Assessment of the Effects of the Biotic and Abiotic Harm Factors on the Amount of Industrial Wood Production with Deep Learning

doi:10.21203/rs.3.rs-2103669/v1

Download PDF

Research Article

Assessment of the Effects of the Biotic and Abiotic Harm Factors on the Amount of Industrial Wood Production with Deep Learning

https://doi.org/10.21203/rs.3.rs-2103669/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 16 Jan, 2023

Read the published version in Environmental Science and Pollution Research →

You are reading this latest preprint version

The protection and sustainability of forest assets is possible with the planned production of forest products with minimum loss. Among the products obtained from forests, industrial wood is the most important raw material for many sectors. For this reason, it is important to determine the factors affecting the industrial wood production amount in order to obtain this raw material in the most efficient way. This article aims to investigate and assess the effects of two biotic and two abiotic harm factors on the amount of industrial wood production by building a deep learning estimation model. These factors are forest fires, outbreaks of insects, outbreaks of diseases, and severe weather events. The study shows that the most harmful factor decreasing the industrial wood production level is outbreaks of diseases. The second effective factor, however, appears to be severe weather events. The third and the fourth factors were determined to be outbreaks of insects and burned forest areas, respectively.

Industrial wood

wood production

sustainable forestry

biotic factors

abiotic factors

deep learning

Forests are important natural resources. In addition, forest products are indispensable for many needs. Wood is the main product of forestry and it is used for many purposes. Apart from the wood used as fuel, the wood used in the production of industrial products is called industrial wood. The annual level of wood production in the world is around 3.4 billion m³, 53% of which is firewood and 47% is industrial wood. United Nations (UN) reports that, while approximately 90% of the firewood is produced and consumed by the developing countries, 79% of industrial wood is produced by the developed countries (UN, 2011). Industrial wood is the most important raw material of the sectors such as construction, furniture, or kitchen sectors, which are directly affected by any increase or decrease in industrial wood production. Therefore, it gains importance to examine harm factors affecting industrial wood production by building an appropriate and accurate estimation model.

In the literature, when the studies estimating the amount of industrial wood or investigating the factors affecting this amount are examined, it is seen that they are not very numerous. Among these studies, Söderberg (1988) described a method of constructing functions that are based on long-term forecasting of timber productions by using a growth simulator. Stage and Renner (1988) made a comparison of the production-forecasting techniques using long-term stand records. MacLean (1990) reviewed the impacts of pests and fire on stand growth and timber production and reported that damaging agents reduced the tree growth, killed the trees, and destroyed their commercial values. Murray and von Gadow (1993) developed a flexible production model for regional timber forecasting depending on age-class based timber resource projection. Vanclay (1996) reviewed the tools, the techniques, and the information sources for estimating timber productions. Pukkala et al. (2003) developed and demonstrated a new method to make forecasts on the future timber supply. Churakov et al. (2011) studied the influence of heartwood rot on industrial wood production in aspen stands. Yildirim (2012) investigated the industrial wood production and consumption in Turkey and concluded that the industrial wood production would decrease in future, as the demand would increase. Moreover, de Moraes Goncalves et al. (2013) studied the integration of genetic and silvicultural strategies for minimization of abiotic and biotic constrains in Brazilian eucalypt plantations. In addition, Atik and Yilmaz (2014) performed a study to determine the factors affecting the industrial wood production in Turkey using various conventional statistical analyses. They examined 26 variables and they found that the most important factors affecting the industrial wood production are fertility, aspect of land, skidding method, stand structure, skidding distance, growing stock, transportation, and harmful abiotic factors. They also reported that insect harm was among the most dangerous forest threats after human intervention. In another study, Atik and Aslan (2015) built an artificial neural network model to predict the industrial wood production yield in forestry operations according to the allowable cut in Günye, Turkey. In addition, Buongiorno (2015) suggests a model to predict timber prices and production. Sierota et al. (2019), however, examined the abiotic and biotic disturbances affecting the forest heath in Poland and reported that forests in Poland are vulnerable to disease and insect outbreaks. Jactel et al. (2009) performed a study about the influences of forest stand management on biotic and abiotic risks of damage. Marais et al. (2022) reviewed the biotic and abiotic influence factors for wood durability in terrestrial and aquatic environments. Kernaghan et al. (2003) examined the biotic and abiotic factors affecting ectomycorrhizal diversity in boreal mixed woods. Tippmann et al. (2006), however, studies the common themes in biotic and abiotic signaling in plants. Additionally, Jones et al. (2019) performed as study about wood decomposition in aquatic and terrestrial systems in the tropics comparing the biotic and abiotic processes. In another study by Jactel and Vodde (2011), prevalence of biotic and abiotic hazards in European forests was studied. They performed the study in 21 European countries for the years 1994 to 2005. Moreover, Erdenebileg et al. (2020) examined the abiotic and biotic drivers of long-term wood decomposition. Wilkaniec et al. (2021) studied the biotic and abiotic factors causing the collapse of Robinia pseudoacacia L. veteran trees in urban places. Additionally, van Lierop et al. (2015) investigated the global forest area disturbance from fire, insect pests, diseases, and severe weather events. Moreover, Polle and Rennenberg (2019) examined the physiological responses to biotic and abiotic stress in the forest trees. Nawrot-chorabik et al. (2022), however, examined the impact of biotic and abiotic stress factors on the development of European ash tissue cultures. In addition, Yuan et al. (2018) performed a study concerning biotic and abiotic determinants of coarse wood productivity in temperate mixed forests. Dossa et al. (2020) quantified the factors affecting wood decomposition across a tropical disturbance gradient. Hennon et al. (2020) provided a framework to evaluate the climate effects on the forest tree diseases. In addition, Leclerc et al. (2020) studied the complex interactions with biotic and abiotic disturbances in managing wildlife habitat.

