Data from: Thinning and prescribed burning increase shade-tolerant conifer regeneration in a fire excluded mixed-conifer forest
Data files
Nov 21, 2023 version files 71.64 KB
Abstract
Fire exclusion and past management have altered the composition, structure, and function of frequent-fire forests throughout western North America. In mixed-conifer forests of the California Sierra Nevada, fire exclusion has exacerbated the effects of drought and endemic bark beetles, resulting in extensive mortality of fire-adapted pine species. Thinning and prescribed fire are widely used in these forests to reduce fuels, moderate fire behavior, and restore ecosystems. Tree regeneration influences future forest composition and structure, and therefore future resilience to disturbances, but long-term effects of thinning and prescribed burning on tree regeneration after prolonged fire exclusion are poorly understood. We measured tree regeneration one year prior to, and periodically for 16 years following thinning and prescribed burning in a mixed-conifer forest in the Sierra Nevada, California, USA. We asked three questions. How did the composition and density of tree regeneration change after thinning and prescribed burning? Did pretreatment vegetation types influence conifer regeneration density after treatments? Did planting after overstory thinning increase regeneration density of native pine species?
Sixteen years after treatments, combined natural regeneration of shade-tolerant white fir (Abies concolor) and incense-cedar (Calocedrus decurrens) averaged 2,032 trees per hectare (tph) after understory thinning, and 7,745 tph after understory thinning combined with prescribed burning, increases of 37% and 146% from pretreatment densities. In contrast, combined natural regeneration of white fir and incense-cedar averaged 497 tph after overstory thinning, 780 tph after overstory thinning with prescribed burning, 113 tph after prescribed burning alone, and 807 tph in untreated controls, all of which were declines from pretreatment densities. Natural regeneration of white fir and incense-cedar was consistently an order of magnitude greater than Jeffrey pine (Pinus jeffreyi) and sugar pine (Pinus lambertiana), whose combined densities 16 years after treatments averaged 37 tph across treatments and did not significantly respond to thinning and/or prescribed burning. Natural conifer regeneration after treatments varied by pre-treatment vegetation type (closed canopy, Ceanothus cordulatus shrub-dominated, and open sparse), with large increases of natural regeneration after understory thinning in closed canopy and Ceanothus shrub vegetation types. Planting increased sugar pine regeneration density after overstory thinning, marginally increased Jeffrey pine regeneration after overstory thinning combined with prescribed burning, and increased white fir regeneration after overstory thinning with and without burning. No treatments reduced white fir and incense-cedar natural regeneration while simultaneously increasing natural pine regeneration, suggesting new thinning, burning, and planting approaches may be required to meet regeneration restoration objectives.
README
This README file was generated on 2023-11-20 by Harold Zald.
GENERAL INFORMATION
- Title of Dataset: Thinning and prescribed burning increase shade-tolerant conifer regeneration in a fire excluded mixed-conifer forest
- Author Information A. Principal Investigator Contact Information Name: Harold Zald Institution: USDA Forest Service, Pacific Northwest Service Station Address: Corvallis, OR USA Email: harold.zald@usda.gov B. Associate or Co-investigator Contact Information Name: Andrew Gray Institution: USDA Forest Service, Pacific Northwest Research Station Address: Corvallis, OR USA Email: andrew.gray@usda.gov
- Date of data collection (multiple dates, range): 2000-2021
- Geographic location of data collection: Teakettle Experimental Forest, California, USA 36°58′N, 119°02′W
- Information about funding sources that supported the collection of the data: USDI Joint Fire Science Program (Project ID 15-1-07-6, 10-1-10-14, 01-3-2-02), California Department of Forestry and Fire Protection as part of the California Climate Investments Program (Grant #8GG14803, #8GG14804), and the USDA Forest Service Pacific Northwest and Pacific Southwest Research Stations
SHARING/ACCESS INFORMATION
- Licenses/restrictions placed on the data: CC0 1.0 Universal (CC0 1.0) Public Domain
- Links to publications that cite or use the data:
Zald, H. S. J., May, C. J., Gray, A. N., North, M. P., Hurteau, M. D. (2024). Thinning and prescribed burning increase shade-tolerant conifer regeneration in a fire excluded mixed-conifer forest. Forest Ecology and Management 551(1): 121531. https://doi.org/10.1016/j.foreco.2023.121531
- Links to other publicly accessible locations of the data: None
- Links/relationships to ancillary data sets: None
- Was data derived from another source? No A. If yes, list source(s): NA
- Recommended citation for this dataset:
Zald, H. S. J., May, C. J., Gray, A. N., North, M. P., Hurteau, M. D. (2023). Data from: Thinning and prescribed burning increase shade-tolerant conifer regeneration in a fire excluded mixed-conifer forest.
