https://www.alice.cnptia.embrapa.br/alice/handle/item/250 Artigo em periódico indexado (CNPMS) Sat, 16 May 2026 20:25:49 GMT 2026-05-16T20:25:49Z https://www.alice.cnptia.embrapa.br/alice/handle/doc/1186660 Título: Machine learning algorithms to predict the crops most susceptible to weed occurrence in integrated crop-livestock systems. Autoria: LUZ, A. L. G.; FERNANDES, A. M. da R.; COELHO, F. V.; OLIVEIRA, M. F. de; ALVARENGA, R. C. Conteúdo: ABSTRACT – The objective of this work was to investigate the use of machine learning algorithms to predict the crops most susceptible to weed occurrence in integrated crop-livestock systems, based on environmental factors of climate, soil, and cropping systems, to establish correlations between these elements and the occurrence of weeds. Three datasets were used for this purpose: the first provided quantitative information on the invasive species, the second contained data about the soil, and the last had records of the region’s climate. The algorithms used were Support Vector Machine, Decision Tree, Random Forest, and K-Nearest Neighbors. The application of machine learning algorithms to predict the susceptibility of crops to weed emergence is technically feasible and effective. The Decision Tree and Random Forest algorithms demonstrated the best performance, with both models achieving 99% accuracy. Robust relationships were established between environmental factors (climate, soil, and planting) and the appearance of invasive species in certain crops. The algorithms reproduced the patterns of weed emergence observed under field conditions. RESUMO –O objetivo deste trabalho foi investigar o uso de algoritmos de aprendizado de máquina para prever as culturas mais suscetíveis à ocorrência de plantas daninhas em sistemas de integração de lavoura-pecuária, com base em fatores ambientais de clima, solo e sistemas de cultivo para estabelecer correlações entre esses elementos e a incidência de plantas invasoras. Três conjuntos de dados foram utilizados para esse propósito: o primeiro forneceu informações quantitativas sobre as espécies invasoras, o segundo continha dados sobre o solo e o último possuía registros do clima da região. Os algoritmos utilizados foram Support Vector Machine, Decision Tree, Random Forest e K-Nearest Neighbors. A aplicação de algoritmos de aprendizado de máquina para prever a suscetibilidade das culturas à emergência de plantas daninhas é tecnicamente viável e eficaz. Os algoritmos Decision Tree e Random Forest demonstraram o melhor desempenho, com ambos os modelos atingindo 99% de acurácia. Relações robustas foram estabelecidas entre os fatores ambientais (clima, solo e plantio) e o surgimento de espécies invasoras em determinadas culturas. Os algoritmos reproduziram com sucesso os padrões de emergência de plantas daninhas observados em condições de campo. Thu, 01 Jan 2026 00:00:00 GMT https://www.alice.cnptia.embrapa.br/alice/handle/doc/1186660 2026-01-01T00:00:00Z https://www.alice.cnptia.embrapa.br/alice/handle/doc/1186312 Título: Degraded soils in the Brazilian semiarid region harbor bacteria with high phosphate solubilization potential. Autoria: BUZÓ, L. N.; FRACETTO, F. J. C.; OLIVEIRA-PAIVA, C. A.; NASCIMENTO, C. W. A. do; SOUZA JÚNIOR, V. S. de; SILVA, W. R. da; SILVA, C. C. G. da; FRACETTO, G. G. M. Conteúdo: The widespread deficiency of available phosphorus (P) in soils, particularly in tropical regions, necessitates the application of phosphate fertilizers due to their high adsorption. However, excessive use engenders significant economic costs and contributes to eutrophication of aquatic ecosystems and the mobilization of heavy metals. One viable alternative is the use of phosphate-solubilizing bacteria (PSB), which can solubilize phosphate through the production of acids and enzymes. This study aimed to assess the cultivable PSB community in different pedogenetic horizons of degraded and preserved Planosols and Luvisols (Planossolos and Luvissolos) in the semiarid and subhumid regions of Brazil, as these soils exhibit a high P availability, particularly in deeper layers. Furthermore, the biotechnological potential of the most promising PSB was evaluated in corn. A total of 170 PSB strains were collected, with 96 bacteria isolated from Planosol and 74 from Luvisol. The majority of the PSB exhibited a medium solubilization index and a broad morphophysiological diversity, and all solubilized calcium phosphate in liquid media. To investigate the biotechnological potential of these bacteria, eight PSB strains were selected based on their ability to solubilize calcium phosphate (>160 mg L-1) and to promote plant growth through mechanisms such as high auxin production, exopolysaccharide synthesis, zinc solubilization, and antagonism against the pathogen Fusarium sp. The