DEVELOPMENT OF ECOLOGICAL

MANAGEMENT SYSTEM FOR PLANTED

FOREST BASED ON ELM DEEP LEARNING

ALGORITHM

Zhe Wang*

School of Electronic and Electrical Engineering, Shanghai University of

Engineering Science, shanghai, 201620, China

whospurp@163.com

Reception: 18/11/2022 Acceptance: 07/01/2023 Publication: 27/01/2023

Suggested citation:

W., Zhe (2023). Development of ecological management system for

planted forest based on ELM deep learning algorithm. 3C Empresa.

Investigación y pensamiento crítico, 12(1), 165-184. https://doi.org/

10.17993/3cemp.2023.120151.165-184

https://doi.org/10.17993/3cemp.2023.120151.165-184

165

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

ABSTRACT

Plantations play a central and lever role in maintaining the ecological balance of the

earth, maintaining the overall function of the terrestrial ecosystem, and promoting the

coordinated development of economic society and ecological construction. In order to

strengthen the ecological management of plantation forests and improve the

ecological level of forest region, the C/S framework is taken as the basic structure,

and the programming mode of business model-user interface controller is used, on

J2EE platform. The ecological management system of a planted forest is constructed

by the evaluation module, the principal component comprehensive analysis module of

ecological function value and the demand prediction module of planted forest based

on extreme learning machine and deep learning algorithm, and runs under the support

of windows system, oracle 15G and above database software. The indexes and

factors affecting the ecological function of plantation forests were evaluated and

analyzed, and the final management decision was given by the prediction module.

The results showed that the plant density significantly affected plant biomass, organic

carbon storage, water content and nutrient accumulation, and the comprehensive

evaluation indexes of four ecological functions increased from 32.69, 31.84, 33.71 and

35.46 to 86.18, 89.46, 89.83 and 88.76, respectively. Although the degree of influence

of the system on lemon strip plants, herbaceous plants, surface litter and soil varies, it

still has good feasibility, effectiveness and practicality, and can assist the scientific

ecological management of artificial plantation forests.

KEYWORDS

Extreme learning machine; Deep learning algorithm; Plantation forests; Ecological

management system; Principal component analysis

https://doi.org/10.17993/3cemp.2023.120151.165-184

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

166

PAPER INDEX

ABSTRACT

KEYWORDS

1. INTRODUCTION

2. DEVELOPMENT AND DESIGN OF ECOLOGICAL

MANAGEMENT SYSTEM FOR ARTIFICIAL PLANTATION

FOREST

2.1. Evaluation module design

2.2. Design of principal component comprehensive analysis module

2.3. Design of demand prediction module of planted forest based on

extreme learning machine deep learning algorithm

3. APPLICATION OF THE SYSTEM

3.1. Overview of the study area

3.2. Selection, measurement and calculation of ecological function

evaluation index of planted forest

4. RESULTS AND ANALYSIS

4.1. Plant biomass, organic carbon storage and water content of

artificial plantation forest

4.2. Nutrient accumulation in plantation forests

5. DISCUSSION

6. CONCLUSION

7. DATA AVAILABILITY STATEMENT

REFERENCES

9. CONFLICT OF INTEREST

https://doi.org/10.17993/3cemp.2023.120151.165-184

167

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

1. INTRODUCTION

As the main body of the terrestrial ecosystem, the forest has always been the focus

of debate [1-2]. The function of the forest ecological environment can be direct or

indirect. It can be tangible or intangible [3]. The function of forest ecology refers to the

natural environmental conditions and effects that are formed by the forest ecosystem

and ecological process and maintained by human survival [4].

The management of forest ecology is a systematic and complex project. At present,

soil erosion, biodiversity reduction, wetland degradation, land desertification, and

other problems are still serious, and the important role of forests in maintaining

national ecological security has not been fully played [5-8]. With the development of

industrial civilization and the aggravation of global warming, the ecological functions of

forests, such as soil and water conservation, water conservation, carbon fixation and

oxygen release, air purification and environment beautification, have attracted

people's attention [9-10].

Recently, there have been many reports on forest ecological functions, especially

on water conservation, carbon sequestration and oxygen release. Therefore,

managing forest ecology objectively, dynamically and scientifically can deepen

people's environmental awareness, strengthen the leading position of forestry

construction in the ecological environment of the national economy, and improve the

level of forest ecological environment. It is of great practical significance to accelerate

the integration of the environment into the national economic accounting system and

correctly handle the relationship between social-economic development and

ecological environmental protection [11-13].

Based on the above research background, researchers in related fields are trying

to achieve good research results in forest ecological management. Such as literature

[14] according to Ukraine and the EU regulatory filing modern requirements for

environmental protection and biodiversity conservation and the suggestion, proposing

to overcome the ecological problems of radioactive pollution of forest ecosystems

from the perspective of environmental management, and create a fire prevention and

forest management system based on science, to prevent the personnel and the public

excessive exposure from various sources, Prevention of secondary diffusion of

radionuclides to relatively clean areas by fire is achieved through the use of

hydrodynamic active water extinguishers and the laying of a polyethylene guanidine

based barrier in front of the fire line. Literature [15] used mixed methods to conduct

ecological management of tropical dry forests on 11079 hectares of land in Colombia.

