Data from: Hurricane-induced habitat change alters growth and microclimate for fungal diseases in the endangered Florida torreya
Data files
Apr 22, 2026 version files 76.74 KB
-
README.md
10.36 KB
-
Torreya_branchsegments_2021.csv
13.43 KB
-
Torreya_cankercounts_2019.csv
341 B
-
Torreya_canopy_soilcompact_2022.csv
2.04 KB
-
Torreya_forestDBH_2019.csv
6.65 KB
-
Torreya_foresttreecounts_2019.csv
449 B
-
Torreya_mortality_branchmort_cankers_2022.csv
39.92 KB
-
Torreya_weatherloggers_2021.csv
3.56 KB
Abstract
Florida torreya (Torreya taxifolia), a critically endangered conifer endemic to steephead ravines of northwest Florida and southwest Georgia, has declined for decades due to a fungal canker disease. Hurricane Michael, a Category 5 storm in 2018, caused extensive canopy loss within this refugial habitat. We investigated how such disturbance influenced tree health and microclimatic conditions relevant to disease dynamics. We monitored 40 wild T. taxifolia individuals for four years following the hurricane, measuring branch growth and mortality, temperature and light conditions, and the abundance and area of fungal cankers. Hurricane-induced treefall was quantified as the diameter at breast height (DBH) of fallen and standing trees within a 10 m radius of each focal individual. Hurricane-induced treefall was associated with increased branch mortality but also increased branch growth, likely due to greater light availability. Canopy loss significantly elevated local temperatures, which were positively associated with canker abundance, though not with total canker area. These results demonstrate that severe disturbance can simultaneously reduce competition and intensify pathogen pressure in rare species. In T. taxifolia, hurricane disturbance appears to alleviate light limitation while creating warmer microclimates that favor fungal disease. This work highlights the complex interplay between natural disturbance and invasive pathogens, with implications for the persistence of endangered plants under global change.
Dataset DOI: 10.5061/dryad.zs7h44jrb
Description of the data and file structure
Dataset Overview
This dataset documents the health, growth, microclimate, and disease dynamics of the critically endangered Florida torreya (Torreya taxifolia) following Hurricane Michael (2018), a Category 5 storm that caused extensive canopy loss in Torreya State Park, Florida. The dataset was generated from a multi-year field study designed to evaluate how hurricane-induced forest disturbance influences individual tree performance and fungal canker disease severity.
Data include repeated measurements of individual tree health, branch growth and mortality, forest structure, microclimate conditions, and canker disease severity. These data enable analyses of how variation in canopy disturbance and local environmental conditions affect both host performance and pathogen dynamics in a rare species.
Dates of Data Collection
- Initial tree selection: October 2018
- Baseline measurements: March 2019
- Tree health monitoring: December 2019, November 2021, December 2022, October 2023
- Growth measurements (branch segments): Fall 2021 (retrospective assignment to 2018–2021 growing seasons)
- Microclimate measurements (temperature, dew point): Continuous logging beginning in 2021 (6-hour intervals; summarized as May–September means)
- Canker disease measurements: 2019 and 2022
Data Spatial Scope
All data were collected within Torreya State Park, Florida, USA, in steephead ravine habitats that support remnant populations of Torreya taxifolia.
Forty focal individuals were distributed across four sampling areas to capture spatial variation:
- Aspalaga (north; n = 7)
- Sweetwater (south; n = 8)
- Central-East (n = 8)
- Central-West (n = 17)
The northern and southern sites are separated by approximately 8.3 km. The southern site lies ~2.6 km from the central sites, and the northern site lies ~3.7 km from them. Central-West sites occur in ravines draining toward the Apalachicola River, while Central-East sites occur in ravines farther from the river, draining toward Rock Creek.
Methods Summary
Focal T. taxifolia individuals were selected in October 2018 to minimize variation in initial size and form (initial height approximately 25–120 cm). Each focal individual represents a single rootstock. Baseline height and condition were recorded in March 2019, when all individuals were alive.