Research gap and motivation of the study

Forests are affected by various biotic or abiotic harm factors. These factors affecting forests also directly affect the production of industrial wood, which is an important raw material. For this reason, it is important to assess the extent and importance of these factors on industrial wood production in order to take precautions when possible. Thus, the objective of our study is to investigate and assess the effects of four biotic and abiotic harm factors, which are forest fires, outbreaks of insects, outbreaks of diseases, and severe weather events, on the amount of industrial wood production by building a deep learning estimation model.

2.1. Artificial neural networks

Artificial neural networks are models that imitate the working mechanism of the nerve cells in the human brain, in a mathematical structure. The basic structure of artificial neural networks is made up of artificial nerve cells, called neurons. To form an artificial neural network, neurons need to work together. Neurons are also given names such as nodes, units, or processing elements. In an artificial neural network, the structure formed by neurons coming together in the same line is called a layer. Layers are also basic parts of artificial neural networks. There are three types of layers in an artificial neural network, which are the input layer, the intermediate (hidden) layer, and the output layer. The layers are also in contact with each other.

Neurons in the input layer are responsible for reception the input data transmitted to the model. However, they do not proceed with the input data. Instead, they only transmit the input data to the hidden layers. The hidden layers are responsible for building the model. Finally, the output layers are obliged to transmit the output values, obtained in the model, to the required environments. While an artificial neural network has only one input and one output layer, the number of hidden layers may be more than one. Any neuron in the hidden layer proceeds with the information it receives from the input layer and transmits it to the neurons in the next layer. However, in some types of artificial neural networks called feedback networks, this transmission is in the form of transmitting to the previous layer or to the input layer. The first step in the working mechanism of a neuron is the reception of the inputs and weights from the outer world or from the other neurons. These inputs and weights may be provided by sample data or by the outputs of other neurons. After the reception of inputs and weights, the neuron expresses this information as an additive model called the net function. The output of the neuron is formed by a function called the activation function whose input is the net function. The general expression of a net function is given in Eq. (1).

$$a=\sum {w}_{i}{x}_{i}$$

where ${x}_{i}$ denotes the input values and ${w}_{i}$ denotes the corresponding weights. There are many different types of activation functions. For example, sigmoid function is a popularly used activation function whose structure is given by Eq. (2).

$$F\left(a\right)=\frac{1}{1-{e}^{a}}$$

where $a$ is the value resulted from the net function in Eq. (1). The result produced by the activation function is the output of the neuron. Although neurons may have input values more than one, they have just one output value. Figure 1 briefly displays the working mechanism of a neuron.

An artificial neural network needs to be trained to produce accurate and consistent results. The training process is to make the network gain the ability of producing the correct outputs by calculating the necessary coefficients with the help of the input values in the sample data set.

Artificial neural network models have some advantages over the classical models. One of this advantaged is that they store the information over the entire network and a few pieces of information loss do not prevent the model from working. Additionally, they can produce outputs without getting affected by errors or corruptions in the data drastically. Moreover, the model can be trained with the help of other examples and previous studies, as the network is also capable of learning from similar networks. Artificial neural networks can work together to complement each other. Thus, they can perform multiple tasks at the same time. Another advantage of the artificial neural networks is that they can solve complex and nonlinear problems more successfully than the conventional approaches. They also do not require any assumptions or prerequisites in terms of data and model structure. Artificial neural networks are also able to learn from data and make decisions. Unlike the other classical models requiring fixed variance, they are still able to model the data when the assumption of fixed variance does not hold.