PSB were identified as follows: PL1A, PL2L (Priestia sp.), PL1B (Neobacillus sp.), PL3J and LU13C (Streptomyces sp.), PL6A (Fictibacillus sp.), and PL18E and LU10E (Bacillus sp.). The PSB were inoculated into corn with the treatments based on the absence of phosphorus (-P) (control), simple superphosphate (SSP), reactive natural phosphate (RP), and the SSP + RP combination. The maximum available P value in the soil was observed in the RP treatment, where LU10E (78.70 mg dm-3) was inoculated, and in the SSP treatment, where LUC13C (69.9 mg dm-3) was inoculated. Isolates LU13C, LU10E, and PL3J exhibited the highest efficiency in phosphorus mobilization, PL1B and PL18E were moderately efficient, while PL1A, PL2L, and PL6A were the least efficient. This study assessed the diversity of phosphate-solubilizing bacteria (PSB) across soil profiles of degraded and vegetated Planosols and Luvisols in the Borborema Province, Northeastern Brazil, resulting in the establishment of the first bacterial collection comprising 170 strains with confirmed potential for phosphorus solubilization and corn growth promotion. Thu, 01 Jan 2026 00:00:00 GMT https://www.alice.cnptia.embrapa.br/alice/handle/doc/1186312 2026-01-01T00:00:00Z https://www.alice.cnptia.embrapa.br/alice/handle/doc/1186075 Título: Growth and morphology of grain sorghum hybrids cultivated in semi-arid region under pre- and post-flowering drought stress. Autoria: BATISTA, P. S. C.; SILVA, K. J. da; MAGALHAES, P. C.; PORTUGAL, A. F.; FERNANDES, J. S. C.; MENEZES, C. B. de; CARVALHO, A. J. de; JESUS, L. N. de Conteúdo: Drought stress is a major factor limiting sorghum yield, especially in regions with irregular rainfall. This study investigated the morphophysiological and growth responses of grain sorghum hybrids to drought stress, comparing the effects at pre- and post-flowering stages. Twenty hybrids were evaluated under well-watered conditions and under drought imposed at the pre- and post-flowering stages, using a randomized complete block design, with three replications. The evaluated traits were plant height, leaf area index, shoot dry weight, and chlorophyll index. Growth parameters, including absolute growth rate, relative growth rate, leaf area ratio, leaf weight ratio, specific leaf area, and net assimilation rate, were also determined. The pre-flowering drought caused the greatest reductions in plant height, leaf area index, and dry weight accumulation, whereas the post-flowering drought had the greatest effect on leaf area ratio, chlorophyll index, and net assimilation rate. The grain yield under pre-flowering water stress was approximately 4,000 kg ha-1, significantly higher than the 2,500 kg ha-1 observed under postflowering stress, suggesting a greater capacity for recovery and grain production when water availability is restored before the reproductive stage. Thu, 01 Jan 2026 00:00:00 GMT https://www.alice.cnptia.embrapa.br/alice/handle/doc/1186075 2026-01-01T00:00:00Z https://www.alice.cnptia.embrapa.br/alice/handle/doc/1185981 Título: Agronomic traits, physiological parameters, chemical composition, and in vitro digestibility of Urochloa brizantha cv. Piatã, Megathyrsus maximus cv. Zuri and Megathyrsus maximus cv. Tamani in silvopastoral systems. Autoria: OLIVEIRA, A. F. de; PEREIRA, W. M. C. R.; GONTIJO NETO, M. M.; VILAS-BOAS, T.; TORRE, F. D.; TEIXEIRA, E. C.; SILVA, E. A. da; RODRIGUES, P. D.; LANA, A. M. Q. Conteúdo: Urochloa species are widely used in silvopastoral systems (SPS), but production losses highlight the need for more productive forages. This study evaluated agronomic, physiological, chemical composition, and in vitro digestibility responses of Urochloa brizantha cv. Piatã and Megathyrsus maximus cvs. Zuri and Tamani in SPS and monoculture (MONO). Forages were sown in MONO or SPS with 166 eucalyptus trees/ha. Nine plots were established per system, giving three replicates per cultivar. Forage accumulation and biomass were 31.8% (35.2 vs. 51.6 kg DM/ha/day) and 32.0% (5735 vs. 8428 kg DM/ha) higher in MONO than SPS. Biomass was 34.1% higher in M. maximus cv. Zuri than U. brizantha cv. Piatã. The leaf/stem ratio was 52.3% lower (3.18 vs. 6.66) in U. brizantha cv. Piatã than M. maximus cultivars. Intercellular carbon was 16.1% (180 vs. 155 μmol CO2/m2 /s) higher in MONO, while water use efficiency was 67.8% lower than in SPS. Megathyrsus maximus cv. Tamani had the highest crude protein (CP) in both systems, and CP in cvs. Tamani and Zuri was 6.43% (140 vs. 149 g DM/kg) and 18.0% (111 vs. 131 g DM/kg) higher in SPS than MONO. Although shading reduces M. maximus production, these cultivars can increase animal production compared with Urochloa in SPS. Thu, 01 Jan 2026 00:00:00 GMT https://www.alice.cnptia.embrapa.br/alice/handle/doc/1185981 2026-01-01T00:00:00Z