Centanni et al. (2025) Assessment of pollutants from the Canale d'Aiedda basin to the sea: SWAT model and Remote Sensing Approach

Identification

• Journal: Ecological Informatics
• Year: 2025
• Date: 2025-09-05
• Authors: Marco Centanni, Rose Mary, Giovanni Romano, Giovanni Francesco Ricci, Ossama M. M. Abdelwahab, Julio Pérez‐Sánchez, Anna Maria De Girolamo, Francesco Gentile
• DOI: 10.1016/j.ecoinf.2025.103410

Research Groups

• Department of Soil, Plants and Food Sciences, University of Bari Aldo Moro, Bari, Italy
• National Research Council, Water Research Institute, Bari, Italy
• Universidad de Las Palmas de Gran Canaria, Gran Canaria, Spain

Short Summary

This study assesses the spatial patterns of pollutants from the Canale d’Aiedda basin to the Mar Piccolo Sea, identifying nutrient sources and their coastal fate by coupling the ecohydrological SWAT model with Sentinel-2 remote sensing. It found that agricultural subbasins are primary sources of nitrogen and phosphorus, with flash floods significantly impacting pollutant delivery and coastal turbidity.

Objective

• To assess spatial patterns of pollutants from the catchment to the sea, identifying the sources of nutrients within a basin, and their fate in the sea.
• To quantify spatial and temporal patterns of sediment and nutrient generation within the Canale d’Aiedda basin.
• To identify critical source areas contributing disproportionately to pollutant loading.
• To characterize the extent, intensity, and deposition patterns of sediment plumes in the receiving Mar Piccolo coastal embayment during a flash flood event.

Study Configuration

• Spatial Scale: Canale d’Aiedda basin (360 km²) and Mar Piccolo Sea (Apulia, Southern Italy). The basin was discretized into 40 subbasins and 271 Hydrological Response Units (HRUs).
• Temporal Scale: SWAT model simulations from 2014 to 2023 (with 3 years warm-up), calibrated using daily flow and discrete sediment/nutrient concentrations from 2017 to 2021. Remote sensing analysis focused on a flash flood event on June 10, 2023, with comparative images from 2022 and before/after the event.

Methodology and Data

• Models used:
  • Soil & Water Assessment Tool (SWAT) ecohydrological model (version 2009)
  • SWAT-Calibration and Uncertainty Programs (SWAT-CUP) with Sequential Uncertainty Fitting (SUFI-2) algorithm for calibration.
• Data sources:
  • Digital Elevation Model (DEM): Puglia Region (8 m spatial resolution).
  • Land Use Map: Puglia Region land use map (1:5000, 21 land uses) integrated with National Agricultural Census data.
  • Soil Map and Database: Puglia Region (2001) ACLA2 project, JRC-ESDAC (1:100000, 500 m spatial resolution, 11 soil types).
  • Point Sources: Wastewater Treatment Plants data from Puglia Water Authority and Regional Agency for Environmental Protection.
  • Meteorological Data: Civil Protection Service – Puglia Region and Regional Agency for Irrigation and Forestry Activities.
  • Agricultural Practices: Farmers’ and dealers’ interviews.
  • Streamflow Data: Daily measurements from 2017 to 2021 at two gauging stations (subbasins 27 and 29).
  • Discrete Water Quality Data: Sediment, Total Nitrogen (TN), and Total Phosphorus (TP) concentrations collected from 2017 to 2021.
  • Remote Sensing Data: Sentinel-2 L2A satellite images (10 m, 20 m, 60 m spatial resolution, 13 spectral bands, 5-day revisit time) processed on Google Earth Engine (GEE).
  • Spectral Indices: Automated Water Extraction Index (AWEI), Normalized Difference Water Index (NDWI), Normalized Difference Turbidity Index (NDTI).

Main Results

• The SWAT model successfully simulated hydrological processes, sediment transport, and nutrient dynamics with satisfactory to good statistical performance (e.g., streamflow R² 0.50-0.60, NSE 0.50-0.58; sediment R² 0.97-1.00, NSE 0.88-0.98; TN R² 0.72-0.75, NSE 0.59-0.73; TP R² 0.83-0.90, NSE 0.77-0.83).
• The highest specific loads of Total Nitrogen (TN) (~10 kg ha⁻¹ y⁻¹) were simulated in northwestern agricultural subbasins (vineyards, olive groves, winter wheat).
• The highest specific loads of Total Phosphorus (TP) (0.7 kg ha⁻¹ y⁻¹) were predicted near the basin outlet, in subbasins dominated by vineyards.
• Temporal distribution showed peak TN and TP loads in April and October, linked to precipitation and agricultural activities. Elevated TP loads in June and August were attributed to Wastewater Treatment Plant (WWTP) discharges during low natural runoff.
• A flash flood event on June 10, 2023 (29.72 mm rainfall, 0.066 m³ s⁻¹ streamflow at outlet) delivered significant loads to the sea: 31.43 tonnes of sediment, 19.4 kg of TN, and 1.43 kg of TP.
• Remote sensing using NDTI detected a significant increase in turbidity in the Mar Piccolo and d’Aiedda estuarine zone after the flash flood, with mean NDTI increasing by more than 0.05 units in the estuarine channel.
• The sediment plume showed an asymmetric spatial distribution, with higher turbidity concentrated in the southeastern area near the d’Aiedda outlet, influenced by local currents.

Contributions

• This study provides a novel integrated methodological framework combining ecohydrological modeling (SWAT) with remote sensing (Sentinel-2 NDTI) to characterize the complete source-to-sea continuum of sediment and nutrient transport in an intermittent Mediterranean watershed.
• It addresses significant knowledge gaps regarding the fate and transport of sediments and associated pollutants across the land-sea interface, particularly in understudied episodic watersheds with intermittent streams.
• The research identifies critical source areas for nutrient pollution within the Canale d’Aiedda basin, offering actionable information for targeted agricultural management interventions.
• It quantifies the significant impact of flash floods on episodic pollutant delivery to coastal environments, highlighting the need for management strategies that account for extreme hydrological events in Mediterranean regions.
• The study demonstrates the complementary nature of SWAT's continuous temporal predictions at specific points and satellite imagery's extensive spatial coverage at discrete intervals, proving valuable for watersheds where conventional monitoring is limited.

Funding

• European Union Next-Generation EU (PIANO NAZIONALE DI RIPRESA E RESILIENZA (PNRR)—MISSIONE 4 COMPONENTE 2, INVESTIMENTO 1.4—D.D. 1032 17/06/2022, CN00000022) within the Agritech National Research Center.

Citation

@article{Centanni2025Assessment,
  author = {Centanni, Marco and Mary, Rose and Romano, Giovanni and Ricci, Giovanni Francesco and Abdelwahab, Ossama M. M. and Pérez‐Sánchez, Julio and Girolamo, Anna Maria De and Gentile, Francesco},
  title = {Assessment of pollutants from the Canale d'Aiedda basin to the sea: SWAT model and Remote Sensing Approach},
  journal = {Ecological Informatics},
  year = {2025},
  doi = {10.1016/j.ecoinf.2025.103410},
  url = {https://doi.org/10.1016/j.ecoinf.2025.103410}
}