However, there are also some disadvantages of the artificial neural networks such as their performances are heavily dependent on the hardware level of the computer used. Additionally, there are not definite rules to follow to build a model except for a few thumbnail rules. Thus, building an appropriate model may depend on personal experiences and trials. As causality is not considered when building a model, they may produce unexpected results. Moreover, there is a necessity of converting all the input values into numerical data, which can be difficult sometimes for some kinds of variables. During the learning phase, it is not possible to observe how artificial neural networks operate. Thus, it is not possible to explain the reason behind the output values. Finally, it cannot be guaranteed that the estimated artificial neural network is the optimum one.

2.2. Multilayer perceptron model

A multilayer perceptron model can be defined as an artificial neural network, which is based on the perceptron model. The perceptron model is simply a single layer artificial neural network. Perceptron itself can also be defined as a type of neuron, where the net function includes the sum of weighted data and a bias value. During the introduction period of artificial neural networks, it was seen that the perceptron model was insufficient in solving nonlinear problems. Thus, Rumelhart et al. (1986) proposed the multilayer perceptron model by adding a new layer of neurons called the hidden layer, which exists between the input layer and the output layer, to the perceptron model. Artificial neural networks having more than two layers are also called deep neural networks. Thus, the multilayers perceptron models are also deep neural network or deep learning models. An example of multilayer perceptron model is demonstrated in Fig. 2.

Multilayer perceptron model is a supervised learning model. The model is trained by a sample input and output data set so that the network learns to produce outputs that are close to the real output values existing in the data set. The difference between the observed data and the estimated data is called error ($\epsilon$). During the training process, the model adjusts the parameters, or the weights and biases, to reduce the error, so that to produce the most accurate outputs. The most important property of the multilayer perceptron model is the backpropagation process, which means recursively adjusting the weights to minimize the difference between the observed and estimated values. During the backpropagation process, the total loss is propagated back into the network to calculate the amount of loss for which each neuron is responsible. Meanwhile, the weights are re-adjusted to minimize the loss and the process continues until the amount of error is stabilized. The error can be measured in various ways such as the mean absolute error (MAE), the root mean squared error (RMSE), the relative absolute error (RAE), and the root relative squared error (RRSE). MAE can be defined as the mean of the absolute values of the error terms. Calculation of MAE is given as follows.

$$MAE=\frac{1}{n}\sum _{i=1}^{n}\left|{\epsilon }_{i}\right|$$

where $n$ is the data size and ${\epsilon }_{i}$ are the error terms. RMSE is the square root of the mean value of the squared error terms, which can be calculated as follows.

$$RMSE=\sqrt{\frac{1}{n}\sum _{i=1}^{n}{\epsilon }_{i}^{2}}$$

RAE and RRSE values, however, can be calculated by the following equations.

$$RAE=\frac{\sum _{i=1}^{n}\left|\widehat{x}-x\right|}{\sum _{i=1}^{n}\left|\stackrel{-}{x}-x\right|}$$

$$RRSE=\sqrt{\frac{\sum _{i=1}^{n}{\left(\widehat{x}-x\right)}^{2}}{\sum _{i=1}^{n}{\left(\stackrel{-}{x}-x\right)}^{2}}}$$

where $x$ is the observed data, $\widehat{x}$ is the estimated data and $\stackrel{-}{x}$ is the mean value of $x$.

The correlation coefficient ($\rho )$ indicates the direction and the degree of the linear relation between the actual data instances and the estimated instances produced by the constructed model. Calculation of the correlation coefficient is as follows.

$$\rho =\frac{\sum _{i=1}^{n}\left[x-E\left(x\right)\right]\left[\right(\widehat{x}-E\left(\widehat{x}\right)]}{\sqrt{{\sum _{i=1}^{n}\left[x-E\left(x\right)\right]}^{2}\sqrt{{\sum _{i=1}^{n}\left[\right(\widehat{x}-E\left(\widehat{x}\right)]}^{2}}}}$$

where $x$ is the observed data and $\widehat{x}$ is the estimated data given by the model. $E\left(x\right) and E\left(\widehat{x}\right)$ are the expected values of $x$ and $\widehat{x}$, respectively. The correlation coefficient $\rho$ is a value between − 1 and 1. A $\rho$ value close to 1 indicates a strong positive linear relationship between the observed and estimated data instances while a $\rho$ value close to -1 is an evidence of a strong negative linear relationship. If the $\rho$ is 0, there is no relationship between $x$ and $\widehat{x}$.

When building artificial neural networks, there are not definite rules to decide about the number of layers, and the number of neurons existing in each layer. These values are highly dependent on personal experiences and trials. However, during the model selection phase, the models with higher correlation coefficients and lower error measures are preferred.

2.3. Study area

In Turkey, the owner and the manager of the state forests is General Directorate of Forestry (GDF), which is divided into regional directorates. In addition, Turkey is among the industrial wood producer countries in the world. Turkish Ministry of Development (MOD) reports that the annual wooden raw material consumption in Turkey is approximately 25.9 million m³. Moreover, 15.5 million m³ (60%) of wood is produced from the state forests by GDF, while 4.8 million m³ (19%) of the production belongs to the private forestry companies. It is declared that 15% (4 million m³) of the wood production comes from illegal harvesting from the state forests, while 6% (1.6 million m³) is obtained by importing (MOD, 2014). Figure 3 presents the annual industrial wood production quantities (m³) in Turkey between 2000 and 2022.