Hurricane-induced forest disturbance was quantified within 10 m radius plots centered on each focal individual. Within each plot, all standing and fallen trees ≥5 cm DBH were measured. Derived metrics include standing basal area, fallen basal area, total basal area, and summed DBH values representing size-weighted tree abundance.
Microclimate and habitat variables measured at each focal tree include air temperature, dew point, overstory cover, and soil compaction. Temperature and humidity were recorded using HOBO data loggers at 6-hour intervals beginning in 2021 and summarized as mean summer values (May–September).
Tree health was monitored across multiple years, with mortality defined as death of all shoots from a rootstock. Growth was estimated using annual branch segment lengths measured on lateral branches, allowing reconstruction of growth across pre- and post-hurricane years (2018–2021). Branch mortality was quantified by counting live and dead branches or scars.
Fungal canker severity was assessed by measuring canker abundance and size (length and width of the three largest cankers per individual). Canker area was estimated assuming an elliptical shape.
Data Structure and Variables
The dataset contains multiple tables representing different components of the study, which may include:
- Torreya individual data: tree ID, location, baseline size, survival status
- Growth data: torreya branch segment lengths by year (assuming proximal branch segment represents the most recent year's growth) per branch limb per tree ID
- Disease data: canker abundance and size measurements for the three largest
- Forest structure data: plot-level counts and DBH metrics for standing and fallen trees
- Microclimate data: temperature, dew point, overstory cover, soil compaction
Because data were collected across multiple years and measurement types, sample sizes may vary among variables due to mortality of focal individuals or equipment limitations. Baseline measurements are available for all initially selected individuals.
Funding
Funding was provided by the U.S. Fish & Wildlife Service (F20AC00357), an anonymous funder, and the Atlanta Botanical Garden.
Additional Notes
- All measurements were collected in situ under natural post-disturbance conditions following Hurricane Michael.
- Focal individuals were not randomly selected but were chosen to standardize initial size and facilitate longitudinal comparison.
- Coordinates are not provided to protect this critically endangered species.
- NA is used to denote missing data or Not Applicable.
Files and variables
File: Torreya_branchsegments_2021.csv
Description: Branch segments lengths measured on each focal torreya individual to estimate annual growth.
Variables
- treeID: identity of focal torreya individual
- limb.direction: cardinal direction of branch limb measured in 2021
- branch.year: branch segment identity by year, assuming most distal segment was produced in 2021. This makes 2019 the segment produced the first year after the hurricane (October 2018), 2020 the second year, and 2021 the third year.
- branch.length: branch segment length in cm
File: Torreya_canopy_soilcompact_2022.csv
Description: Percent canopy cover and soil compaction measured at each focal torreya individual.
Variables
- treeID: identity of focal torreya individual. One tree is listed as "unmarked"
- location: location within Torreya State Park
- canopy.north: precent canopy cover measured in four cardinal directions around each tree ID in 2022
- canopy.south: precent canopy cover measured in four cardinal directions around each tree ID in 2022
- canopy.east: precent canopy cover measured in four cardinal directions around each tree ID in 2022
- canopy.west: precent canopy cover measured in four cardinal directions around each tree ID in 2022
- soil.north: compressive soil strength (tons/sq. ft.) measured using a pocket penetrometer in four cardinal directions around each tree ID in 2022
- soil.south: compressive soil strength (tons/sq. ft.) measured using a pocket penetrometer in four cardinal directions around each tree ID in 2022
- soil.east: compressive soil strength (tons/sq. ft.) measured using a pocket penetrometer in four cardinal directions around each tree ID in 2022
- soil.west: compressive soil strength (tons/sq. ft.) measured using a pocket penetrometer in four cardinal directions around each tree ID in 2022
File: Torreya_forestDBH_2019.csv
Description: The diameter at breast height of fallen and standing forest trees around focal torreya individuals.