With the participation of 64 experts, the Delphi method was applied to conduct

quantitative research on the ecological situation from 2018 to 2020. The results

showed that all knowledge management practices identified had a certain impact on

ecological management. It also contributes to the generation, transformation, and

mobilization of scientific knowledge on each component of the ecological restoration

process of tropical dry forests. Literature [16] aims at how to realize the maximization

of the value of forest ecological resources to provide theoretical reference to the

sustainable development of the global forest resources, using the theory of ecological

capital is discussed how to define the concept of forest ecological resources

https://doi.org/10.17993/3cemp.2023.120151.165-184

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

168

capitalization, analyzed from two aspects of forest ecological resources capitalization

of inner motivation, and how people change the use of forest resources, the results

show that the realization path of capitalization of forest ecological resources can

provide theoretical reference for maximizing the value of forest ecological resources

and realizing the sustainable development of global forest resources. Literature [17] in

30 provinces as a sample, constructed the index system of forest ecological security

and the efficiency of forest management, with the help of the CCR model, the coupling

coordination model and the spatial panel model, from 2003 to 2017 China's provincial

forest ecological safety and the management efficiency of the forestry coupling

coordination degree and its temporal pattern characteristics and influence factors in

the measurement and analysis, The results show that, in terms of forest ecological

security, the index increases as a whole, and the coupling coordination degree

changes from near uncoordinated to intermediate coordinated. This study can provide

a theoretical reference framework for China's forest ecological management

decisions. According to the consensus on the ecological environment in previous

studies and the characteristics of the study area. Literature [18] established a

quantitative evaluation index system for the comprehensive ecological environment of

forest ecosystem nature reserve based on water, air, soil and biological environment,

constructed a weightless cloud model, and provided a weightless evaluation

mechanism. The results showed that the results of this study can provide theoretical

support for the evaluation of forest ecosystem nature reserves and general evaluation

when the weight is difficult to determine or uncertain. Literature [19] in human forest

botanical garden as the research object, from the regional environmental quality,

development conditions and the scenic area characteristic value of three standard

level selected 23 indicators, to evaluate the ecological tourism development potential,

using the analytic hierarchy process (AHP) to determine the weight of each index, and

using the fuzzy comprehensive evaluation method to three standard level and

evaluate the ecological tourism development potential, The conclusion suggested that

scenic spots should maintain their advantages, enhance tourism characteristics,

strengthen ecological civilization construction, and promote the in-depth development

of ecological tourism. The above research results have established a relatively

comprehensive and applicable forest ecological evaluation index system from different

emphases, which has a certain promoting effect on improving the forest ecological

environment and enhancing the ecological level.

Human initial afforestation is mainly in order to production needs, as a part of the

agricultural production, with the progress of social productivity, timber demand

increasing, under the natural state of forest resource components are insufficient to

satisfy the human production and life, people began planting plantation, planted forest

ecological level and function is becoming more and more attention [20]. Therefore,

this article is based on C/S architecture and the J2EE platform, in the Struts

framework with the business model - user interface - controller programming model,

under the support of construct artificial planting forest ecological management system,

through the synergy evaluation module, the ecological function value of the principal

component comprehensive analysis module and the depth of the extreme learning

machine learning algorithm of planted forests demand forecast module, Various

https://doi.org/10.17993/3cemp.2023.120151.165-184

169

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

indicators and factors affecting the ecological function of planted forests were

evaluated and analyzed in order to better understand the biodiversity of planted

forests and its regulation approaches and mechanisms on ecological function and to

provide scientific basis and technical means for guiding the ecological management

and sustainable management of planted forests [21].

2. DEVELOPMENT AND DESIGN OF ECOLOGICAL

MANAGEMENT SYSTEM FOR ARTIFICIAL

PLANTATION FOREST

Based on the C/S framework and J2EE platform, the business model -user

interface-controller programming mode of the struts framework is adopted and design

the environmental management system for artificial planting forests as shown in

Figure 1. The system includes an evaluation module, principal component

comprehensive analysis module of ecological function value and planting forest

demand prediction module of extreme learning machine deep learning algorithm. The

three modules work together to evaluate and analyze the indicators and factors that

affect the ecological function of planted forests and provide management and support

for decision-makers. The evaluation unit mainly includes three units: single ecological

function value evaluation of all planted forest species, total ecological function value

evaluation of single planted forest species, and total ecological function value

evaluation of all planted forest species. The system server uses windows system,

oracle 15G or above database, Tomcat6.0.67 or above server, JDK2.0 or above

development package; The client uses IE 9.0 or higher.

Figure 1. architecture diagram of the environmental management system of artificial plantation

forest

Ecological management system

for artificial planting forests

Evaluation module Principal Component

Synthesis Analysis Module

Planting demand forecast

module

Principal

compone

analysis

strategies

Demand

predictio

n model

based on

extreme

learning

machine

deep

learning

algorithm

Evaluatio

n unit of

total

ecologica

l function

value of

all

planted

forest

species

Unit for

rating the

total

ecologica

l function

value of

forest

species in

a single

planted

forest

A single

ecologica

l function

value

evaluatio

n unit for

all

planted

forest

species

https://doi.org/10.17993/3cemp.2023.120151.165-184

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

170

2.1. EVALUATION MODULE DESIGN

The ecological function value evaluation algorithm of the four units of the evaluation

module is described as follows:

(1) Single ecological function value evaluation unit of all planted forest species: this

unit includes the ecological function of water conservation, soil conservation, and

carbon dioxide fixation of planted forest species. The value evaluation formulas of

different ecological functions are as follows:

(1)

Where, is the ecological function value of water conservation of dominant tree

species in artificial plantation forests.

