Artificial neural networks: Modeling tree survival and mortality in the Atlantic Forest biome in Brazil

Abstract

Models to predict tree survival and mortality can help to understand vegetation dynamics and to predict effects of climate change on native forests. The objective of the present study was to use Artificial Neural Networks, based on the competition index and climatic and categorical variables, to predict tree survival and mortality in Semideciduous Seasonal Forests in the Atlantic Forest biome. Numerical and categorical trees variables, in permanent plots, were used. The Agricultural Reference Index for Drought (ARID) and the distance-dependent competition index were the variables used. The overall efficiency of classification by ANNs was higher than 92% and 93% in the training and test, respectively. The accuracy for classification and number of surviving trees was above 99% in the test and in training for all ANNs. The classification accuracy of the number of dead trees was low. The mortality accuracy rate (10.96% for training and 13.76% for the test) was higher with the ANN 4, which considers the climatic variable and the competition index. The individual tree-level model integrates dendrometric and meteorological variables, representing a new step for modeling tree survival in the Atlantic Forest biome.

Necromass Carbon Stock in a Secondary Atlantic Forest Fragment in Brazil

“…Despite this low mortality, the volume of necromass with DBH > 22.5 cm represented 37.32% of the total necromass produced. It is expected that with the forest successional advance there is an increase in tree mortality due to senescence [47][48][49], although other factors such as anthropogenic disturbances [50] or climatic disturbances such as heavy rainfall [44,51], extreme drought [9,52], and El Niño [53] can also boost tree mortality. In addition, the increased presence of necromass in later stages of decomposition indicates that tree mortality may have occurred at more distant times.…”

“…The main driver responsible for this fragmentation was the pressure exerted by human activities, such as agriculture, logging, and urban growth [4][5][6]. These activities together with climate change have affected forest dynamics, altering carbon storage and the tree mortality rate [7][8][9].…”

Identification of Significative LiDAR Metrics and Comparison of Machine Learning Approaches for Estimating Stand and Diversity Variables in Heterogeneous Brazilian Atlantic Forest

“…The use of ANNs for the estimation of forest variables has been growing, and several studies have been developed in Brazilian forests using this technique as an alternative to OLS. Many of these studies are mainly concerned with estimating variables such as height, volume, the shape of the trunk and tapering, and prognosis of yield and production in forests of Eucalyptus spp, Pinus spp, and Tectona grandis, as listed by [81], in Black Wattle plantations [18], in native forests for prediction of the diametric distribution [82,83] and biomass [28], both in the Amazon Forest and in the Atlantic Forest biome, aimed at estimating surviving individuals and mortality within the forest [84].…”

Artificial Neural Networks and Linear Regression Reduce Sample Intensity to Predict the Commercial Volume of Eucalyptus Clones

“…In forestry, ANNs have been applied in several studies [18]. For instance, for modelling forest growth and dynamics [14], predicting height of individual trees [19], diametric distribution [20], prediction of biomass above ground [21], prognosis of diameter [22], volume of stems and branches [23], and modeling of survival and mortality [24]. ANNs are computer systems that are parallel distributed and composed of simple mathematical processing units.…”