Dinâmica da biomassa radicular e carbono em floresta tropical seca no semiárido brasileiro

Abstract

The Seasonally Dry Tropical Forests (SDTFs) are known for their rich floral diversity even under conditions of water scarcity, and therefore have a high potential for biomass production. However, these enviroments are still poorly researched and studied, especially regarding their relationship with carbon dynamics, particularly for root biomass. Thus, the most commonly used method to understand the relationship between carbon reservoirs and forest biomass dynamics, including root biomass, is by estimating biomass using direct or indirect methods. The present study aimed to evaluate the relationship between intra- and inter-annual dynamics of root biomass and soil organic carbon dynamics in SDTFs in the Brazillian Semiarid region. Data on root and soil biomass were collected in the field during the rainy, intermediate, and dry seasons, with six samples collected per season, totaling 24 samples. Root biomass samples were processed, weighed, and analyzed for root morphological parameters obtained through scanning and WinRhizo software. Soil samples were analyzed in the laboratory for total soil and root organic carbon, and soil organic matter. All data was statistically analyzed in terms of temporal dynamics, Spearman correlation, and principal component analysis (PCA). To estimate biomass using indirect methods, multivariate statistical analysis was used to generate allometric equations using RLM, and a prediction model was created using spectral data obtained with a contact probe and FieldSpec Pro FR 3 spectroradiometer (350- 2500 nm). For the study period, two peculiarities were observed in the temporal behavior of root biomass dynamics regarding seasonality, as a consequence of sensitivity to variations in rainfall between the rainy, intermediate, and dry periods. It was also observed that the proportional values of coarse roots exceeded those of fine roots, with an inverted behavior observed from 2019 compared to the first four years of the study (2015-2018). The data for total soil organic carbon, organic matter, and root organic carbon did not show statistically significant temporal variance, but PCA and Spearman correlation analysis indicated that soil carbon is primarily influenced by fine roots, while root carbon is primarily influenced by coarse roots. Using spectral data obtained through PLRS, a root biomass prediction model for SDTFs was obtained, with R² adj = 0.79 and RPD = 2.04. With respect to allometric equations, it was possible to use the morphological parameters of average diameter and root volume to predict BIOT (p<0.001; R²=0.71) and BIO 2 (p<0.001; R² adj=0.80), but not BIO1. Therefore, although soil carbon is primarily a result of the decomposition of fine roots, the analyses indicated that this does not necessarily mean that the input of biomass in the soil between seasons will significantly affect soil carbon in SDTFs. Additionally, it was possible to generate root biomass estimation models in SDTFs using reflectance spectroscopy and allometric relationships established using morphological parameters, but further validation with new data is required to increase the reliability of these models and enable their use in future studies on belowground carbon stocks.