Variables
- treeID: identity of focal torreya individual
- fallentotaldbh: total diameter at breast height (DBH) of all fallen trees above 5 cm DBH within 10 m in 2019
- standingtotaldbh: total diameter at breast height (DBH) of all standing trees above 5 cm DBH within 10 m in 2019
File: Torreya_mortality_branchmort_cankers_2022.csv
Description: The mortality status of focal torreya individuals, branch mortality, and canker sizes measured in 2022.
Variables
- site: location within Torreya State Park
- treeID: identity of focal torreya individual
- num.liveshoots: number of living stems emerging from the ground in 2022
- marked.live: length of the longest stem in 2022
- num.branchalive: number of lateral branches alive in 2022
- num.branchdead: number of lateral branches dead (or branch scars) in 2022
- num.cankers: number of cankers on longest stem (not including lateral branches)
- length.cank1: length of one of the three (four if present) largest cankers in cm
- width.cank1: width of one of the three (four if present) largest cankers in cm
- length.cank2: length of one of the three (four if present) largest cankers in cm
- width.cank2: width of one of the three (four if present) largest cankers in cm
- length.cank3: length of one of the three (four if present) largest cankers in cm
- width.cank3: width of one of the three (four if present) largest cankers in cm
- length.cank4: length of one of the three (four if present) largest cankers in cm
- width.cank4: width of one of the three (four if present) largest cankers in cm
File: Torreya_cankercounts_2019.csv
Description: Number of cankers counted on torreya individuals in 2019
Variables
- treeID: identity of focal torreya individual
- cankers2019: number of cankers counted on longest stem in 2019
File: Torreya_foresttreecounts_2019.csv
Description: Number of fallen and standing trees within 10 m of focal torreya individual in 2019
Variables
- treeID: identity of focal torreya individual
- fallentrees: total number of fallen trees above 5 cm DBH within 10 m in 2019
- standingtrees: total number of standing trees above 5 cm DBH within 10 m in 2019
File: Torreya_weatherloggers_2021.csv
Description: Weather data per torreya tree measured with data loggers (HOBO MX2301 Temp/RH Data Logger). Loggers were set up in October 2021 and set to record every 6 hours continuously starting from 12:00 AM until October 2022. Due to inconsistent equipment performance outside of the summer season, only data between May and September were deemed of high quality for analysis.
Variables
- treeID: identity of focal torreya individual
- season: season for which data was summarized
- meantemp: mean temperature in degrees Celsius for the season
- meandp: mean dew point for the season
- maxtemp: maximum temperature for the season
- mintemp: maximum temperature for the season
Code/software
CSV files can be viewed using spreadsheet software like Microsoft Excel or Google Sheets.
Recording Florida torreya baseline health condition immediately post-hurricane
In October 2018, we selected 40 torreya (Torreya taxifolia) individuals in Torreya State Park for a multi-year monitoring study of the effects of hurricane-induced forest change. Focal trees were not selected randomly. Instead, individuals were chosen to minimize initial variation in size and form to facilitate fair comparisons through time. A focal individual represents a single torreya rootstock, which may produce multiple shoots at the soil surface. Selection criteria included a single visible shoot and an approximate initial height between 25–120 cm; only two selected individuals had more than one shoot emerging from the rootstock at the time of selection
In March 2019, five months after the hurricane and before the new growing season, all focal individuals were observed to be green and alive. Baseline height was then measured as the length of the longest shoot from the soil surface to the shoot tip, excluding leaves and following the contour of the shoot.