Dryad Digital Repository. https://doi.org/10.5061/dryad.2bvq83bx1
DATA & FILE OVERVIEW
- Data frame List: All data files are data frame objects within the R data file TEF_regeneration.RData
A) gridpt.count
B) gridpt.envars
- Relationship between files, if important: Both data files have a common variable gridpt
- Additional related data collected that was not included in the current data package: None
- Are there multiple versions of the dataset? No A. If yes, name of file(s) that was updated: NA i. Why was the file updated? NA ii. When was the file updated? NA
#########################################################################
DATA-SPECIFIC INFORMATION FOR: gridpt.count
- Number of variables: 15
- Number of cases/rows: 14670
- Variable List:
- trt: Treatment combination of prescribed burning and mechanical thinning. Categorical variable with six levels (UN = unburned, not thinned control, UC = unburned, understory CASPO thinning, US = unburned, overstory shelterwood thinning, BN = burned, not thinned control, BC = burned, understory CASPO thinning, BS = burned, overstory shelterwood thinning). CASPO refers understory thinning following guidelines in the California spotted owl report (Verner et al. 1992. https://doi.org/10.2737/PSW-GTR-133), removing intermediate sized trees 25–76 cm diameter at breast height (DBH, 1.37 m) while retaining at least 40 % canopy cover, retaining 40 % basal area, or no harvest of trees greater than 76 cm DBH, whichever constraint was most restrictive.
- plot: 4 hectare plot identification code. Categorical variable with 16 levels. There are 3 replicate plots for each level of treatment combination in the trt variable.
- gridpt: Gridpoint identification code. Categorical variable with 402 levels, with first three characters denoting which plot a gridpoint is in, followed by a numeric identifier ranging from 01 to 49. For each treatment combination, two replicate plots each have nine gridpoints (01-09) positioned in a regular grid with 50 m x 50 m spacing. For the third replicate plot in each treatment combination there are 49 gridpoints on regular grid with 25 m x 25 m spacing.
- quad: Quadrat of 3.5 m radius tree regeneration microplots centered on gridpoints. Categorical variable with 4 levels (NE = Northeast, SE = Southeast, SW = Southwest, NW = Northwest)
- species: Tree species code. Categorical variable with eight levels (ABCO = White fir, Abies concolor, ABMA = Red fir, Abies magnifica, CADE = Incense-cedar, Calocedrus decurrens, PIJE = Jeffrey pine, Pinus Jeffreyii, PILA = Sugar pine, Pinus lambertiana, PREM = Bitter cherry, Prunus emarginata, QUCH = Canyon live oak, Quercus chrysolepis, QUKE = California black oak, Quercus kelloggii).
- origin: Origin of tree regeneration. Categorical variable with two levels (N = natural regeneration, P = planted regeneration). Planted regeneration was 2 year old bare root stock, planted in 2002, in only the overtstory thinned treatments (trt = US or trt = BS).
- year: Year of regeneration measurement. Numeric variable with values 2000, 2002, 2003, 2004, 2005, 2011, 2016, 2017, 2018, 2019, 2021.
- szcl: Tree regeneration size class. Categorical variable with four levels (A = height greater or equal than 5 cm and less than 50 cm\, B = height greater or equal to 50 cm and less than 137 cm\, C = height greater or equal to 137 cm and diameter at 137 cm height less than 5 cm\, D = height greater or equal to 137 and diameter at 137 cm height greater or equal then 5 cm). Note that during first measurement year (year = 2000) tree regeneration was only tallied for regeneration less than 5 cm in diameter (szcl = A | B |C). Inclusion of szcl = D was in subsequent measurement years to account for regeneration that grew greater than 5 cm diameter.
- count: Count of regeneration for a given gridpt, quad, species, origin, and szcl. Numeric variable from 0 to 327.
- azi: Azimuth from gridpoint center to planted tree regeneration. Numeric variable from 1-359. Azimuth assumes magnetic declination of 14 degree east.
