The water footprint of cleaning PV farms in West Africa
Start : Sep/Oct 2023
Keywords : Dust cycle, Solar energy, Water resources, Spatiotemporal analysis, Climate projections

Profile and skills required
Master 2 or Engineer Diploma in Geography, Energy Systems, Earth or Climate sciences with a good knowledge of statistics and time series analysis. A good knowledge of the software R (or equivalent) and strong interest in developing scripts is also required. Ability to work with spatial data would be appreciated. Ability and interest to work in a team and with African partners. Motivation for field experiments in Ivory Coast and Senegal. Good knowledge of English for reading articles, report writing.

Scientific context
The potential for photovoltaic (PV) production in sub-Saharan Africa is large and a massive development of PV production installations is ongoing in many sub-Saharan regions. With the aim to foster energy access in the continent, these installations typically take the form of large PV farms, connected to national grids, or of small PV systems to produce electricity for mini-grids in isolated rural areas.
The production to be achieved from PV installations can be however lower than that expected to due dust. Dust in the air first reduces the solar irradiance that reaches the panels. Dust deposit and accumulation on the panels decreases next their production efficiency (Maghami et al. 2016, Mani and Pillai 2010, Sarver et al. 2013). In the West African context, dust transport from the Sahara region can be especially important, and the soiling losses in production may be significant. Losses can be mitigated by cleaning operations of solar panels. If the required cleaning strategies have to be defined so that energy production losses are minimized, they have also to minimize the footprint on regional water resources and not conflict with other water uses (crop irrigation, drinking water supply), especially in areas where water resource is already scarce (You et al. 2018, Micheli et al 2021).

Objectives
The aim of this PhD is to assess the potential reduction that dusts may have on the PV production potential in WA and to provide a water footprint / soiling reduction assessment of different cleaning strategies. This assessment will be obtained with a simulation framework to be developed for different strategies of soiling cleaning within the PhD. It will rely on a local dynamic model of dust deposit on a panel, to be adapted for the WA context based on local observations of soiling (Javed et al. 2017, Coello and Boyle 2019, You et al. 2018). It will allow simulating the time evolution of the soiling accumulation on the panels of a “virtual” PV farm, as a result of both atmospheric dust deposition fluxes and dust removal from wind, precipitation self-cleaning and eventually artificial cleaning operations.

Collaborative context
The PhD is part of the NETWAT ANR project, a collaborative research project, coordinated by IGE and involving researchers in atmospheric and hydrological sciences, as well as practitioners in solar energy. NETWAT mainly aims to develop solutions to ensure an optimized photovoltaic (PV) production in West Africa without compromising the sustainability of the water resource management in the region. In addition to the water footprint evaluation of solar farms, NETWATT also aim to improve the understanding of the West African dust cycle and its direct/indirect effects on solar resource and production and to develop an innovative PV production forecasting chain and build reliable and efficient decision-making tools to optimize power grid management and solar production. NETWAT benefits from partnerships with African academics and national grid managers and solar farms operators. Local observations will be collected within the project for pilot sites in the area (SNO INDAF & AMMA).

Description of the work

• Local observations and empirical modelling of soiling accumulation
  A bibliography will first be conducted to identify a relevant dust accumulation / removal modelling approach. The model will be adapted to local conditions and calibrated against the observations of soiling accumulation obtained from the in-situ field experiments. Different “theoretical” cleaning strategies will be proposed and followed on-field, including “self-cleaning only” strategy (removal is only produced from rain events if any) and “scheduled” or “soiling-conditioned” cleaning operations with different cleaning-rules. Data to be used for the calibration will be continuous dust deposition fluxes estimates from simulations of an atmospheric model, and measurements of soiling losses and soiling accumulation over the panels.
• Mapping soiling losses of PV production over West Africa
  The calibrated model will be used to simulate the dynamic of soiling accumulation and the time evolution of resulting soiling losses that would have been obtained for the recent years (i.e. 20 last years) for a “virtual” PV farm. The simulation will be produced for each location of a regular 0.125° grid covering the whole WA region. This will allow estimating the mean efficiency of PV farms to be expected anywhere in the region when accounting for dusts dynamics. Simulations will then be performed for the “theoretical" cleaning strategies, allowing to assess the efficiency of each on the mean PV production to be expected for any location.
• Mapping the water footprint of PV production over West Africa
  The water footprint associated with the PV production will be defined from the water amount required for cleaning operations and compared with the water resource locally available from rainfall, or needed for other water uses such as agriculture. The water footprint will be estimated for any location in WA for illustrative PV production systems of small size (off-grid PV panel farms in a village) and large size (utility scale PV farms) for the recent climate conditions, and mapped over WA for different cleaning strategies. The optimal cleaning strategies will be compared based on the trade-off between water footprint minimization and/or energy production maximization. The water footprint of the cleaning strategies will also be compared with the water recharge in the region, as well as other water needs in the regions, including crop water needs, drinking water consumption or evaporation from water reservoirs.