When Fig. 3 is examined, it is seen that the levels of the industrial wood production display a decreasing trend from 2000 to 2003. However, except for the slight decreases in 2009, 2013, and 2017, it is possible to say that the industrial wood production levels tend to increase every year.

Figure 4 shows the regional directorates in Turkey and the industrial wood production densities in these regions as of 2020 (GDF, 2020).

It can be seen in Fig. 4 that industrial wood production is denser in the western and the mid-western parts of Turkey. At the same time, it is clear that the areas where the most intensive production is carried out are generally the areas near the sea. Figure 5 presents the amounts of industrial wood (m³) produced in the regional directorates as of 2020.

Figure 5 suggests that the highest level of industrial wood production is observed in Kastamonu directorate, which significantly differs from the other directorates. The first four directorates Kastamonu, Bolu, Zonguldak, and Amasya having the highest production levels are all in the Black Sea region that forms the northern part of Turkey. Bursa and İzmir, which have the second highest production levels, exist in the western part of Turkey. The third highest production levels, however, are observed in the directorates Muğla, Adana, and Antalya, which are in the Mediterranean region forming the southern part of Turkey.

Figure 6 presents the kinds and the percentages of the industrial wood products produced from coniferous and non-coniferous wood in Turkey.

Figure 6 indicates that log that is produced from coniferous wood is the most popular product (42%). The second and third most popular products produced by using coniferous wood are fiber-chip wood (32%) and pulpwood (17%), respectively. When non-coniferous wood is considered, however, the most popular product appears to be fiber-chip wood (56%). The second and the third popular products produced by using non-coniferous wood are log (29%) and pulpwood (8%), respectively.

2.4. Data and variables

The data were collected from various resources. In the study, two reports of FAO were used to derive the data belonging to the variables (Koc et al., 2014; Özel et al., 2020). Additionally, a forestry sector report provided by The Union of Chambers and Commodity Exchanges of Turkey (UCCET) was also referred (UCCET, 2011). Additionally, two reports of GDF were also used (GDF, 2012; GDF, 2020). Moreover, a report about Turkish forestry published by Foresters' Association of Turkey (FOT) and the forestry statistics published by Ministry of Agriculture and Forestry (MAF) were referred to collect the data used in the study (FOT, 2019; MAF, 2019).

The data involve one dependent variable, which is the industrial wood production level, and four dependent variables, which are burned forest area, outbreaks of insects, outbreaks of diseases and severe weather events. The available data consist of 18 cases belonging to the years between 2000 and 2017. The variables, their units of measurement, and descriptions are provided in Table 1.

Table 1

The variables employed in the study
Variables	Descriptions
Industrial wood production	The yearly level of industrial wood production (m³)
Burned forest area	The yearly amount of burned forest areas (1000 ha) in forest fires
Outbreaks of insects	The yearly amount of harmed forest areas (1000 ha) due to outbreaks of insects
Outbreaks of diseases	The yearly amount of harmed forest areas (1000 ha) due to outbreaks of various diseases
Severe weather events	The yearly amount of harmed forest areas (1000 ha) due to severe weather events

Figure 7 presents the yearly amounts of harmed forest areas with respect to burned forest area, outbreaks of insects, outbreaks of diseases and severe weather events factors.

3.1. Development of the deep learning model

Using the collected data, an artificial neural network model based on the multilayer perceptron architecture was developed to estimate the yearly industrial wood production level in Turkey. The flowchart provided in Fig. 8 demonstrates the model development phases of the study.

Weka (2021) software was used to build the artificial neural network model. During the model validation stage, the data were split as 75% for training and 25% for testing. After trying various parameter combinations, the artificial neural network model with two hidden layers, each containing four and two neurons, respectively gave the best performance scores. An iteration period of 500 epochs was applied to estimate the model. Figure 9 demonstrates the layers and the neurons of the constructed artificial neural network model, based on the multilayer perceptron architecture.

Table 2 displays the actual and the estimated instances with the corresponding errors calculated for the testing data.

Table 2

The actual instances, the estimated instances, and the errors for the testing data
Instant number	Actual industrial wood production (m³)	Estimated industrial wood production (m³)	Error
1	4,913,900	5,995,542.74	1,081,642.74
2	9,299,000	5,607,923.50	-3,691,076.50
3	12,569,000	11,391,383.56	-1,177,616.44
4	16,637,598	15,885,947.22	-751,650.78

As 25% of the 18 instances were allocated for the training set, the training set involves four cases. When the results provided in Table 2 are examined, it is seen that the actual data used for testing and the corresponding estimated values are satisfactorily close to each other. The performance scores for the constructed artificial neural network model are presented in Table 3.