Measuring hurricane-induced forest change
We used natural variation in hurricane-related forest damage across locations within Torreya State Park to examine the effects of Hurricane Michael on torreya health and habitat. Hurricane damage was quantified using two complementary metrics derived from tree size and condition: total basal area of fallen and standing trees, which reflects forest structural conditions, and summed diameter at breast height (DBH), which reflects the size-weighted abundance of trees. Fallen trees included all living or dead trees uprooted or broken near the trunk base, while standing trees included living, damaged, or dead trees that remained upright following the hurricane. Basal area was calculated for each standing and fallen tree from DBH measurements and summed to derive plot-level estimates of standing basal area and fallen tree basal area; their sum (total basal area) was used as a proxy for pre-hurricane forest structural conditions. To characterize size-weighted tree abundance, hurricane damage was also summarized using summed DBH of standing trees, fallen trees, and their combined total.
We set up 10 m radius circular plots (314 m2) centered on each of the 40 focal individuals. Within each circular plot, we measured the diameter of all standing and fallen trees, with DBH greater than 5 cm, at 1.4 m from the ground, and on the upslope side of the tree or the root flare (for fallen trees).
To understand how post-hurricane environmental conditions might alter the severity of torreya canker disease, we measured post-hurricane temperature, dew point (air moisture), soil compaction, and overstory cover experienced by each of the focal individuals. Temperature and dew point was measured using data loggers (HOBO MX2301 Temp/RH Data Logger), set to record every 6 hours continuously starting from 12:00 AM in 2021. Daily mean summer temperatures were calculated with the readings from May to September. Soil compaction was measured in compressive soil strength (tons/sq. ft.) using a pocket penetrometer (AMS Pocket Penetrometer Model G 281, 1" Adapter Foot). Overstory cover was measured at the four cardinal directions approximately 1 m from each of the torreya focal individuals using a spherical densiometer held at waist height, to obtain a mean overstory cover value at each torreya individual.
Measuring post-hurricane torreya mortality, growth, and branch mortality
We visited focal torreya individuals in December 2019, November 2021, December 2022, and October 2023 to monitor their health and mortality. Mortality was defined as death of all shoots from a rootstock. In October 2023, all 40 focal tree locations were visited, to confirm that all previously recorded dead trees had not re-emerged.
In the fall of 2021, we estimated annual torreya growth rates by measuring the lengths of branch segments of each of the focal individuals. Although focal individual shoot height was recorded at the start of the study, previous studies have shown that only about 30% of trees increase in height per year, compared to 70 % lateral branches showing growth (Schwartz and Hermann, 1993). We used the lengths between branch scars on each focal individual, henceforth ‘branch segments’, as a non-invasive way to estimate branch annual primary growth for conifers in seasonal climates (Niinemets and Lukjanova, 2003; Lemay et al., 2019). Annual observations of torreya individuals grown at the Atlanta Botanical Garden corroborate that in mature individuals branch growth occurs in annual flushes. For each torreya individual, we recorded annual branch segment lengths on four of the longest lateral branches, selected from the four cardinal directions of the torreya. Only about 70 % of lateral branches produce growth each year (Schwartz and Hermann, 1993) and the growth of branch segments may not be independent of preceding growth. To factor this uncertainty in our models, data on branch segments was recorded separately for each torreya branch. For each branch, the lengths of up to four branch segments were measured starting from the youngest and most distal segment, and progressing proximally to older growths. Each branch segment was assigned an estimated year of growth, which included three post-hurricane growing seasons (2021, 2020, 2019) and one pre-hurricane growing season (2018). We quantified the branch mortality by counting the number of lateral live and dead branches or branch scars on the leading shoot of each focal individual.
Measuring post-hurricane canker disease severity
To measure how the hurricane influenced the severity of the torreya canker disease, we recorded the abundance and area of cankers on focal individuals in 2022. Consistent with data collected in 2019, we counted the abundance of cankers along a shoot of each focal individual. In individuals with multiple shoots, we tallied only the cankers on the longest shoot. Multiple cankers that had fused were counted as a single canker. In addition, we measured the length and width of the three largest cankers using digital calipers. Canker area was estimated by calculating the area of an ellipse: π * a * b, where a = length / 2 and b = width / 2.