Siqueira et al. (2025) Estimation of Kcb for Irrigated Melon Using NDVI Obtained Through UAV Imaging in the Brazilian Semiarid Region

Identification

• Journal: AgriEngineering
• Year: 2025
• Date: 2025-10-10
• Authors: J.M. Siqueira, Gertrudes Macário de Oliveira, Pedro Rogério Giongo, José Henrique da Silva Taveira, Edgo Jackson Pinto Santiago, Mário de Miranda Vilas Boas Ramos Leitão, Lígia Borges Marinho, Wagner Martins dos Santos, Alexandre Maniçoba da Rosa Ferraz Jardim, Thieres George Freire da Silva, Marcos Vinícius da Silva
• DOI: 10.3390/agriengineering7100340

Research Groups

• Department of Technology and Social Sciences (DTCS), State University of Bahia (UNEB), Brazil
• Institute of Agricultural Sciences and Sustainability (IACAS), State University of Goiás (UEG), Brazil
• Department of Statistics and Informatics (DEINFO), Federal Rural University of Pernambuco (UFRPE), Brazil
• Department of Agricultural and Environmental Engineering, University of Vale do São Francisco (UNIVASF), Brazil
• Department of Agricultural Engineering, Federal Rural University of Pernambuco, Brazil
• Institute of Biosciences, São Paulo State University (UNESP), Brazil
• Chapadinha Science Center, Federal University of Maranhão, Brazil

Short Summary

This study estimated the basal crop coefficient (Kcb) for irrigated melon using Normalized Difference Vegetation Index (NDVI) derived from Unmanned Aerial Vehicle (UAV) imagery in the Brazilian semiarid region. It demonstrated that UAV-derived NDVI provides reliable Kcb estimates across warm and cool seasons, supporting precision irrigation and efficient water management.

Objective

• To estimate the basal crop coefficient (Kcb) using UAV-derived NDVI, providing a practical tool for water management in melon cultivation under semiarid conditions in northern Bahia.

Study Configuration

• Spatial Scale: Experimental area of the Department of Technology and Social Sciences (DTCS) at the State University of Bahia (UNEB), in Juazeiro municipality, northern Bahia, Brazil (09°24′50″ S latitude, 40°30′10″ W longitude, 368 m altitude).
• Temporal Scale: Two distinct periods: warm (October–December 2019) and cool (June–August 2020).

Methodology and Data

• Models used:
  • Penman-Monteith method (FAO-56) for reference evapotranspiration (ETo).
  • Hargreaves-Samani method for ETo (for comparison in T3).
  • Random Forest classifier for pixel-by-pixel classification.
  • Linear regression models for Kcb estimation from NDVI.
  • Polynomial regression for Leaf Area Index (LAI) and NDVI relationship.
• Data sources:
  • UAV imagery (Phantom 4 with Mapir Survey3 multispectral sensor: Red, Green, Near-Infrared bands).
  • Field experiments with two melon cultivars ('Gladial' and 'Cantaloupe').
  • Water balance drainage lysimeter evapotranspiration meters for crop evapotranspiration (ETc).
  • Meteorological data (air temperature, relative humidity, solar radiation, rainfall).
  • Leaf area index (LAI) measurements (average number of leaves per plant).

Main Results

• UAV-derived NDVI enabled reliable estimation of melon Kcb across seasons.
• Estimated Kcb values for the 'Gladial' cultivar ranged from 2.753 to 3.450 during the warm period and from 0.815 to 0.993 during the cool period.
• For the 'Cantaloupe' cultivar, Kcb values ranged from 3.087 to 3.856 during the warm period and from 1.118 to 1.317 during the cool period.
• NDVI-based Kcb estimates showed strong correlations with field data (r > 0.80).
• Linear regression models between Kc-FAO and NDVI were statistically significant (p < 0.05) for both cultivars in the warm period (R² = 0.83 for 'Gladial', R² = 0.81 for 'Cantaloupe').
• During the cool period, an inverse correlation was observed, with significance only for 'Gladial' (R² = 0.81).
• Model performance for Kcb estimation was classified as "good" (rRMSE between 10% and 20%) for the warm period and "very good" (rRMSE < 10%) for the cool period.
• Global accuracy of Random Forest classification ranged from 51.78% to 86.68% (warm period) and 74.69% to 88.41% (cool period), with Kappa coefficients indicating moderate to substantial agreement (0.43 to 0.75).
• Drip irrigation system delivered 1.75 L h⁻¹ at 1.0 kgf cm⁻² (approximately 98.1 kPa) pressure, with a distribution uniformity coefficient of 90%.
• Average solar radiation was 20.9 MJ m⁻² day⁻¹ during the warm period and 14.2 MJ m⁻² day⁻¹ during the cool period.
• Average air temperature was 28.5 °C (warm) and 23.8 °C (cool), with average relative humidity of 48.9% (warm) and 65% (cool).

Contributions

• Demonstrated the effectiveness of UAV-derived NDVI for accurate Kcb estimation in melon cultivation, particularly in semiarid regions.
• Provided specific Kcb ranges for 'Gladial' and 'Cantaloupe' melon cultivars across different seasons (warm and cool) in the Brazilian semiarid.
• Offered a practical and efficient remote sensing tool for improved water management and precision irrigation in melon production, addressing a critical need for efficient water resource utilization in the region.
• Highlighted the influence of climatic factors (solar radiation, temperature, relative humidity) on Kcb and NDVI-Kcb relationships across seasons.

Funding

• This research received no external funding.

Citation

@article{Siqueira2025Estimation,
  author = {Siqueira, J.M. and Oliveira, Gertrudes Macário de and Giongo, Pedro Rogério and Taveira, José Henrique da Silva and Santiago, Edgo Jackson Pinto and Leitão, Mário de Miranda Vilas Boas Ramos and Marinho, Lígia Borges and Santos, Wagner Martins dos and Jardim, Alexandre Maniçoba da Rosa Ferraz and Silva, Thieres George Freire da and Silva, Marcos Vinícius da},
  title = {Estimation of Kcb for Irrigated Melon Using NDVI Obtained Through UAV Imaging in the Brazilian Semiarid Region},
  journal = {AgriEngineering},
  year = {2025},
  doi = {10.3390/agriengineering7100340},
  url = {https://doi.org/10.3390/agriengineering7100340}
}