(2)

Where, is the ecological function value of soil conservation of dominant tree

species in artificial plantation forests.

(3)

Where, is the ecological function value of fixed carbon dioxide of dominant

tree species in artificial plantation forests.

(2) Total ecological function value evaluation of monoculture forest species: this unit

contains the total ecological function of different dominant tree species, and the value

evaluation formula is as follows:

(4)

Where, is the corresponding ecological function value of oxygen release of

dominant tree species in the planted forest. is the ecological function value of

biological storage energy of each dominant tree species in the planted forest; is

the ecological function value of biodiversity of each dominant tree species in the

plantation.

(3) Total ecological function value evaluation of all planted forests and species: the

total ecological function value evaluation algorithm of this unit is as follows:

(5)

Evaluation module by running the business model - user interface - controller

programming software, all sorts of forest of planted forest respectively calculated

single, single planting forest ecological function value of total sorts of forest ecological

function value and all sorts of forest of planted forest ecological function value of three

indicators, namely the complete management system of planted forest ecological

function value assessment, the evaluation function of ecological management system

of artificial plantation forest was realized.

https://doi.org/10.17993/3cemp.2023.120151.165-184

1,2,...,

hysy i

i n

M V

=∑

1,2,...,

bytr i

i n

M T

=∑

1,2,...,

co i

i n

M Q

∑

1 2dy i i i i i i

M V T Q Q E K= + + + + +

1 2

1,2,..., 1,2,..., 1,2,..., 1,2,..., 1,2,..., 1,2,...,

z i i i i i i

i n i n i n i n i n i n

M V T Q Q E K

= = = = = =

= + + + + +

∑ ∑ ∑ ∑ ∑ ∑

171

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

2.2. DESIGN OF PRINCIPAL COMPONENT COMPREHENSIVE

ANALYSIS MODULE

The principal component analysis method [22] was adopted to comprehensively

and systematically understand the comprehensive strength of the ecological function

value of different planted forest types, and many indicators reflecting the ecological

function value characteristics of planted forest types were considered from different

aspects. The program flow chart of the principal component analysis algorithm is

shown in Figure 2. The specific description is as follows:

(1) Data standardization of ecological value evaluation index;

(2) The correlation between ecological value evaluation indicators;

(3) Calculate the covariance matrix of standardized data;

(4) Calculate all the eigenvalues of the covariance matrix, calculate the

eigenvectors, determine the number of principal components according to the

cumulative ratio of eigenvalues;

(5) Calculate principal component load value and factor score coefficient matrix to

determine the principal component expression; To calculate the comprehensive score

is to carry out the principal component score, sort the scores according to the size of

the score value, and output the most important first several ecological value

evaluation indicators.

With the support of Windows system, Oracle 15G or above database,

Tomcat6.0.67 or above server, JDK2.0 or above development package and other

software, the algorithm program of principal component analysis can run according to

the process shown in Figure 2. Principal component scoring is carried out according

to the obtained comprehensive score. Several important evaluation indexes of

ecological value were obtained, and the analysis function of ecological management

system of artificial plantation forest was realized.

https://doi.org/10.17993/3cemp.2023.120151.165-184

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

172

Figure 2. program flow chart of principal component analysis algorithm

2.3. DESIGN OF DEMAND PREDICTION MODULE OF PLANTED

FOREST BASED ON EXTREME LEARNING MACHINE DEEP

LEARNING ALGORITHM

The prediction module is mainly used for quantitative prediction and analysis of the

demand for artificial plantation forests. The change in plantation density is influenced

by the social economy, population, natural environment and other factors [23]. So

choose gross GDP ($one hundred million), at the end of the total population (ten

thousand people), non-agricultural population (ten thousand people), fiscal revenue

(one hundred million yuan), financial expenditure (one hundred million yuan), the first

industry (one hundred million yuan), the second industry (one hundred million yuan),

the third industry (one hundred million yuan), the output value of industrial output

value ($one hundred million), construction (one hundred million yuan), the per capita

net income (yuan), fixed asset investment ($one hundred million), Urban input (100

million yuan), rural input (100 million yuan), annual precipitation (100 million cubic

meters) and other indicators constitute the driving index system of the change of

artificial planting forest density (10 thousand trees/ha). There is no simple linear

relationship between each driving factor and the density of the planted forest, and

there is a correlation in time. Therefore, an extreme learning machine deep learning

Begin

Find the covariance array and the feature root

Computes feature vectors

The cumulative contribution rate is calculated,

and the number of principal components is given

Calculate the principal component score

and arrange it by size to give the result

End

https://doi.org/10.17993/3cemp.2023.120151.165-184

173

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

algorithm is adopted to build the demand time series prediction model with the nearest

neighbor domain theory [24] as the core, as shown in Figure 3. The prediction process

is described as follows:

(1) Extraction of training samples: in order to complete the prediction of planting

forest density at time , samples most similar to the prediction sequence

of should be extracted from all training samples as the recombination samples,

and recombination samples should be used as the input of the local model. The

process of finding the nearest neighbor is the process of measuring the similarity

between the prediction sequence and all the training samples. Through this step,

the best prediction model of planting forest density can be built.