Table 3

Model performance scores
Metrics	Score
Correlation coefficient	92%
Mean absolute error (MAE)	1,675,496.6
Root mean squared error (RMSE)	2,046,078.9
Relative absolute error (RAE)	44.70%
Root relative squared error (RRSE)	47.43%

When the model performance scores in Table 3 are examined, it is seen that the correlation coefficient, calculated for the actual values and the estimated values, appears to be 92%. This coefficient value is quite close to 1. In this case, it is possible to conclude that the degree of linearity between the actual and the estimated instances is quite high. This means that the estimated instances are quite close to the actual instances, which implies that the estimation performance of the constructed artificial neural network model is quite satisfactory. Figure 10 presents the comparison of the actual and the estimated instances for the testing data.

It can be seen in Fig. 10 that the actual and the estimated values are satisfactorily close to each other. Thus, the constructed artificial neural network successfully estimates the industrial wood production level.

3.2. Assessment of the harm factors

To assess the effects of the harm factors on industrial wood production, an attribute selection study on the constructed deep learning model is required. Thus, the abiotic and biotic harm factors, which form the variables of the model, and their ranked attributes that are based on the calculated correlation coefficients are presented in Table 4.

Table 4

The attributes of the harm factors
Inputs	Ranked attributes
Outbreaks of diseases (ha)	0.86
Severe weather events (ha)	0.45
Outbreaks of insects (ha)	0.29
Burned forest areas (ha)	0.26

When the ranked attribute weights provided in Table 4 are examined, it is seen that industrial wood production level is decreased mostly by outbreaks of diseases in forests. The second effective factor decreasing industrial wood production quantity appears to be severe weather events. The third effective harmful factor, however, is outbreaks of insects occurring in forest areas. Finally, the least effective harm factor decreasing industrial wood production level turns out to be burned forest areas in fire events.

When the findings of our study are compared with other studies, some similarities and differences stand out. For instance, Atik and Yilmaz (2014) only consider the abiotic harm factors among the other harm factors. In our study, however, as well as two abiotic factors (forest fires and severe weather events), two biotic factors (outbreaks of diseases and outbreaks of insects) are also taken into consideration. Moreover, they report that harmful abiotic factors are among the most important factors affecting the industrial wood production level. In our study, however, the most important factor reducing the industrial wood production level appeared to be a biotic factor, which is the outbreak of diseases. In addition, Atik and Yilmaz (2014) also state that insect harm is the most dangerous forest threat after human intervention. On the contrary, our study suggests that there are more harmful factors than insect harm affecting the industrial wood production level, such as outbreaks of diseases and severe weather events. Similarly, de Moraes Goncalves et al. (2013) also report that susceptibility to diseases are among the most important selective factors for genotype allocation. Moreover, in accordance with the results of our study, Sierota et al. (2019) state that outbreaks of insects and diseases caused serious problems in Polish forests over the past 30 years. They also found out that outbreaks of insects often co-occurred with increased incidences of fungal diseases. However, Sierota et al. (2019) also considered drought and atmospheric pollution factors, which were not included in our study, among abiotic harm factors. Jactel and Vodde (2011), in their study involving the forests of 21 European countries, determined that the most harmful factors to forests were biotic factors with a rate of 60%, followed by abiotic hazard factors with a rate of 20% and anthropic hazard factors with a rate of 20%. This result is partially compatible with the result of our study, such that a biotic factor, which is outbreaks of diseases, was found to be the most harmful factor also in our study. Meanwhile, Erdenebileg et al. (2020) discuss the UV light, which is another factor not included in this study, among the abiotic factors affecting long-term wood decomposition. Hennon et al. (2020) defines a term called abiotic diseases, which does not require the presence of a pathogen. Abiotic diseases are explained as drought, water inundation, excessive heat and freezing factors that push the physiological limits of trees excessively. On the other hand, in our study, these factors are grouped under the common title severe weather events. Two other abiotic harm factors that were not included in our study are examined by Nawrot-chorabik et al. (2022). These factors are heavy metal pollution and salinity in soils. However, while the impact of the salinity factor turned out to be insignificant in the study, the impact of heavy metal pollution was found to be significant on the development of European ash (Fraxinus excelsior L.). In addition, the results provided by van Lierop et al. (2015) partially agree with the results of our study, such that fire, insects, diseases, and severe weather events, which are also included in our study, are used and it is stated that the second most damaging factor to forests worldwide is severe weather events. However, in the aforementioned study, insects were reported as the most harmful factor and diseases as the third harmful factor, which is the opposite of the order in our study. Moreover, van Lierop et al. (2015) also suggest that the order of importance of forest fires change with respect to the continents.

