Modeling of tree recruitment by artificial neural networks after wood harvesting in a forest in eastern Amazon rain forest

Reis, Leonardo Pequeno; Souza, Agostinho Lopes de; Reis, Pamella Carolline Marques dos Reis; Mazzei, Lucas; Leite, Helio Garcia; Soares, Carlos Pedro Boechat; Torres, Carlos Moreira Miquelino Eleto; Silva, Liniker Fernandes da; Ruschel, Ademir Roberto; Rêgo, Lyvia Julienne Sousa

doi:10.5902/1980509825808

Acessibilidade / Reportar erro

Brasil

Español English

sumário « anterior atual seguinte »

Sumário

Artigos • Ciênc. Florest. 29 (2) • Apr-Jun 2019 • https://doi.org/10.5902/1980509825808 copy

Modeling of tree recruitment by artificial neural networks after wood harvesting in a forest in eastern Amazon rain forest

Modelagem do recrutamento de árvores por redes neurais artificiais após a colheita de madeiras em floresta no leste da Amazônia

Authorship SCIMAGO INSTITUTIONS RANKINGS

Abstract

Recruitment models in tropical forests are important for studies on forest management sustainability because they provide adequate support to recovery of wood stocks. The objective of this work was to estimate recruitment after wood harvesting by using an artificial neural network (ANN) model. The study area is located at Tapajós National Forest (55° 00’ W, 2° 45’ S), Pará state. In 64 ha of the study area, in 1979, an intensive harvest of 72.5 m³ ha^-1 was carried out. In 1981, 36 permanent plots of 50 m x 50 m were randomly installed. These plots were measured in 1982, 1983, 1985, 1987, 1992, 1997, 2007, 2010 and 2012. For recruitment modeling, the variables of the target subplot and its vicinity were considered. The estimates obtained in ANN training and generalization were evaluated by statistics: correlation () and root mean square error (RMSE) were determined: RMSE 35.6% and 0.89. Recruitment tendency could be modeled over time in tropical forests after wood harvesting.

Keywords:
Ingrowth; Artificial intelligence; Forest management

Table 1 Variables used for training artificial neural networks (ANN) in a forest after harvesting.

ANN	Input variables	Number of training sessions	Output
1	BA, N, P, PC, FC, PAI_dbh, dbh_max, BA₁, BA₂, BA₃, BA₄, BA₅, BA₆, BA₇, BA₈, N₁, N₂, N₃, N₄, N₅, N₆, N₇, N₈, PAI_dbh1, PAI_dbh2, PAI_dbh3, PAI_dbh4, PAI_dbh5, PAI_dbh6, PAI_dbh7, PAI_dbh8, dbh_max1, dbh_max2, dbh_max3, dbh_max4, dbh_max5, dbh_max6, dbh_max7, dbh_max8	300	NR
2	BA, N, P, PC, FC, PAI_dbh, dbh_max	300	NR
3	BA, N, P, PC, FC, PAI_dbh, dbh_max, BA₁, BA₂, BA₃, BA₄, BA₅, BA₆, BA₇, BA₈	300	NR
4	BA, N, P, PC, FC, PAI_dbh, dbh_max, N₁, N₂, N₃, N₄, N₅, N₆, N₇, N₈	300	NR
5	BA, N, P, PC, FC, PAI_dbh, dbh_max, PAI_dbh1, PAI_dbh2, PAI_dbh3, PAI_dbh4, PAI_dbh5, PAI_dbh6, PAI_dbh7, PAI_dbh8	300	NR
6	BA, N, P, PC, FC, PAI_dbh, dbh_max, dbh_max1, dbh_max2, dbh_max3, dbh_max4, dbh_max5, dbh_max6, dbh_max7 dbh_max8	300	NR
Total		1800

Table 2 Linear correlation between the input and output variables used for training and testing the ANNs.

Variables	NR	PC	P	N	BA	PAI_dbh	dbh_max
NR	1.0000
PC	0.2682*	1.0000
P	0.3468*	0.8050*	1.0000
N	-0.1743*	-0.0150^ns	-0.0339*	1.0000
BA	-0.0469*	0.1448*	0.1138*	0.2632*	1.0000
PAI_dbh	0.2008*	-0.3636*	-0.3483*	-0.1455*	-0.1257*	1.0000
dbh_max	-0.0181^ns	0.1201*	0.0968*	0.0701*	0.8957*	-0.1056*	1.0000

Table 3 Training precision measurements and artificial neural network (ANN) testing for the estimation of tree recruitment.

ANN	Index	Architecture MLP	RMSE		RMSE		Activation functions
ANN	Index	Architecture MLP	RMSE		RMSE		Intermediate layer	Output layer
1	2	41-61-1	34.8791	0.9027	35.6462	0.8932	Hyperbolic tangent	Exponential
2	1	9-18-1	43.1114	0.8489	38.4479	0.8723	Logistic	Exponential
3	5	17-33-1	46.6002	0.8177	38.5054	0.8732	Logistic	Logistic
4	5	17-32-1	41.4371	0.8621	35.9185	0.8908	Logistic	Exponential
5	3	17-28-1	38.1384	0.8825	36.0299	0.8900	Hyperbolic tangent	Logistic
6	3	17-35-1	43.0030	0.8493	39.0262	0.8692	Logistic	Exponential

Universidade Federal de Santa Maria Av. Roraima, 1.000, 97105-900 Santa Maria RS Brasil, Tel. : (55 55)3220-8444 r.37, Fax: (55 55)3220-8444 r.22 - Santa Maria - RS - Brazil
E-mail: cienciaflorestal@ufsm.br

Acompanhe os números deste periódico no seu leitor de RSS

[TFN1] Continuous input variables: BA = basal area (m² ha^-1); N = number of trees ha^-1; P = period in years between measurements; PC = period in years after the harvest; PAIdbh = Annual Periodical Increase on dbh, in mm year^-1; dbhmax: Maximum diameter found in the subplot (mm); Categorical input variables: FC = forest class, composed of Mature Forest, Forest under Construction and Clearings; Continuous output: NR = number of recruited trees; 1, 2, 3, ..., 8: Variables of the adjacent subplots.