In order to select nearest neighbors of from set as recombination samples,

a method is needed to calculate the similarity of two sets of sequences. Assume that

and are two time series:

(6)

and are difference sequences of two time series and respectively:

(7)

Assuming that is the standardized Euclidean distance between and

[25], is the standardized Euclidean distance between and , we

use the mixed Euclidean distance to calculate the similarity of the two groups of time

series:

(8)

is the mixed Euclidean distance of and . We calculated the mixed

Euclidean distance of each element in the set and , finally obtained mixed

Euclidean distances, and selected elements with the shortest distance as the

extracted training samples.

(2) Prediction model derivation of extreme learning machine deep learning

algorithm: the deep learning algorithm of extreme learning machine integrates the

idea of self-coding [26] and encodes the output by minimizing reconstruction error so

that the output can approach the original input infinitely. This structure provides an

abstract representation of the input and thus captures the deep features of the original

input. Figure 3 describes the prediction modeling process of the algorithm for the

output planting forest density at time . The extracted training sample is

t s

Y+

t s+

t s

Y+

( )

1 2

N A A

( )

1 2

N F F

( )

1 2

N A A

{ }

t s

Y+

t s+

{ }

https://doi.org/10.17993/3cemp.2023.120151.165-184

1 1 1 1 1

2 1 1 1 1

, , , , , , ,

a a a a a q a q a q a q

b b b b b q b q b q b q

A X Y X Y Y X Y

+ + + −+−+ +

⎧⎡ ⎤

⎪ ⎣ ⎦

⎨⎡ ⎤

⎪⎣ ⎦

⎩

…

, X

…

, X

1 1 1 1 1

2 1 1 1 1

, , ,

a a a a a q a q a q a q

b b b b b q b q b q b q

F X X Y Y X Y Y

+ + + + −+ + −

⎧⎡ ⎤

=− − − −

⎪ ⎣ ⎦

⎨⎡ ⎤

=− − − −

⎪⎣ ⎦

⎩

…

, X

…

, X

( ) ( ) ( )

1 2 1 2

1 2

E E

N A A N F F

N A A +

=，

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

174

taken as the input of the network. It is assumed that the network is composed of the

hidden layer of layer, and

represents the weight parameters to

be learned in the network.

Figure 3. architecture diagram of a prediction model based on extreme learning machine deep

learning algorithm

Each layer in the network can be decoupled as an independent extreme learning

machine [27], and the target output of each extreme learning machine is equal to the

input of the extreme learning machine. In this way, the low-dimensional representation

of the input data can be obtained, that is, the hidden layer output of the extreme

learning machine, and this output can be used as the input of the next extreme

learning machine. The output weight

of the extreme learning machine is calculated

by the following formula:

(9)

Where, is the rule item;

is the output matrix of the hidden layer of an extreme

learning machine.

Then the weight parameter is calculated by the following formula:

(10)

{ }

1 2 1

, ,

W W W W

=…

∧

( )

W k =…

, h

https://doi.org/10.17993/3cemp.2023.120151.165-184

T T

H H H Tβλ

−

∧⎛ ⎞

= +

⎜ ⎟

⎝ ⎠

( )

W kβ

∧

= = …

, h

175

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

Finally, the feature expression is obtained, that is, the output of the hidden layer of

layer , as the hidden layer of an independent extreme learning machine, and the

output weight of the extreme learning machine is obtained by similar calculation.

The weight parameter of the network is calculated by the following formula:

(11)

According to the obtained network weight parameter , the

prediction model architecture of extreme learning machine deep learning algorithm is

defined. After the training sample data is input into the input layer of the prediction

model, layer by layer calculation is carried out according to the determined network

structure, and of planted forest density can be calculated, which provides decision

basis for ecological management of planted forest.

3. APPLICATION OF THE SYSTEM

3.1. OVERVIEW OF THE STUDY AREA

An artificial plantation forest area with an area of about 2,000 square kilometers

and a planting density of 9.64 million trees per hectare in 2017 was selected as the

research target. The research area is located at 110°13 '~111°35' east longitude and

36°46 '~37 °52' north latitude, 983-1684m altitude and mountainous area. It is a

typical hilly and gully region with complex terrain and closed traffic. The main type of

soil is the yellow soil developed from the parent material of loess, with fine particles

and deep soft soil, which is conducive to farming. The main vegetation types include

agricultural land crops, grasslands, and shrubland. The main grassland species are

alfalfa, long miscanthus, thyme, ice grass, iron pole, pig hair down, Chinese

asparagus herb, two cracks, and so on. Shrub species mainly include Salix

psammophila, peach, apricot, caragana korshinskii. Caragana korshinskii is widely

planted in this region because of its drought resistance, cold resistance, sand

resistance and good soil and water conservation. There are large typical experimental

and demonstration areas of the caragana korshinskii plantation. The system was put

into use in 2018. According to the prediction model of extreme learning machine deep

learning algorithm, the reasonable planting density of artificial planting forest in this

region is 23.24 million trees/ha. Caragana korshinskii will be planted in 2021.

