Banca de DEFESA: ANA LIVIA ARAUJO DE AZEVEDO

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : ANA LIVIA ARAUJO DE AZEVEDO
DATE: 27/02/2026
TIME: 08:30
LOCAL: Laboratório Didático de Geociências
TITLE:

Soil Survey of the Raimundo Nonato Settlement, Jucurutu (RN), Brazil: An Integration of Conventional and Ethnopedological Approaches

KEY WORDS:

soil mapping; pedometrics; ethnopedology; rural settlement; semi-arid soils

PAGES: 112
BIG AREA: Ciências Humanas
AREA: Geografia
SUMMARY:

The implementation of large-scale water infrastructure projects in the semi-arid region of Northeastern Brazil generates both benefits and challenges, particularly due to regional socioeconomic restructuring and the relocation of communities affected by dam construction. In the Raimundo Nonato Agrovillage, a settlement established to accommodate families displaced by the Oiticica Dam in Jucurutu, Rio Grande do Norte State, family farming and livestock production constitute the primary economic activities. In this context, understanding soil as a supporting resource becomes essential for farmers seeking to reconstruct their productive practices and establish new territorial roots.This study aimed to analyze the spatial distribution of soils in the Raimundo Nonato Agrovillage using an integrated methodological approach that articulates farmers’ local knowledge with conventional soil survey techniques and digital soil mapping tools. The pedological survey was conducted at a semi-detailed scale (1:25,000). Sampling was guided by geomorphometric covariates derived from LiDAR data and implemented through the Conditioned Latin Hypercube Sampling method, which defined the observation and collection points. Laboratory analyses were performed on these samples, limited to pH, electrical conductivity, particle-size distribution, and sodium and potassium contents. In addition, representative soil profiles were described and sampled across distinct landscape positions, with subsequent physical and chemical analyses and classification according to the Brazilian Soil Classification System. Environmental covariates were incorporated into machine learning algorithms to predict the spatial distribution of soil attributes, thereby supporting the delineation of mapping units. The ethnopedological phase addressed settlers’ environmental perceptions, local soil classification systems, and soil–landscape relationships. Farmers emphasized rainfall irregularity and identified water availability as the main limiting factor for agricultural production. They distinguished shallow and stony soils—locally referred to as “barro branco” and “lajeiro”—from deeper and sandier soils, such as “ariuço,” which are considered more suitable for cultivation. Overall, soils in the study area exhibit predominantly eutrophic character, sandy texture, slightly acidic pH, high-activity clays, and relatively low values of base sum and cation exchange capacity compared to regional standards. Identified soil classes include Orthic Vertic Haplic Luvisol, Lithic Entisols, Fluvent Entisols, and Regolithic Entisols, with Lithic Entisols predominating across different topographic positions and frequently associated with rocky outcrops. Relevant pedodiversity was observed, with land-use limitations related to shallow effective depth, high stoniness, and low clay content. Profiles located in floodplain environments displayed greater granulometric heterogeneity and morphological similarities among themselves. Predictive modeling results showed divergent performance, with satisfactory explanatory capacity only for coarse fragments and coarse sand, whereas the prediction of other soil attributes yielded performance metrics below ideal thresholds. Nevertheless, the predicted spatial distribution patterns were consistent with field observations. Spatial analysis demonstrated strong topographic and parent material control over the variability of soil attributes.

COMMITTEE MEMBERS:
Externo à Instituição - DAVI FEITAL GJORUP
Presidente - 1818226 - JOAO SANTIAGO REIS
Externo à Instituição - RAFAEL GOMES SIQUEIRA
Interna - 1726169 - SARA FERNANDES FLOR DE SOUZA
Externa à Instituição - SIMONE CARDOSO RIBEIRO - URCA