In conclusion, in this study, the assessment of the effects of biotic and abiotic harm factors on the amount of industrial wood production as well as the estimation of the yearly amount of industrial wood production was made through an artificial neural network, which is based on a multilayer perceptron deep learning model. Four kinds of biotic and abiotic forest harm factors, which are burned forest area, outbreaks of insects, outbreaks of diseases, and severe weather events, were taken into consideration. The estimated artificial neural network model shows that deep learning models can be used in industrial wood production level estimation very efficiently, as they have no restrictions like linearity or fixed variance, which are required by some conventional estimation models. The deep learning model developed in this study has quite satisfactory performance metrics such as a high correlation coefficient of 92% and a low relative absolute error score of 44.70%. Moreover, the study shows that the most significant harm factor decreasing the industrial wood production level is outbreaks of diseases. The second most harmful factor, however, appears to be severe weather events. The third and the fourth factors were determined to be outbreaks of insects and burned forest areas, respectively. As a continuation of this study, it is possible to make a comparison between the results provided by deep learning and regression models, which can be constructed using the same or different harm factors.

Compliance with Ethical Standards

Funding: This study was not funded by anyone or any establishment.

Conflict of Interest: The sole author of this manuscript declares no conflict of interests.

Ethical Approval: This article does not contain any studies with human participants or animals performed by the author.

Author Contributions: The sole author of the manuscript is solely responsible for all the contributions made in the article.

Consent to Participate: The sole author of the manuscript consents to participate.

Consent to Publish: The sole author of the manuscript consents to publish.

Availability of Data and Materials: The data used in this study is publicly available in the reference link provided.