3.2. SELECTION, MEASUREMENT AND CALCULATION OF

ECOLOGICAL FUNCTION EVALUATION INDEX OF

PLANTED FOREST

In order to verify the ecological management effect of the system on the region, the

ecological function of the forest region is evaluated from four aspects [28-31]: plant

biomass, organic carbon storage, nutrient accumulation, and water retention capacity.

∧

W+

{ }

1 2 1

, ,

W W W W

=…

t s

Y+

https://doi.org/10.17993/3cemp.2023.120151.165-184

Wβ

∧

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

176

The evaluation system and the measurement method of secondary indicators are

shown in Table 1. The first-level indicators include (1) plant biomass: caragana

korshinskii and herbaceous aboveground biomass and litter biomass; (2) organic

carbon storage: the aboveground and ground litter of caragana korshinskii and

herbaceous plants and soil organic carbon storage; (3) total nitrogen and total

phosphorus reserves of caragana korshinskii plants and herbaceous plants

aboveground and litters; soil nutrient index: soil total nitrogen, nitrate nitrogen,

ammonium nitrogen, and available phosphorus contents; (4) water content of

caragana korshinskii plants and herbaceous plants aboveground, litter and soil.

Table 1. evaluation system and component determination method

The contents of different components were determined, and the first-order index of

ecological function was measured using average value method[32-33]. Assuming that

the actual measured value of the secondary index of sample plot is , and the

mean value and standard deviation of the secondary index among all sample plots

of the same factor are and respectively, the calculation formula for the score of

the secondary index of sample plot is as follows:

(12)

Thus, the ecological functional comprehensive index of each level index can be

deduced:

(13)

Where, is the number of all second-level indicators contained in plot .

4. RESULTS AND ANALYSIS

Primary indicators secondary indicators Determination method

biomass Aboveground biomass Model estimation method

Aboveground and litter biomass Weighing method

Organic carbon storage Aboveground and surface litter and soil

organic carbon storage

The external heating method

of potassium dichromate

oxidation

Accumulation of nutrients

Total nitrogen storage in aboveground and

litter

Concentrated sulfuric acid -

hydrogen peroxide digestion

Total phosphorus storage in aboveground

and litter

Vanadium molybdenum

yellow colorimetry

Contents of total nitrogen, nitrate nitrogen

and ammonium nitrogen in soil

Semi-trace Kjeldahl nitrogen

determination

Soil available phosphorus content

Sodium bicarbonate extraction

- molybdenum antimony

resistance colorimetric method

Water retention Water content of aboveground and litter Weighing method

Soil moisture content

Xij

jσ

https://doi.org/10.17993/3cemp.2023.120151.165-184

( )

Zij Xij uj jσ=−/

EMFi Zij f= /

∑

177

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

4.1. PLANT BIOMASS, ORGANIC CARBON STORAGE AND

WATER CONTENT OF ARTIFICIAL PLANTATION FOREST

The aboveground biomass of caragana korshinskii, herbaceous plants and litters

were selected as the parameters to evaluate the plant biomass of the planted forest.

Figure 4 shows that plant biomass is significantly affected by planting density of

planted forests. Before the application of ecological management system, the

aboveground biomass and litter biomass of caragana korshinskii and herbage were

only 8.19t/hm2, 6.47t/hm2 and 5.84t/hm2

, respectively. After applying the system for

ecological management, through comprehensive evaluation of the ecological function

of the sample land, they are 10.67t/hm2, 2.51t/hm2, 1.34t/hm2, 12.55t/hm2,

respectively. After the application of ecological management system, organic carbon

storage increased significantly, reaching 30.43t/hm2

in 2021, the aboveground

biomass and litter biomass of herbage increased, and the aboveground biomass and

litter biomass of herbage increased, reaching 40.64t/hm2, 8.76t/hm2 and 7.66t/hm2

2021, respectively. The aboveground organic carbon storage of caragana korshinskii

and herbage, litter organic carbon storage and soil organic carbon storage were

selected as parameters to evaluate the ecological function of organic carbon storage.

Caragana korshinskii, herbage, litter and soil organic carbon storage were 7.19t/hm2,

6.48t/hm2, 26.71t/hm2, respectively. Water retention is an important ecological function

of planted forests. In this study, four indexes of caragana korshinskii plants,

herbaceous plants, ground litters and soil water content were selected as parameters

to evaluate the water retention capacity of caragana korshinskii plantations. As shown

in figure 4, you can see that plantation planting density of caragana korshinskii and

plant litter moisture content is less affected, there was no significant difference before

and after the application in the system, but the herbs and soil moisture content in

2017 and 2021 significant difference under different planting density and planting

density increase, herbaceous plants and soil water content is increased greatly, in

2021, they will reach 62.58% and 59.43% respectively. To sum up, the environmental

management system is based on the results of the evaluation of the single ecological

function value of all planted forests, the total ecological function value of all planted

forests and the total ecological function value of all planted forests. By reasonably

adjusting the planting density in the forest region, the coverage rate of surface

vegetation was expanded, and the ecological function indexes were improved. The

comprehensive indexes of plant biomass, organic carbon storage and water content

of the planted forest increased from 32.69, 31.84 and 35.46 in 2017, respectively.

