Browsing by Subject "Random Forest"
Now showing items 1-2 of 2
-
(2014)In forest inventories, more and more detailed information about the constantly growing stock is intended to obtain both at national and at private forests level. At present, in forest planning the features describing wood quality are rarely estimated from standing trees since there are limited resources for the precise measurements of the trees due to high expenses. The principal aim of this study was to determine the precision whereby the externally reviewed predictive features of the internal quality of a log-size pine wood can be estimated manually using Terrestrial Laser Scanning (TLS). The examined features were tree height, diameter at breast height, upper diameter as well as the heights of the lowest dead and living branch. The second main objective was to determine the precision whereby the tree class can be predicted based on measured and derived tree attributes. The derived attributes were the volume of the wood, crown ratio, the relation of dead branched and branch free part of the tree to the tree height, and form factor. For forecasting the nearest neighbor method was used where the search for the nearest neighbors was performed using the Random Forest -method. The relative accuracy (RMSE %) of TLS data in relation to the reference field data was found to be 7.54% (bias -6.16%) for the tree height, 6.39% ( -2.46%) for the breast height diameter, 10.01% (0.40%) for the upper diameter, 9.21% ( -5.99%) for the height of the lowest living branch and 34.95% ( -1.47%) for the height of the lowest dead branch. On the prediction of the tree class indicating the stem quality, the TLS data reached 78 % classification accuracy (5 tree classes). With harsher three tree class categorization 87% classification accuracy was reached. Based on the results can be said that quality factors, such as the lowest branches can be measured from the TLS data with reasonably adequate accuracy. Also the prediction of the tree class turns out decently (5 classes) and with harsher categorization (3 classes) well. The forecasting method described in this study can still be improved for example by the automatic interpretation of the laser scanning data, as well as combining several laser scanning points from the examined target. The most potential near future application is that TLS data can work as reference for airborne laser scanning because for this purpose the harsher categorization accuracy seems to be already very promising.
-
(2024)Soil rutting in forest operations is a critical phenomenon, characterized by depressions or tracks on the forest floor, often caused by heavy machinery use such as logging equipment. These disturbances can have profound impacts on forest health and ecosystem integrity, disrupting soil structure, compacting soil layers, and altering drainage patterns. Consequently, soil erosion may occur, leading to the loss of topsoil, nutrient depletion, and reduced soil fertility, ultimately affecting forest vegetation growth and vitality, while also posing aesthetic concerns. This study aims to explore the relationship between soil rutting and various influencing factors by utilizing Random Forest machine learning model to predict soil rut severity based on these predictors within forest stands across Finland. Encompassing 40 forest stands comprising 420 sample plots and 425 rut measurement points, the study aims to produce soil rut susceptibility maps to aid forest managers and stakeholders in identifying high-risk areas and implementing targeted mitigation strategies. The study seeks to understand the dynamic relationship between soil rutting and its controlling factors, while also determining the most influential contributors to severity of rut in forest operations in both mineral soils and peatland forest stands. The rut severity index (RSI) which was defined as the number of rut occurrences exceeding 10 cm in depth within a 30m width measurement area, served as our main target for prediction. The findings indicate that tree volume in mineral soils and aspect in peatlands are the most important variables explaining rut occurrences. Furthermore, disparities in susceptibility to soil rutting are observed among different soil types, particularly between mineral soils and peatlands. The Random Forest predictive models showed better performances in mineral forest stands than in peatland stands, with their ability to predict approximately 72% of the variability in rut severity index in mineral soils and 26% in their counterparts. Accuracy assessment was conducted using the Mean Squared Error (MSE), a measure of the average squared difference between predicted and actual rut severity values. A lower MSE indicates a closer alignment between predictions and actual values. Our study found that the Random Forest model achieved a MSE of 0.839 for mineral soils, indicating more precise predictions compared to peatlands, which exhibited a higher MSE of 25.341. This study underscores the importance of comprehending the interplay between soil rutting and its controlling factors for sustainable forest management, as well as the application of machine learning techniques, providing robust capabilities for future research, enabling more accurate predictions and informed decision-making in forest management practices.
Now showing items 1-2 of 2