Enhancing Irrigation Reservoirs through Floating Photovoltaic: A Satellite Mapping-Based Multifunctional Water-Energy Strategy
Abstract
The potential for floating photovoltaics (FPVs) in Mendoza, Argentina, is quantified by mapping irrigation reservoirs with Sentinel-2 imagery and estimating energy yield, water evaporation mitigation, and avoided greenhouse gas (GHG) emissions. A composite Water Index (WI) is developed by linearly combining the NDWI and MNDWI, both derived from Sentinel-2 multispectral satellite imagery. A total of 380 control points (70% calibration, 30% validation) are used to fit WI parameters. Performance is evaluated against MNDWI, AWEIsh, EWI, and a 10-meter deep-learning land-use/land-cover (LULC) map. The new WI achieves the highest accuracy (approximately 98.2%) with the fewest omissions and outperforms all comparators. The application to the Oasis Norte Oeste (ONO) region identifies 524 waterbodies, totaling 5.92 km². After feasibility screening, 3.55 km² (60%) is deemed suitable for FPV deployment. Scenario analysis indicates that covering 10% of the feasible surface would install approximately 6.7 MWp and yield 11,776 MWh per year, scaling up to 60.6 MWp and 117,757 MWh per year at 90% coverage. Avoided GHG emissions are calculated using the CDM ACM0002/TOOL07 methodology, with Argentina’s grid factor ranging from 3,415 to 34,150 tCO₂e per year across scenarios. These findings demonstrate that FPV can provide land-neutral electricity generation and substantial climate co-benefits in a water-stressed basin. The WI further offers a robust and transferable workflow for reservoir detection and FPV site screening.
Keywords
Floating Photovoltaics, Multispectral Satellite Imagery, Reservoir mapping, Evaporation mitigation, Renewable Energy, Water Scarcity, Distributed Generation