Atik, A. and Aslan, F. 2015. Estimation of annual industrial wood production level in forestry operations with the artificial neural network. AJAEES 7(4): 1-17. https://doi.org/10.9734/AJAEES/2015/20028
Atik, A. and Yılmaz, B. 2014. Factors affecting industrial wood, material production in Turkey’s natural beech forests. Ann For Res 57(1): 149-162. http://dx.doi.org/10.15287/afr.2014.182
Buongiorno, J. 2015. Global modelling to predict timber production and prices: the GFPM approach. Forestry (Lond) 8(3): 291-303. https://doi.org/10.1093/forestry/cpu047
Churakov, B. P., Kornilina, V. V. and Zamaldinov, I. T. 2011. The influence of heartwood rots on industrial wood production in aspen stands (in Russian). Lesovedenie (2): 19-24. http://old.cepl.rssi.ru/lesovedenie2011_2_churakov_en.htm
de Moraes Goncalves, J. L., Alvares, C. A., Higa, A. R., Silva, L. D., Alfenas, A. C., Stahl, J., ... & Epron, D. (2013). Integrating genetic and silvicultural strategies to minimize abiotic and biotic constraints in Brazilian eucalypt plantations. Forest ecology and management, 301, 6-27.
Dossa, G. G., Paudel, E., Schaefer, D., Zhang, J. L., Cao, K. F., Xu, J. C., & Harrison, R. D. (2020). Quantifying the factors affecting wood decomposition across a tropical forest disturbance gradient. Forest ecology and management, 468, 118166.
Erdenebileg, E., Wang, C., Ye, X., Cui, Q., Du, J., Huang, Z., ... & Cornelissen, J. H. (2020). Multiple abiotic and biotic drivers of long‐term wood decomposition within and among species in the semi‐arid inland dunes: A dual role for stem diameter. Functional Ecology, 34(7), 1472-1484.
FOT (2019). Foresters' Association of Turkey, Turkish Forestry: 2019 (in Turkish), Kuban Matbaacılık Yayıncılık, Ankara, Turkey. https://www.google.com/url?sa=t&rct=j&q=&esrc=
s&source=web&cd=&ved=2ahUKEwiKtvXa4sbwAhXsqJUCHRCzBEgQFjAAegQIAxAD&url=
https%3A%2F%2Fwww.ormancilardernegi.org
%2Fdosyalar%2Ffiles%2Frevize_rapor7%2520web.pdf&usg=AOvV
aw1pv7FXomMiICiTzTRyVMO3 Accessed 8 September 2022
GDF 2012. General Directorate of Forestry, Department of Business and Marketing, 2012 report (in Turkish), Ankara, Turkey. https://9lib.net/document/zkw6ovj1-orman-genel-m%C3%BCd%C3%BCrl%C3%BC
%C4%9F%C3%BC-i%CC%87%C5%9Fletme-ve
-pazarlama-dai%CC%87re-ba%C5%9Fkanli%C4%9Fi.html Accessed 8 September 2022
GDF (2020). General Directorate of Forestry, End of November 2020 stamp, production, sales, stock status report (in Turkish), https://www.ogm.gov.tr/tr/
e-kutuphane-sitesi/UretimS
atisveStokFaaliyetleri/2020%20YILI%20KASIM%20SONU%20
DAMGA%20%C3%9CRET%C4%
B0M%20SATI%C5%9E%20STOK%20DURUMU.xlsx Accessed 8 September 2022
Hennon, P. E., Frankel, S. J., Woods, A. J., Worrall, J. J., Norlander, D., Zambino, P. J., ... & Shaw III, C. G. (2020). A framework to evaluate climate effects on forest tree diseases. Forest Pathology, 50(6), e12649.
Jactel, H., Nicoll, B. C., Branco, M., Gonzalez-Olabarria, J. R., Grodzki, W., Långström, B., ... & Vodde, F. (2009). The influences of forest stand management on biotic and abiotic risks of damage. Annals of Forest Science, 66(7), 701-701.
Jactel, H., & Vodde, F. (2011). Prevalence of biotic and abiotic hazards in European forests. EFI Technical Report 66 (518128), 2011.hal-02804469. https://hal.inrae.fr/hal-02804469/document
Jones, J. M., Heath, K. D., Ferrer, A., Brown, S. P., Canam, T., & Dalling, J. W. (2019). Wood decomposition in aquatic and terrestrial ecosystems in the tropics: contrasting biotic and abiotic processes. FEMS microbiology ecology, 95(1), fiy223.
Kernaghan, G., Widden, P., Bergeron, Y., Légaré, S., & Paré, D. (2003). Biotic and abiotic factors affecting ectomycorrhizal diversity in boreal mixed‐woods. Oikos, 102(3), 497-504.
Koc, M., Demirci, M., Karagoz, G., Kalayci, K., Ozel, A., Tas, S. and Kayacan, B. 2014. Global forest resources assessment 2015 country report Turkey, FAO publication, Rome, Italy. http://www.fao.org/3/a-az358e.pdf
Leclerc, M. A. F., Daniels, L. D., & Carroll, A. L. (2021). Managing wildlife habitat: Complex interactions with biotic and abiotic disturbances. Frontiers in Ecology and Evolution, 9, 613371.
MacLean, D. A. 1990. Impact of forest pests and fire on stand growth and timber yield: implications for forest management planning. Can J For Res 20(4): 391-404. https://doi.org/10.1139/x90-057
MAF (2019). Forestry Statistics 2019 (in Turkish). https://www.ogm.gov.tr/tr/e-kutuphane/resmi-istatistikler Accessed 8 September 2022
Marais, B. N., Brischke, C., & Militz, H. (2022). Wood durability in terrestrial and aquatic environments–A review of biotic and abiotic influence factors. Wood Material Science & Engineering, 17(2), 82-105.
MOD (2014). Tenth development plan (2014-2018) sustainable forest management: special commission report 2023 (in Turkish). Ministry of Development, Ankara, Turkey. https://www.sbb.gov.tr/wp-content/uploads/2022/08/Onuncu-Kalkinma-Plani-Surdurulebilir-Orman-Yonetimi-Ozel-Ihtisas-Komisyonu-Raporu.pdf Accessed 8 September 2022
Murray, D. M. and von Gadow, K. 1993. A flexible yield model for regional timber forecasting. Southern J Appl Forestry 17(3): 112-115. https://doi.org/10.1093/sjaf/17.3.112