Increasing to 86.18, 89.46 and 88.76 in 2021, there is an obvious synergistic effect

between planting density and ecological function in the forest region, and artificial

planting forest has a better ability to provide and maintain multiple ecological functions

simultaneously.

https://doi.org/10.17993/3cemp.2023.120151.165-184

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

178

Figure 4. schematic diagram of ecological function change of plantation forest

4.2. NUTRIENT ACCUMULATION IN PLANTATION FORESTS

Total nitrogen and total phosphorus reserves of caragana korshinskii and

herbaceous plants, total nitrogen and total phosphorus reserves of litters, and soil

nutrient content indexes (soil total nitrogen, available phosphorus, nitrate-nitrogen,

ammonium nitrogen) were selected as the parameters to evaluate the ecological

function of nutrient accumulation in planting areas. Figure 5 (a) shows that the

planting density of caragana korshinskii significantly affected the plant total nitrogen

storage of the planted forest. The total nitrogen storage of caragana korshinskii was

significantly higher than that of herbaceous plants and litters. Before the system was

put into use, the planting density was small, and the total nitrogen reserves of

caragana korshinskii, herbaria korshinskii and litters were low, only 135.46kg/hm2,

11.13kg/hm2 and 8.54kg/hm2

. However, after the system was put into use, the total

nitrogen reserves of three kinds of plants were greatly increased. In 2021, the

ecological function of total nitrogen accumulation was 763.08kg/hm2, 25.66kg/hm2

and 51.73kg/hm2

, respectively. The ecological function of caragana korshinskii was

the most obvious before and after application of the system. Compared with total

nitrogen, the total phosphorus reserves of plants in the planted forests with different

planting densities changed more gently, but the functional degree of plants was still

that the total phosphorus reserves of caragana korshinskii were significantly higher

than that of herbaceous plants and litters. Before the use of systematic management

of forest ecology, the total phosphorus reserves of the three are 8.46kg/hm2, 1.23kg/

hm2 and 0.64kg/hm2

, respectively. After management, total phosphorus reserves

reach 36.16kg/hm2, 1.43kg/hm2 and 2.22kg/hm2

, respectively. There was no

significant difference in total phosphorus storage of herbaceous plants and litters

under different planting densities.

It can be seen from Figure 5 (b) that with the increase in plantation density, soil total

nitrogen content also showed a corresponding upward trend, but the increase was not

2017 2021

Plant biomass

t/hm

2017 2021

Moisture content

2017 2021

Organic carbon storage

t/hm

Caragana microphylla

Herbal

Litter

Soil

https://doi.org/10.17993/3cemp.2023.120151.165-184

179

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

obvious, only increased by 0.08g/kg. Therefore, the application of an ecological

management system had no significant effect on soil total nitrogen content. Before the

application of the system, the contents of soil available phosphorus, nitrate-nitrogen, and

ammonium nitrogen content were only 1.03mg/kg, 3.11mg/kg and 0.37mg/kg. When the

system gave a reasonable planting density and implemented it, the contents of the three

nutrients in the soil were as high as 13.56mg/kg, 18.38mg/kg and 10.84mg/kg,

respectively. This is because, with the support of a principal component analysis strategy

and extreme learning machine deep learning algorithm, the management system

reasonably increases the number of caragana korshinskii plants per unit area, and makes

the biomass of caragana korshinskii and litters at a high level. Therefore, it directly

promotes the increase of total nitrogen and total phosphorus reserves of caragana

korshinskii plants and litters. In addition, by increasing the surface vegetation coverage

and improving the input of soil resources, the nutrient contents of soil such as nitrogen and

phosphorus were increased, and the nutrient accumulation ecological functional

composite index of the cultivated forest increased from 33.71 in 2017 to 89.83 in 2021.

(a) plant and litter nutrient content

(b) soil nutrient content

Figure 5. schematic diagram of nutrient accumulation

2017 2021

140

800

Total nitrogen reserves

kg/hm

Caragana microphylla

Herbal

Litter

2017 2021

Total phosphorus reserves

kg/hm

2017 2021

Total nitrogen g/kg

Rapid available phosphorus mg/kg

Nitrate nitrogen mg/kg

Ammonium nitrogen mg/kg

https://doi.org/10.17993/3cemp.2023.120151.165-184

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

180

5. DISCUSSION

This study focuses on the study of the environmental management system of

artificial planting forests. The data input into the system, through the function

evaluation and prediction, helps to achieve the best combination of forest species

selection, planting area and ecological environmental benefits. The next step for

further in-depth research is as follows:

(1) Combined with 3S technology, global positioning system is used for real-time

positioning, remote sensing for data collection and update, geographic information

system for spatial analysis and comprehensive processing, so as to rapidly update

ecological related data and reduce the cost of manpower and material resources;

(2) For the model base of this study, optimization model and early warning model

need to be added in the future. From the perspective of ecological economics, it is

expected to think about how to arrange the species structure of planted forest and

analyze the optimization between its related functions and generated services.

Forewarning of the loss of ecological function of plantation forest;

(3) For the system, there is currently a lack of expert experience and practical

knowledge, which needs to be enriched and strengthened in the future. In addition,

artificial intelligence technologies such as neural networks, support vector machines

and time series analysis can also be used to acquire knowledge in the system.

