Espectrometria de reflectância e sequenciamento molecular do gene 16S rRNA no estudo da crosta biológica do solo no bioma Caatinga

Abstract

The ecosystems of arid and semi-arid regions are strongly affected by desertification. Biological soil crusts (BSC or biocrusts) develop especially in these environments. Recently, research around the world has explored the microbial diversity of BSC and its role in biogeochemical cycles. However, in the Caatinga biome, research with BSC is still incipient. Therefore, the objective was to obtain the composition and microbial diversity of BSC in the Caatinga and investigate the functionality of the BSC, map the coverage and establish relationships between BSC elements and Remote Sensing (RS) data. For this, i) sequencing of the 16S rRNA gene and; ii) hyperspectral RS data. 110 samples of BSC were collected in the area of the Nucleus of Desertification, in the city of Irauçuba/CE. Subsequently, spectral analyzes were performed on the BSC, using the FieldSpec Pro FR 3® spectroradiometer. The reflectance factors and the spectral behavior of BSC were also evaluated. Furthermore, the continuum removed was (CR) of the BSC spectra, multi and hyperspectral indices of BSC were calculated and linear regression models were generated to estimate C, N, P and pH contents in BSC. Molecular analyzes were performed on 34 samples of BSC, in which the total DNA of the samples was extracted and, later, the sequencing of the 16S rRNA gene was performed on the Illumina Miseq® platform. Data were grouped into operational taxonomic units (OTUs), which were taxonomically classified using the SILVA database (138), with the help of QIIME (v. 1.9). Thus, indices of microbial diversity and richness (number of OTUs), relative abundance and metabolic functionality of BSC were generated. The most abundant bacterial phyla in BSC were Cyanobacteria, Chloroflexi, Proteobacteria and Actinobacteria, the most abundant class and order in the crusts were Oxyphotobacteria and Nostocales, respectively. For the reflectance spectra, the BSC showed different reflectance factors, with very low values in the visible region (VIS). A subtle flattening was observed between 600 and 700 nm, attributable to absorption by photosynthetic pigments in BSC. The CRCIA classified most samples as BSC-Cyanobacteria. RC extracted 3 absorption nuclei characteristic of CBS. The best estimates of the linear regression models were for 1Der and RC, in relation to pH, with R2 adjusted in the calibration of 0.718 and 0.571; and validation RMSE of 0.182 and 0.159, respectively. Although, still in the initial stages, the results indicate a promising perspective, encouraging further research related to BSC in the Caatinga biome.