Nawrot-Chorabik, K., Sułkowska, M., Osmenda, M., Mohytych, V., Surówka, E., & Latowski, D. (2022). The Impact of biotic and abiotic stress factors on development of European ash tissue cultures. Forests, 13(1), 59.
Özel, A, Koç, M., Cevirme, M., Özkal, M.K. and İpek, N. 2020. Global forest resources assessment 2020 report Turkey, FAO publication, Rome, Italy. http://www.fao.org/3/cb0081en/cb0081en.pdf
Polle, A., & Rennenberg, H. (2019). Physiological responses to abiotic and biotic stress in forest trees. Forests, 10(9), 711.
Pukkala, T., Ketonen, T. and Pykäläinen, J. 2003. Predicting timber harvests from private forests—a utility maximisation approach. For Policy Econ 5(3): 285-296. https://doi.org/10.1016/S1389-9341(02)00062-X
Rumelhart, D. E., Hinton, G. E. and Williams, R. J. 1986. Learning representations by back-propagating errors. Nature 323(9): 533-536. http://www.cs.toronto.edu/~hinton/absps/naturebp.pdf
Sierota, Z., Grodzki, W., & Szczepkowski, A. (2019). Abiotic and biotic disturbances affecting forest health in Poland over the past 30 years: Impacts of climate and forest management. Forests, 10(1), 75.
Söderberg, U. 1988. 2, Individual tree models, Functions for forecasting of timber yields. In: Forest Growth Modelling and Prediction Volume 1. Ek, A.R.; Shifley, S.R.; Burk, T.E. (Eds.). North Central Forest Experiment Station, Forest Service, U.S. Department of Agriculture, St. Paul, Minnesota, USA. https://books.google.com.tr/books?hl=tr&lr=&id=EglMK_62M-gC&oi=fnd&pg=PA151&dq=+Functions+for+forecasting+of+timber+yields&ots=svzZNdbvE-&sig=6rI6yQJpm-ER0b9h4BVjFW7JnCI&redir_esc=y#v=onepage&q=Functions%20for%20forecasting%20of%20timber%20yields&f=false Accessed 8 September 2022
Stage, A. R. and Renner, D. L. 1988. 12, Evaluation and validation of models, Comparison of yield-forecasting techniques using long-term stand histories. In: Forest Growth Modelling and Prediction Volume 2. Ek, A.R.; Shifley, S.R.; Burk, T.E. (Eds.). North Central Forest Experiment Station, Forest Service, U.S. Department of Agriculture, St. Paul, Minnesota, USA. https://books.google.com.tr/books?hl=tr&lr=&id=U6Cq5OSAuuQC&oi=fnd&pg=PA810&dq="Comparison+of+production-forecasting+techniques+using+long-term+stand+histories"&ots=Dznf2EfqNs&sig=ubUsPwu-c1VBgIh8_3iaNhMdPtI&redir_esc=y#v=onepage&q&f=false Accessed 8 September 2022
Tippmann, H. F., Schlüter, U., & Collinge, D. B. (2006). Common themes in biotic and abiotic stress signalling in plants. Floriculture, ornamental and plant biotechnology, 52-67.
UCCET (2011). Turkey Forest Products Assembly Sector Report (in Turkish), The Union of Chambers and Commodity Exchanges of Turkey, Sar Matbaası, Ankara, Turkey. https://www.tobb.org.tr/Documents/yayinlar/2012/Orman%20Urunleri%20Rapor_2011.pdf Accessed 8 September 2022
UN. 2011. State of Europe’s forests 2011 status & trends in sustainable forest management in Europe. Forest Europe Liaison Unit, Oslo, Norway. https://www.google.com/url?sa=t&rct=j&q=&
esrc=s&source=web&cd=&ved
=2ahUKEwjM3_m81cbwAhWlMX0KHW3rCMwQFjAAe
gQIBBAD&url=https%3A%2F%2Fwww.unece.org%2
Ffileadmin%2FDAM%2
Fpublications%2Ftimber%2FForest_Europe_report
_2011_web.pdf&usg=AOvVaw2A-cR_gDkRKrnCaud1ZlnC
Accessed 8 September 2022
Vanclay, J. K. 1996. Estimating sustainable timber production from tropical forests (A Discussion Paper prepared for the World Bank). Center for International Forestry Research. https://doi.org/10.17528/cifor/000071
van Lierop, P., Lindquist, E., Sathyapala, S., & Franceschini, G. (2015). Global forest area disturbance from fire, insect pests, diseases and severe weather events. Forest Ecology and Management, 352, 78-88.
Weka 2021. Weka software, http://old-www.cms.waikato.ac.nz/ml/weka/
Wilkaniec, A., Borowiak-Sobkowiak, B., Irzykowska, L., Breś, W., Świerk, D., Pardela, Ł., ... & Wielgus, K. (2021). Biotic and abiotic factors causing the collapse of Robinia pseudoacacia L. veteran trees in urban environments. Plos one, 16(1), e0245398.
Yıldırım, H. T. 2012. Industrial wood production and consumption in Turkey and some future projections. Afr J Bus Manag 6(6): 2261-2266. https://academicjournals.org/article/article1380726067_Yildirim.pdf
Yuan, Z., Ali, A., Wang, S., Gazol, A., Freckleton, R., Wang, X., ... & Loreau, M. (2018). Abiotic and biotic determinants of coarse woody productivity in temperate mixed forests. Science of the Total Environment, 630, 422-431.

Download PDF

Journal Publication

published 16 Jan, 2023

Read the published version in Environmental Science and Pollution Research →

Editorial decision: Major Revision
02 Nov, 2022
Reviewers agreed at journal
07 Oct, 2022
Editor invited by journal
06 Oct, 2022
Editor assigned by journal
29 Sep, 2022
First submitted to journal
26 Sep, 2022

You are reading this latest preprint version

Assessment of the Effects of the Biotic and Abiotic Harm Factors on the Amount of Industrial Wood Production with Deep Learning

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Materials And Methods

2.1. Artificial neural networks

2.2. Multilayer perceptron model

2.3. Study area

2.4. Data and variables

3. Results And Discussions

3.1. Development of the deep learning model

3.2. Assessment of the harm factors

4. Conclusions

Declarations

References

Status:

Journal Publication

Version 1