6. CONCLUSION

Afforestation is one of the main ways of ecological restoration and land

reclamation. Whether the restoration mode is suitable or not needs long-term

observation to see its restoration effect. The ecological health of the planted forest

also varies with the recovery time. Therefore, this paper takes C/S as the basic

framework and J2EE as the development platform. With the support of principal

component analysis method and extreme learning machine deep learning algorithm,

the Struts framework and business model-user-interface controller programming

mode are adopted in this paper. An environmental management system consisting of

an evaluation module, principal component comprehensive analysis module and

planting demand prediction module has been developed. Through carrying out

experimental research activities on the effect of ecological management of forest

region by the system in the study area, the following three conclusions have been

obtained:

(1) Single system USES all sorts of the forest of planted forest ecological function

value evaluation, the single planting Lin total sorts of forest ecological function value

evaluation, all sorts of the forest of planted forest ecological function value evaluation

three units constructing evaluation module, to provide better practice ecological

management database and decision basis, and planted forest density and obvious

synergistic effect between different ecological functions.

(2) Using principal component analysis comprehensive system understanding of

different sorts of the forest of planted forest ecological function value of

https://doi.org/10.17993/3cemp.2023.120151.165-184

181

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

comprehensive strength, help considering from different sides reflects growing forest

category characteristics of many types of ecological function value indicators, the

artificial planting forests can have better ability to provide and maintain multiple

ecological functions at the same time.

(3) Based on the extreme learning machine deep learning algorithm, the demand

prediction module of planted forest designed by extreme learning machine takes the

nearest neighbor domain theory as the core, making the output infinitely close to the

original input, and giving the reasonable planting density of planted forest in this

region as 23.24 million trees/ha. The biomass, organic carbon storage, nutrient

accumulation and water content in the study area were increased to 57.06t/hm2,

70.81t/hm2, 924.38kg/hm2 and 216.03% from 20.5t/hm2, 27.07t/hm2, 171.21kg/hm2

and 130.35%. The corresponding ecological functional composite index increased

from 32.69, 31.84, 33.71, 35.46 to 86.18, 89.46, 89.83, 88.76, respectively.

7. DATA AVAILABILITY STATEMENT

The original contributions presented in the study are included in the article/

supplementary material, further inquiries can be directed to the corresponding author.

REFERENCES

(1) Vinnepand M, Fischer P, Zeeden C, et al.

Decoding geochemical signals of

the Schwalbenberg Loess- Palaeosol-Sequences -A key to Upper

Pleistocene terrestrial ecosystem responses in western Central Europe.

2021.

(2) Wang H H, Chu H L, Dou Q, et al. Seasonal Changes in Pinus tabuliformis

Root-Associated Fungal Microbiota Drive N and P Cycling in Terrestrial

Ecosystem[J]. Frontiers in Microbiology, 2021, 11:3352-.

(3) Le S, Wu Y, Guo Y, et al. Game Theoretic Approach for a service function

chain routing in NFV with coupled constraints[J].

Circuits and Systems II:

Express Briefs, IEEE Transactions on, 2021, PP(99):1-1.

(4) Li, H., Deng, J., Feng, P., Pu, C., Arachchige, D., and Cheng, Q., (2021) Short-

Term Nacelle Orientation Forecasting Using Bilinear Transformation and

ICEEMDAN Framework. Front. Energy Res. 9, 780928.

(5) Z Németová, S Kohnová. Mathematical modeling of soil erosion processes

using a physically-based and empirical models: Case study of Slovakia

and central Poland[C]. Adaptive Hardware and Systems. Central Library of the

Slovak Academy of Sciences, 2021.

(6) Wu Y, Guo Y, Toyoda M.

Policy Iteration Approach to the Infinite Horizon

Average Optimal Control of Probabilistic Boolean Networks[J]. IEEE

Transactions on Neural Networks and Learning Systems, (99):1-15.

(7) Zhan Q, Zhao W, Yang M, et al.

A long-term record (1995-2019) of the

dynamics of land desertification in the middle reaches of Yarlung Zangbo

River basin derived from Landsat data[J]. Geography and Sustainability,

2021, 2(1).

https://doi.org/10.17993/3cemp.2023.120151.165-184

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

182

(8) Zhang Y, Qian T, Tang W.

Buildings-to-distribution-network integration

considering power transformer loading capability and distribution network

reconfiguration[J]. Energy, 2022, 244.

(9) Chen L. Study on carbon fixation and oxygen release ability of urban

greening tree species based on spatial and temporal dynamic analysis[J].

International Journal of Global Energy Issues, 2020, 42(3/4):244.

(10) Li, H., Deng, J., Yuan, S., Feng, P., and Arachchige, D., (2021) Monitoring and

Identifying Wind Turbine Generator Bearing Faults using Deep Belief

Network and EWMA Control Charts. Front. Energy Res. 9, 799039.

(11) Hahn R G. Renal water conservation and the volume kinetics of fluid

induced diuresis; a retrospective analysis of two cohorts of elderly men[J].

Clinical and Experimental Pharmacology and Physiology, 2020.

(12) Heinzelmann G, Sontheim P, Haas A, et al. Hand-Portable Garden, Forestry

and/or Construction Processing Device and Method for Operating a Hand-

Portable Garden, US20210291400A1[P]. 2021.