- dist: Distance in meters from gridpoint center to planted tree regeneration. Numeric variable from 0.1 to 3.5 m.
- dia: Tree regeneration diameter in cm. Numeric variable from 0.1 to 15.2 cm.
- len: Length/height of tree regeneration in cm. Numeric variable from 6 to 480 cm.• comments: qualitative field observation comments. Text variable
- Data.Corrections: Data corrections that derivate from original field data sheets. Text variable
- Missing data codes: NA
- Specialized formats or other abbreviations used: None
#########################################################################
DATA-SPECIFIC INFORMATION FOR: gridpt.envars
- Number of variables: 17
- Number of cases/rows: 402
- Variable List:
- gridpt: Gridpoint identification code. Categorical variable with 402 levels, with first three characters denoting which plot a gridpoint is in, followed by a numeric identifier ranging from 01 to 49. For each treatment combination, two replicate plots each have nine gridpoints (01-09) positioned in a regular grid with 50 m x 50 m spacing. For the third replicate plot in each treatment combination there are 49 gridpoints on regular grid with 25 m x 25 m spacing.
- canopy: Pretreatment percent canopy cover above gridpoints. Numeric variable from 0 to 100 percent, estimated using digital hemispherical photography taken during the pretreatment summer of 2000. For details see Gray et al. (2005) https://doi.org/10.1093/forestscience/51.3.198.
- litter: Percent cover of litter within 1.78 m radius microplot centered on each grid point, measured during the pretreatment summers of 1998–1999. Numeric variable from 0 to 20 percent.
- bare: Percent cover of bare ground within 1.78 m radius microplot centered on each grid point, measured during the pretreatment summers of 1998–1999. Numeric variable from 0 to 95 percent.
- rock: Percent cover of rock within 1.78 m radius microplot centered on each grid point, measured during the pretreatment summers of 1998–1999. Numeric variable from 0 to 100 percent.
- cwd: Percent cover of coarse woody debris (woody pieces greater or equal than 20 cm diameter) within 1.78 m radius microplot centered on each grid point, measured during the pretreatment summers of 1998–1999. Numeric variable from 0 to 84 percent.
- sticks: Percent cover of sticks (woody pieces less than 20 cm diameter, excluding litter) within 1.78 m radius microplot centered on each grid point, measured during the pretreatment summers of 1998–1999. Numeric variable from 0 to 85 percent.
- arpa: Percent cover of shrub Arctostaphylos patula within 1.78 m radius microplot centered on each grid point, measured during the pretreatment summers of 1998–1999. Numeric variable from 0 to 95 percent.
- ceco: Percent cover of shrub Ceanothus cordulatus within 1.78 m radius microplot centered on each grid point, measured during the pretreatment summers of 1998–1999. Numeric variable from 0 to 100 percent.
- syro: Percent cover of shrub Symphoricarpos mollis within 1.78 m radius microplot centered on each grid point, measured during the pretreatment summers of 1998–1999. Numeric variable from 0 to 55 percent.
- prem: Percent cover of shrub/small tree Prunus emarginata within 1.78 m radius microplot centered on each grid point, measured during the pretreatment summers of 1998–1999. Numeric variable from 0 to 55 percent.
- riro: Percent cover of shrub Ribes roezlii within 1.78 m radius microplot centered on each grid point, measured during the pretreatment summers of 1998–1999. Numeric variable from 0 to 40 percent.
- patch: Vegetation patch type classification. Categorical variable with three levels (CECO dominated, closed canopy, open).
- h2o.oct.1998: Volumetric percent soil water content in the upper 15 cm of soil measured in mid-Oct. 1998. Numeric variable from 1.72 to 48.3 percent. Soil moisture measured using time domain reflectometry. For details, see Zald et al. 2008. https://doi.org/10.1016/j.foreco.2008.04.022.
- h2o.may.1999: Volumetric percent soil water content in the upper 15 cm of soil measured in mid-May. 1999. Numeric variable from 3.8 to 50 percent.
- soil.depth: Soil depth to bedrock in cm. Numeric variable from 0 to 257 cm.
- Missing data codes: NA
- Specialized formats or other abbreviations used: None
Methods
See methods section of Zald et al. 2024. Thinning and prescribed burning increase shade-tolerant conifer regeneration in a fire excluded mixed-conifer forest. Forest Ecology and Management, 551(1) 121531. https://doi.org/10.1016/j.foreco.2023.121531