living mature whitebark pine remained, in which case we sampled a smaller number of trees.
At sites that experienced recent mountain pine beetle outbreaks (arid: CDT, Herd Lake, and Poverty Flat), we first determined the diameter distribution of trees killed by the beetle by measuring the diameter at breast height (DBH, 1.4 m above ground level) of approximately 100 mountain pine beetle-killed whitebark pines along the transects. We then measured the DBH of 35-40 living whitebark pines (hereafter ‘living’) and 35-40 beetle-killed trees according to this diameter distribution along the same transects. When whitebark pine occurred in clumps (common at some sites due to seed caching by the bird Clark’s nutcracker, Nucifraga columbiana), only the stem with the largest DBH was measured. Each tree was cored to the pith at 1.4 m above the ground using a 5.2 mm diameter increment borer. Two cores were taken from each tree on opposing sides, with care taken to avoid scars and other abnormal features. For living trees, we also rated each tree for WPBR infection severity and assigned a rating from 0 (no infection) to 18 (full infection) using the rating system of Six and Newcomb (2005).
The minimum DBH of whitebark pine attacked by mountain pine beetle at the sites that had experienced an outbreak was 23 cm. Therefore, at sites where mountain pine beetle mortality was negligible (mesic: Marshall Mountain and Schweitzer), we sampled living trees with a DBH of 23 cm and larger along the transects to allow comparisons with trees at sites that experienced outbreaks. These trees were also rated for WPBR infection severity.
Tree cores were air-dried and then mounted and sanded following standard procedures (Speer, 2010). Each ring was measured to the nearest 0.001 mm on a Velmex measuring stage using J2X software (Voortech Consulting, NH, USA).