(13) Wang Y, Zhang D, Wang Y. Evaluation Analysis of Forest Ecological Security

in 11 Provinces (Cities) of the Yangtze River Economic Belt[J].

Sustainability, 2021, 13.

(14) Maglyovana T, Dolin V. KEY ISSUES FOR ECOLOGICAL MANAGEMENT OF

RADIOACTIVE CONTAMINATED FOREST ECOSYSTEMS IN UKRAINE[J].

2020.

(15) Torres-Romero F, Acosta-Prado J C. Knowledge Management Practices and

Ecological Restoration of the Tropical Dry Forest in Colombia[J]. Land,

2022, 11.

(16) Forum N R. The concept delimitation, the value realization process, and the

realization path of the capitalization of forest ecological resources. 2021.

(17) Chen N, Qin F, Zhai Y, et al. Evaluation of coordinated development of

forestry management efficiency and forest ecological security: A

spatiotemporal empirical study based on China's provinces[J]. Journal of

Cleaner Production, 2020:121042.

(18) Xiang M, Lin X, Yang X, et al. Ecological Environment Evaluation of Forest

Ecosystem Nature Reserves Using an Unweighted Cloud Model[J]. Water,

2020, 12(7):1905.

(19) Yin C, Pan W.

Study on the Evaluation System of Forest Ecological

Protection Using Big Data Technology and Adaptive Fuzzy Logic

System[J]. Journal of Physics: Conference Series, 2021, 1952(3):032030 (6pp).

(20) Jani H, Meder R, Hamid H A, et al. Near infrared spectroscopy of plantation

forest soil nutrients in Sabah, Malaysia, and the potential for microsite

assessment[J]. Journal of Near Infrared Spectroscopy, 2021, 29(3):148-157.

(21) Mitsuru, Toyoda, Yuhu. Mayer-Type Optimal Control of Probabilistic Boolean

Control Network With Uncertain Selection Probabilities[J]. IEEE

transactions on cybernetics, 2019.

(22) Lu C, Feng J, Chen Y, et al. Tensor Robust Principal Component Analysis

with A New Tensor Nuclear Norm[J]. IEEE Transactions on Pattern Analysis

and Machine Intelligence, 2020, 42(4):925-938.

https://doi.org/10.17993/3cemp.2023.120151.165-184

183

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

(23) Thiffault N, Hoepting M K, Fera J, et al. Managing plantation density through

initial spacing and commercial thinning: yield results from a 60-year-old

red pine spacing trial experiment[J]. Canadian Science Publishing, 2021(2).

(24) Tang S, Yang Y, Ma Z, et al. Nearest Neighborhood-Based Deep Clustering

for Source Data-absent Unsupervised Domain Adaptation[J]. 2021.

(25) Ding Y, Liang A, Ma K, et al.

Research on Optimal Strategy of Residential

Buildings Energy Based on Standardized Euclidean Distance Measure

Similarity Search Method[J]. IOP Conference Series Earth and Environmental

Science, 2021, 651(2):022052.

(26) Cai J, Dai X, Hong L, et al. An Air Quality Prediction Model Based on a Noise

Reduction Self-Coding Deep Network[J].

Mathematical Problems in

Engineering, 2020, 2020(3):1-12.

(27) Li B, Chen H, Tan T. PV Cell Parameter Extraction Using Data Prediction–

Based Meta-Heuristic Algorithm via Extreme Learning Machine[J]. Frontiers

in Energy Research, 2021, 9:693252.

(28) Ingraffia R, Amato G, Iovino M, et al.

Polyester microplastic fibers in soil

increase nitrogen loss via leaching and decrease plant biomass

production and N uptake. 2022.

(29) Liu Z, Huang F, Wang B, et al.

Impacts of mulching measures on crop

production and soil organic carbon storage in a rainfed farmland area

under future climate[J]. Field Crops Research, 2021, 273:108303-.

(30) KY Gülüt, Duymu E, Solmaz L et al. Nitrogen and boron nutrition in grafted

watermelon II: Impact on nutrient accumulation in fruit rind and flesh[J].

PLoS ONE, 2021, 16(5):e0252437.

(31) Cla B, Wl A, Bl A, et al. Comprehensive analysis of ozone water rinsing on

the water-holding capacity of grass carp surimi gel[J]. LWT, 2021, 150.

(32) Frayssinet, M., Esenarro, D., Juárez, F. F., y Díaz, M. (2021). Methodology

based on the NIST cybersecurity framework as a proposal for

cybersecurity management in government organizations.

3C TIC.

Cuadernos de desarrollo aplicados a las TIC, 10(2), 123-141. https://doi.org/

10.17993/3ctic.2021.102.123-141

(33) Shen Siqi.(2021). Multi-attribute decision-making methods based on normal

random variables in supply chain risk management.

Applied Mathematics

and Nonlinear Sciences (1). https://doi.org/10.2478/AMNS.2021.2.00147.

9. CONFLICT OF INTEREST

The authors declare that the research was conducted in the absence of any

commercial or financial relationships that could be construed as a potential conflict of

interest.

https://doi.org/10.17993/3cemp.2023.120151.165-184

3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376

Ed. 51 Iss.12 N.1 January - March, 2023

184