Diego Vasconcelos

Mr. Diego Vasconcelos

Tecnology and Innovation, SENAI - Brazilian National Service for Industrial Training, Campo Grande, Mato Grosso do Sul, Brazil

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2022 in Brazilian Journal of Biology
Verified
Journal Article
Open access
Diego VasconcelosDiego Vasconcelos
Diego Vasconcelos
TKT. B. S. Künast
T. B. S. Künast
EPEVALDO MARTINS PIRES
EVALDO MARTINS PIRES
JCJ. N. Corassa
J. N. Corassa
Popularly known as the yellow eucalyptus beetle, Costalimaita ferruginea (Fabricius, 1801) (Coleoptera: Chrysomelidae), in the adult phase it feeds on the leaves of the myrtaceae, leaving them laced, a fact that impairs the productivity of plantations. Therefore, this work aims to model the potential distribution of the beetle in climate change scenarios for Brazil. The biotic data were collected through a review of the literature and entomological collections, totaling 88 georeferenced points of occurrence of the insect in the country. The abiotic data (19 biovariables, precipitation and minimum, average and maximum temperature) in the Wordcllim database, represent the periods: current (1970-2000), middle (2041-2060), and final (2061-2080) of the century, with representation in two projections of climatic anomalies (RCP 4.5 and RCP 8.5). To determine the most important variables for the models, the Jackknife test was performed in the Maxent software, which resulted in five biovariables, namely: annual precipitation (35.2%), temperature seasonality (15.2%), annual temperature variation (13.7%), seasonality of precipitation (8.7%) and an average temperature of the coldest quarter (7.4%). Subsequently, the Openmodeller software was used and five algorithms were tested to determine which model represents the prediction of areas of suitability for the occurrence of the insect. The algorithm that best represented the appropriate areas was the Envelope Score (AUC = 0.808), corroborating the occurrence data collected. The prediction shows that the Pampa biome, in the RCP8.5 scenario for the period between 2061-2080, will become fully suitable for the occurrence of this defoliator beetle, unlike the Amazon, which presents retraction in areas suitable for the occurrence of the beetle for the same period. In this sense, commercial eucalyptus plantations implemented in climatologically suitable areas for the occurrence of this insect must be monitored periodically.

27 June 2026 in Remote Sensing Applications: Society and Environment
Verified
Journal Article
Closed access
RLRômullo Oliveira Louzada
Rômullo Oliveira Louzada
Diego VasconcelosDiego Vasconcelos
Diego Vasconcelos
GSGabriel Ferreira da Silva
Gabriel Ferreira da Silva
FVFernanda Vitorio Vieira
Fernanda Vitorio Vieira
JCJunior da Silva Camargo
Junior da Silva Camargo
TRThiago Gentil Ramires
Thiago Gentil Ramires

4 November 2026 in Plants
Verified
Journal Article
Open access
PDF file
JAJailson Vieira Aguilar
Jailson Vieira Aguilar
SRSuellen Rodrigues Ramalho
Suellen Rodrigues Ramalho
AFAlana Martini Ferreira
Alana Martini Ferreira
KGKamila De Sousa Gomes
Kamila De Sousa Gomes
Jessica Carolina Medina GallardoJessica Carolina Medina Gallardo
Jessica Carolina Medina Gallardo
Desireé Soares SilvaDesireé Soares Silva
Desireé Soares Silva
The Pantanal biome harbors a high diversity of aquatic macrophytes that remain largely unexplored as potential sources of plant-derived agricultural biostimulants. This study evaluated six aquatic macrophytes from the Pantanal: Eleocharis cellulosa, Pistia stratiotes, Thalia geniculata, Pontederia sp., Pontederia rotundifolia, and Pontederia azurea as potential sources of plant-derived biostimulants. Aqueous and methanolic extracts were prepared by dynamic maceration and tested on soybean seed treatment. FTIR spectroscopy was used to characterize functional groups. Final germination percentage ranged from 86% to 97% across treatments, and no extract significantly reduced this endpoint compared with the respective controls, indicating the absence of detectable phytotoxic effects under the tested conditions. P. rotundifolia, T. geniculata, and P. azurea promoted increases in biomass and length (up to 0.44 g of epicotyl dry mass and 14.62 cm of root length), emerging as promising biostimulant source. E. cellulosa and Pontederia sp. showed neutral or negative effects on most growth variables. P. stratiotes remained neutral. FTIR analysis revealed distinct spectral profiles between extraction methods. Aqueous extracts showed prominent absorption bands associated with carbohydrate and polysaccharide-related functional groups, whereas methanolic extracts exhibited additional bands consistent with phenolic, aromatic, ester, and lipid-associated functional groups. The aqueous route stood out for producing bioactive extracts without organic solvents, a key advantage for sustainable scaling.

