Have a personal or library account? Click to login
Comparing the effectiveness of random forest and generalized linear models in predicting ungulate browsing impact on Kyiv Polissya’s young pine forests Cover

Comparing the effectiveness of random forest and generalized linear models in predicting ungulate browsing impact on Kyiv Polissya’s young pine forests

Open Access
|Mar 2025

Figures & Tables

Figure 1.

Location of the study area: 1 – Teterivske FE, 2 – Dniprovsko-Teterivske SFHE, 3 – Vyschedubechanske FE. Damage degree: low, up to 30% of shoots are damaged; moderate, 30–50% of shoots are damaged; high, <51% of shoots are damaged
Location of the study area: 1 – Teterivske FE, 2 – Dniprovsko-Teterivske SFHE, 3 – Vyschedubechanske FE. Damage degree: low, up to 30% of shoots are damaged; moderate, 30–50% of shoots are damaged; high, <51% of shoots are damaged

Figure 2.

Photo of the damage degree in Scots pine plants: A – low, up to 30% of shoots browsed; B – moderate, 30–50% of shoots browsed; C – high, 51% or more of shoots browsed
Photo of the damage degree in Scots pine plants: A – low, up to 30% of shoots browsed; B – moderate, 30–50% of shoots browsed; C – high, 51% or more of shoots browsed

Figure 3.

Scaled variable importance plot from a RFM selection process to achieve the smallest out-of-bag error for Vyschedubechanske FE, Dniprovsko-Teterivske SFHE and Teterivske FE. Panels represent the importance of variables: moose, deer, red_deer – ungulates density index; prop_area_Pine, Oak and R_Oak – the proportion of the young forest area of Scots pine, English oak and red oak. Higher values indicate increased variable importance
Scaled variable importance plot from a RFM selection process to achieve the smallest out-of-bag error for Vyschedubechanske FE, Dniprovsko-Teterivske SFHE and Teterivske FE. Panels represent the importance of variables: moose, deer, red_deer – ungulates density index; prop_area_Pine, Oak and R_Oak – the proportion of the young forest area of Scots pine, English oak and red oak. Higher values indicate increased variable importance

Detailed information about variables according to the damage inventory data from ungulate browsing at three experimental enterprises_ The number of measurements (n) represents the total number of surveyed plots_ Indicators are presented in the respective mean (±SD) for the three experimental enterprises for each variable

VariablesVyschedubechanske FETeterivske FEDniprovsko-Teterivske SFHE
26,746.00, ha32,169.00, ha18,477.00, ha
Young forest area (up to 10 years old), ha914.001,390.00210.00
Percentage of young forest area, %3.424.321.13
Number of measurements (n)1444190
Area of young forests damaged by ungulates, ha0.25 (±0.27)0.50 (±0.46)0.33 (±0.31)
Average age of trees, years3.38 (±1.01)2.85 (±1.01)3.02 (±1.37)
Proportion of areas with damages by ungulates to the total area of young forest0.027 (±0.03)0.036 (±0.03)0.159 (±0.15)
Proportion of areas with damage of the main species to the areas with damages0.25 (±0.05)0.49 (±0.12)0.27 (±0.07)
Average damage degree, score1.91 (±0.47)2.37 (±0.77)1.94 (±0.71)
Proportion of areas with low damage0.026 (±0.018)0.022 (±0.019)0.145 (±0.129)
Proportion of areas with moderate damage0.026 (±0.032)0.027 (±0.022)0.189 (±0.175)
Proportion of areas with high damage0.041 (±0.028)0.045 (±0.039)0.110 (±0.071)
Moose density index0.0040.0050.005
Red deer density index0.0090.0070.015
Roe deer density index0.0180.0210.017
Proportion of Scots pine area0.71 (±0.01)0.83 (±0.02)0.68 (±0.02)
Proportion of silver birch0.11 (±0.02)0.15 (±0.02)
Proportion of English oak0.04 (±0.01)0.04 (±0.01)0.17 (±0.02)
Proportion of red oak0.23 (±0.02)

Accuracy assessment of the RFM and GLM prediction models for Vyschedubechanske FE, DniprovskoTeterivske SFHE and Teterivske FE

Prediction modelsVyschedubechanske FETeterivske FEDniprovskoTeterivske SFHE
MSER2MSER2MSER2
RFM0.00050.870.00190.770.00380.89
GLM0.00030.920.00140.820.00670.80

Significance of features according to the results of the generalized linear model_ Values represent model coefficients indicating the effect size and direction of each variable on the predicted degree of browsing damage

FeaturesVyschedubechanske FE, n = 144Teterivske FE, n = 41Dniprovsko-Teterivske SFHE, n = 90
SEz-valuep-valueSEz-valuep-valueSEz-valuep-value
Intercept–1.72 (±1.16)–1.490.14–4.66 (±21.84)–0.2140.83–2.01 (±1.50)–1.3340.18
Moose density957.66 (±5276.23)0.1820.18135.28 (±989.14)0.1370.90–301.88 (±817.16)–0.3690.71
Deer density–555.56 (±3820.72)–0.1450.8817.52 (±1982.96)–0.0090.99–13.39 (±58.41)–0.2290.82
Roe deer density–49.84 (±442.62)–0.1130.91–69.27 (±1455.44)–0.0480.96179.92 (±465.09)0.3870.70
Age of plants0.03 (±0.27)0.1270.900.01 (±0.95)0.0180.99–0.10 (±0.27)–0.3760.71
Proportion of Scots pine, %–1.03 (±0.86)–1.2020.231.23 (±13.45)0.0910.92–0.005 (±1.14)–0.0040.99
Proportion of silver birch, %1.04 (±13.66)0.0770.94
Proportion of English oak, %–0.80 (±1.50)–0.5370.591.22 (±13.77)0.0890.92–0.23 (±0.99)–0.2390.81
Proportion of red oak, %–0.87 (±0.75)–1.1520.25
Forest type–0.03 (±0.50)–0.0630.95–0.01 (±0.72)0.0160.990.06 (±0.25)0.2240.82
Damage degree, score0.005 (±0.55)0.0100.99–0.06 (±0.67)–0.0900.93–0.05 (±0.37)–0.1480.88
Pearson chi20.347 0.125 0.98
Pseudo R-squ.0.028 0.040 0.121
DOI: https://doi.org/10.2478/ffp-2025-0001 | Journal eISSN: 2199-5907 | Journal ISSN: 0071-6677
Language: English
Page range: 1 - 11
Submitted on: Mar 9, 2024
|
Accepted on: Jan 2, 2025
|
Published on: Mar 7, 2025
In partnership with: Paradigm Publishing Services
Publication frequency: 4 issues per year

© 2025 Natalia Vysotska, Oleksandr Khromuliak, Oleksandr Borysenko, Maksym Rumiantsev, Iryna Yashchuk, Oleksandr Kipran, published by Forest Research Institute
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.