The structural and functional impacts of invasive Psidium cattleianum in forests on the Island of Hawaiʻi
Data files
Dec 18, 2025 version files 1.30 GB
-
191105_coopercenter_native1_georef_.csv
65.20 KB
-
191105_coopercenter_native1_georef_.las
30.94 MB
-
191105_coopercenter_native2_georef_.csv
65.72 KB
-
191105_coopercenter_native2_georef_.las
26.60 MB
-
191108_coopercenter_mixed1_georef_c.csv
98.07 KB
-
191108_coopercenter_mixed1_georef_c.las
50.92 MB
-
191108_coopercenter_mixed2_georef_c.csv
32.28 KB
-
191108_coopercenter_mixed2_georef_c.las
17.15 MB
-
191111_coopercenter_invasive1_geore.csv
64.91 KB
-
191111_coopercenter_invasive1_geore.las
40.90 MB
-
191111_coopercenter_invasive2_geore.csv
63.64 KB
-
191111_coopercenter_invasive2_geore.las
39.12 MB
-
191120_olaa_native1_georef_classifi.csv
61.97 KB
-
191120_olaa_native1_georef_classifi.las
48.74 MB
-
191120_olaa_native2_georef_classifi.csv
60.54 KB
-
191120_olaa_native2_georef_classifi.las
41.82 MB
-
191202_olaa_mixed_20to40_georef_cla.csv
72.45 KB
-
191202_olaa_mixed_20to40_georef_cla.las
35.60 MB
-
191207_olaahwy11_invasive1_georef_c.csv
30.41 KB
-
191207_olaahwy11_invasive1_georef_c.las
35.47 MB
-
191207_olaahwy11_invasive2_georef_c.csv
52.89 KB
-
191207_olaahwy11_invasive2_georef_c.las
51.78 MB
-
191210_olaa_mixed_0to20_georef_clas.csv
69.36 KB
-
191210_olaa_mixed_0to20_georef_clas.las
39.33 MB
-
191211_laupahoehoe_invasive_0to20_r.csv
69.51 KB
-
191211_laupahoehoe_invasive_0to20_r.las
55.20 MB
-
191213_laupahoehoe_invasive_20to40_.csv
82.79 KB
-
191213_laupahoehoe_invasive_20to40_.las
46.30 MB
-
191215_laupahoehoe_mixed1_registere.csv
111.14 KB
-
191215_laupahoehoe_mixed1_registere.las
63.06 MB
-
191215_laupahoehoe_mixed2_registere.csv
112.29 KB
-
191215_laupahoehoe_mixed2_registere.las
52.05 MB
-
191219_Laupahoehoe_native_0to22_reg.csv
102.12 KB
-
191219_Laupahoehoe_native_0to22_reg.las
40.98 MB
-
191223_Laupahoehoe_native_22to40_re.csv
91.10 KB
-
191223_Laupahoehoe_native_22to40_re.las
36.85 MB
-
191227_havo_mixed_registered_georef.csv
149.17 KB
-
191227_havo_mixed_registered_georef.las
101.92 MB
-
191229_havo_nativeec_registered_geo.csv
146.81 KB
-
191229_havo_nativeec_registered_geo.las
77.76 MB
-
200103_havo_invasive_registered_geo.csv
114.10 KB
-
200103_havo_invasive_registered_geo.las
83.46 MB
-
200105_kalopa_native1_registered_ge.csv
81.04 KB
-
200105_kalopa_native1_registered_ge.las
49.04 MB
-
200105_kalopa_native2_registered_ge.csv
66.34 KB
-
200105_kalopa_native2_registered_ge.las
45.74 MB
-
200106_kalopa_mixed_registered_geor.csv
172.92 KB
-
200106_kalopa_mixed_registered_geor.las
103.77 MB
-
200107_kalopaanalogue_invasive_regi.csv
141.13 KB
-
200107_kalopaanalogue_invasive_regi.las
83.26 MB
-
README.md
5.32 KB
Abstract
During the past century, the proliferation of invasive species has contributed to the loss of biodiversity and ecosystem degradation. In forests, invasive tree species can alter ecosystem function, but the underlying mechanisms of these changes are not fully understood. We use the ongoing invasion of P. cattleianum on the Island of Hawaiʻi to test the hypotheses that invasive structural changes drive changes to forest evapotranspiration (ET). The aim of our study is first to quantify the structural changes to native ʻōhiʻa -dominated forest impacted by a gradient of P. cattleianum invasion. Our results suggest that invasive P. cattleianum causes significant changes to the vegetation density and structure of native forest on the Island of Hawai’i, including increased vegetation area index (VAI), decreased mean leaf height, and decreased structural heterogeneity. Second, we strove to understand the functional implications of structural changes through a biophysical modeling simulation, testing the sensitivity of ET to canopy structure under contrasting scenarios. Modeling the functional impact of structural change, we found that plots with P. cattleianum invasion importance value (IVinv) above 0.35 have a higher likelihood to increase ET compared to plots with P. cattleianum invasion less than 0.35 IVinv. Modeled increases in ET due to invasion ranged from 19% and 123% relative to native transects. The large variation in ET increases is caused by structural variation because the modeling scenarios did not include potential species differences in leaf physiology. Diagnostic scenario modeling shows the effect size of increased leaf area on modelled ET is constrained by the structural arrangement, i.e., vertical distribution, of the increased vegetation. Thus, an invasion structure that increases vegetation density in taller, more sunlit forest strata will lead to a greater increase in ET compared to an invasion structure that increases vegetation density in the shaded forest understory. Overall, we conclude that the vertical distribution of vegetation is an important factor shaping the impact of invasive P. cattleianum on the forest water balance.