4 November 2026 in Forests
Verified
Journal Article
Open access
PDF file
RLRômullo Oliveira Louzada
Rômullo Oliveira Louzada
.S. Da Silva
. Da Silva
SCStaci Arielli Caires Correia
Staci Arielli Caires Correia
MCMurillo Couto
Murillo Couto
Diego VasconcelosDiego Vasconcelos
Diego Vasconcelos
RTRenato Mauricio Tavares
Renato Mauricio Tavares
Machine-learning models trained on remote sensing data are widely used to map ecosystem carbon stocks, but their validation often ignores spatial autocorrelation, inflating apparent predictive skill. We tested whether this concern, established for large-scale satellite-based biomass mapping, also holds for UAV-LiDAR carbon estimation at the plot scale, comparing continuous structural metrics, segmentation-derived crown metrics, and Sentinel-2 spectral variables for predicting aboveground biomass (AGB) and soil organic carbon (SOC) across 29 inventory plots in a Cerradão-dominated Pantanal–Cerrado ecotone, Brazil. Random Forest models were evaluated under fourteen predictor scenarios and two allometric formulations. A single spatial partition yielded seemingly reasonable accuracy for AGB (best R2 = 0.27), whereas SOC showed weak apparent skill, even under a single partition (best rRMSE = 33.7%, R2 = −0.14); repeating validation across 20 independent spatial partitions, combined with leakage-free nested variable selection, showed predictive skill collapsing to near-zero or negative R2 for all 98 tested scenario–response combinations (mean R2: −0.44 to −0.003), a pattern statistically indistinguishable from a null model built on permuted response data. Variable selection stability differed from full-dataset importance rankings: raw spectral bands and canopy vertical-complexity metrics were the most consistently retained predictors across partitions for AGB and SOC, respectively, even though a derived spectral index and a lower-canopy structural metric dominated single-model importance for each response. Cross-pool coupling tests showed no meaningful gain from using AGB to predict SOC or vice versa. Empirical variogram analysis further indicated that the spatial blocking scheme adequately separated the two response variables, although residual spatial dependence could not be excluded for a subset of predictors. Validation strategy, rather than sensor combination or vegetation representation, was the primary determinant of apparent predictive skill, cautioning against overinterpreting single-partition accuracy in small, heterogeneous tropical landscapes.

24 July 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
RLRômullo Oliveira Louzada
Rômullo Oliveira Louzada
.S. Da Silva
. Da Silva
SCStaci Arielli Caires Correia
Staci Arielli Caires Correia
MCMurillo Couto
Murillo Couto
Diego VasconcelosDiego Vasconcelos
Diego Vasconcelos
RTRenato Mauricio Tavares
Renato Mauricio Tavares
Machine-learning models trained on remote sensing data are widely used to map ecosystem carbon stocks, but their validation often ignores spatial autocorrelation, inflating apparent predictive skill. We tested whether this concern, established for large-scale satellite-based biomass mapping, also holds for UAV-LiDAR carbon estimation at the plot scale, comparing continuous structural metrics, segmentation-derived crown metrics, and Sentinel-2 spectral variables for predicting aboveground biomass (AGB) and soil organic carbon (SOC) across 29 inventory plots in a Cerradão-dominated Pantanal–Cerrado ecotone, Brazil. Random Forest models were evaluated under fourteen predictor scenarios and two allometric formulations. A single spatial partition yielded seemingly reasonable accuracy for AGB (best R² = 0.27), whereas SOC showed weak apparent skill even under a single partition (best rRMSE = 33.7%, R² = −0.14); repeating validation across 20 independent spatial partitions, combined with leakage-free nested variable selection, showed predictive skill collapsing to near-zero or negative R² for all 98 tested scenario–response combinations (mean R²: −0.44 to −0.003). Variable selection remained comparatively stable across partitions, with a Sentinel-2 shortwave-infrared band and canopy vertical-complexity metrics consistently retained, and cross-pool coupling tests showed no meaningful gain from using AGB to predict SOC or vice versa. Validation strategy, rather than sensor combination or vegetation representation, was the primary determinant of apparent predictive skill, cautioning against overinterpreting single-partition accuracy in small, heterogeneous tropical landscapes.

27 August 2026 in Forest Ecology and Management
Verified
Journal Article
Closed access
RLRômullo Oliveira Louzada
Rômullo Oliveira Louzada
Diego VasconcelosDiego Vasconcelos
Diego Vasconcelos
ABAndré Nogueira Bozza
André Nogueira Bozza
JCJoão Gabriel Chaib
João Gabriel Chaib
RTRenato Maurício Barbosa Tavares
Renato Maurício Barbosa Tavares