Model and Data
Estimates will be based on multi-year simulations forced with weather conditions ERA5 reanalyses and on dust deposit estimates from the CHIMERE atmospheric model. This will allow accounting for the interannual variability of weather conditions.

Practical aspects
Supervision The work will be co-supervised by the IGE (Institute of Environmental Geosciences) and LISA (Laboratoire Interuniversitaire des Systèmes Atmosphériques) laboratories, and in close collaboration with the company SteadySun.
Location : Grenoble, France, with long stays in Creteil
Missions for observations at measurement stations in solar farms in West Africa
Contacts :
–  Louise Crochemore (IGE)
–  Benoît Hingray(IGE)
–  Béatrice Marticorena (LISA)

References
Besson, P., Muñoz, C., Ramírez-Sagner, G., Salgado, M., Escobar, R., Platzer, W., 2017. Long-Term Soiling Analysis for Three Photovoltaic Technologies in Santiago Region. IEEE Journal of Photovoltaics 7, 1755–1760. https://doi.org/10.1109/JPHOTOV.2017.2751752
Chini, C. M. & Peer, R. A. M. The traded water footprint of global energy from 2010 to 2018. Scientific Data 8, 7 (2021).
Coello, M., Boyle, L., 2019. Simple Model for Predicting Time Series Soiling of Photovoltaic Panels. IEEE Journal of Photovoltaics 9, 1382–1387. https://doi.org/10.1109/JPHOTOV.2019.2919628
Javed, W., Guo, B., Figgis, B., 2017. Modeling of photovoltaic soiling loss asa function of environmental variables. Solar Energy 157, 397–407. https://doi.org/10.1016/j.solener.2017.08.046
Maghami, M.R., Hizam, H., Gomes, C., Radzi, M.A., Rezadad, M.I., Hajighorbani, S., 2016. Power loss due to soiling on solar panel : A review. Renewable and Sustainable Energy Reviews 59, 1307–1316. https://doi.org/10.1016/j.rser.2016.01.044
Mani, M., Pillai, R., 2010. Impact of dust on solar photovoltaic (PV) performance : Research status, challenges and recommendations. Renewable and Sustainable Energy Reviews 14, 3124–3131. https://doi.org/10.1016/j.rser.2010.07.065
Micheli, L., Fernández, E.F., Almonacid, F., 2021. Photovoltaic cleaning optimization through the analysis of historical time series of environmental parameters. Solar Energy 227, 645–654. https://doi.org/10.1016/j.solener.2021.08.081
Micheli, L., Smestad, G.P., Bessa, J.G., Muller, M., Fernández, E.F., Almonacid, F., 2022. Tracking Soiling Losses : Assessment, Uncertainty, and Challenges in Mapping. IEEE Journal of Photovoltaics 12, 114–118. https://doi.org/10.1109/JPHOTOV.2021.3113858
Sarver, T., Al-Qaraghuli, A., Kazmerski, L.L., 2013. A comprehensive review of the impact of dust on the use of solar energy : History, investigations, results, literature, and mitigation approaches. Renewable and Sustainable Energy Reviews 22, 698–733. https://doi.org/10.1016/j.rser.2012.12.065
You, S., Lim, Y.J., Dai, Y., Wang, C.-H., 2018. On the temporal modelling of solar photovoltaic soiling : Energy and economic impacts in seven cities. Applied Energy 228, 1136–1146. https://doi.org/10.1016/j.apenergy.2018.07.020

Updated on 2 mars 2023