This dryad repository contains:
- Point clouds of forest structure derived from terrestrial LiDAR for each sample site (.las files).
- The vegetation area index (VAI) for each 1 m3 voxel corresponding to each point cloud dataset (.csv files).
Dataset DOI: 10.5061/dryad.80gb5mm1g
Description of the data and file structure
Files and variables
There are 25 .csv files listed below:
191105_coopercenter_native1_georef_.csv
191108_coopercenter_mixed1_georef_c.csv
191207_olaahwy11_invasive1_georef_c.csv
191207_olaahwy11_invasive2_georef_c.csv
191202_olaa_mixed_20to40_georef_cla.csv
191211_laupahoehoe_invasive_0to20_r.csv
191210_olaa_mixed_0to20_georef_clas.csv
191120_olaa_native2_georef_classifi.csv
191120_olaa_native1_georef_classifi.csv
191219_Laupahoehoe_native_0to22_reg.csv
200105_kalopa_native2_registered_ge.csv
191229_havo_nativeec_registered_geo.csv
191227_havo_mixed_registered_georef.csv
191105_coopercenter_native2_georef_.csv
191213_laupahoehoe_invasive_20to40_.csv
191111_coopercenter_invasive1_geore.csv
191223_Laupahoehoe_native_22to40_re.csv
191108_coopercenter_mixed2_georef_c.csv
191215_laupahoehoe_mixed2_registere.csv
200103_havo_invasive_registered_geo.csv
191215_laupahoehoe_mixed1_registere.csv
191111_coopercenter_invasive2_geore.csv
200106_kalopa_mixed_registered_geor.csv
200105_kalopa_native1_registered_ge.csv
200107_kalopaanalogue_invasive_regi.csv
Description of .csv files:
Naming convention: The first 6 digits indicate the date the data was collected in yymmdd format. The name after the first underscore corresponds to the study site. The label after the second underscore corresponds to the transect type (native, mixed, or invasive) and number (some transects had to be split into 2 sections). If the name specifies a number range (i.e. 0to20), that indicates the portion of the 40m transect the data represents (in m). The remaining descriptive words in the file names include abbreviations of "georeferenced", "classified", and "registered". These are processing steps that apply to all the datasets and do not identify any important differences among datasets. Words were cut off according to a limit on file name characters.
Description: The .csv files contain a 3D voxelized matrix in which the transect's point cloud has been summarized into cubic meter voxels. The vegetation area index is the normalized LiDAR point density found in each cubic meter voxel. These .csv files are the input data processed by forestR to derive structural summary metrics.
Variables
- : An index column
- X: The value corresponds to x (m) in a 3D grid
- Y: The value corresponds to y (m) in a 3D grid
- Z: The value corresponds to z (m) in a 3D grid
- V1: The vegetation area index (VAI) (dimensionless), the normalized LiDAR point density in each cubic meter voxel.
There are 25 .las files listed below:
191108_coopercenter_mixed2_georef_c.las
191105_coopercenter_native2_georef_.las
191105_coopercenter_native1_georef_.las
191111_coopercenter_invasive2_geore.las
191111_coopercenter_invasive1_geore.las
191202_olaa_mixed_20to40_georef_cla.las
191108_coopercenter_mixed1_georef_c.las
191207_olaahwy11_invasive1_georef_c.las
191120_olaa_native2_georef_classifi.las
191120_olaa_native1_georef_classifi.las
191210_olaa_mixed_0to20_georef_clas.las
191219_Laupahoehoe_native_0to22_reg.las
191223_Laupahoehoe_native_22to40_re.las
191213_laupahoehoe_invasive_20to40_.las
191207_olaahwy11_invasive2_georef_c.las
191211_laupahoehoe_invasive_0to20_r.las
191215_laupahoehoe_mixed2_registere.las
191215_laupahoehoe_mixed1_registere.las
200105_kalopa_native1_registered_ge.las
200105_kalopa_native2_registered_ge.las
191229_havo_nativeec_registered_geo.las
200103_havo_invasive_registered_geo.las
191227_havo_mixed_registered_georef.las
200107_kalopaanalogue_invasive_regi.las
200106_kalopa_mixed_registered_geor.las
Description of .las files:
Naming convention: The first 6 digits indicate the date the data was collected in yymmdd format. The name after the first underscore corresponds to the study site. The label after the second underscore corresponds to the transect type (native, mixed, or invasive) and number (some transects had to be split into 2 sections). If the name specifies a number range (i.e. 0to20), that indicates the portion of the 40m transect the data represents (in m). The remaining descriptive words in the file names include abbreviations of "georeferenced", "classified", and "registered". These are processing steps that apply to all the datasets and do not identify any important differences among datasets. Words were cut off according to a limit on file name characters.
Description: The .las files contain the point clouds derived from terrestrial laser scanning (TLS) for each transect. These .las files have been normalized, thinned, and georeferenced according to the processing steps outlined in the supplement to the manuscript. These .las files are the input data required to derive the voxelized .csv files described above.
Code/software
.csv files can be opened using Google Sheets or other open-source software.
.las files can be opened with free software such as .lastools or Cloud Compare.
Access information
Other publicly accessible locations of the data:
- N/A
Data was derived from the following sources:
- Field collection
