Agrivoltaics : Opportunities in Africa
Using Python to show the possibilities of this technology to help poorer rural communities tackle food and energy security challenges
What are agrivoltaic (AV) systems? A merging of agriculture with raised photovoltaic (PV) solar generation. Evidenced impacts include greater crop yields due to shade from the panels reducing heat and water loss stress, improved PV generation efficiencies due to cooler panels, and reduced irrigation demands due to reduced evaporation and increased soil water retention, which in turn offer food, energy and water security. This has become especially vital in regions most significantly impacted by rising global temperatures.
Can this be used across all crops?
No, there is very little information about the shade tolerance of crops beneath PV modules. However, as shown below it is known that at higher light intensities the photosynthesis (growth) rate reaches saturation after which further increases in irradiance have little effect on the photosynthetic rate.
Certain crops such as wheat, rice, potato and lettuce have lower light saturation points, occurring at a lower Photosynthetically Active Radiation (PAR) levels, making these crops ideal for agrivoltaic applications. Studies have actually found that using AV systems can increase crop yield, which is most likely due to this phenomenon and a reduction in heat stress [1].
Why do we need Agrivoltaics?
Because most agrivoltaic experiments are located in the Global North, the extent of their potential to provide socioeconomic, environmental, and livelihood benefits in the Global South is only estimated at this time. However, such benefits are likely to be greatest in parts of the Global South, where there is abundant solar radiation, a huge need for decentralised energy solutions, extensive food security challenges, and an increase in the frequency of droughts due to climate change. This is an unparalleled opportunity for AV systems to bring sustainable livelihood benefits along with food and energy security.
Modelling an Agrivoltaic System in Ethiopia
A 0.23 acre (76 kWp) agrivoltaic system in Ethiopia’s Lake Tana Basin was modelled to better understand the potential benefits and operation of this system.
A North-South array orientation was used to ensure a more even distribution of sunlight throughout the day for the underlying crops. The PVLIB python library was used to estimate the spatial-temporal interception of direct and diffuse sunlight of both the fixed-tilt panels and the ground beneath using a view factors approach. Moreover, bifacial panels (double-sided panels) were considered to see the additional benefits they could provide in collecting the reflected diffuse and direct irradiance from the surrounding surfaces. As shown below the introduction of the panels results in noticeable reduction in the irradiance reaching the plants, however still tending to rise above the daily useful PAR for lettuce at circa. 213 𝑊/𝑚₂ [2].
Plant growth model
Lettuce (Great-lakes cultivar) was the crop of choice which in Ethiopia is commonly planted after the end of rainy season in September and harvested by the end of the year. With reduced rainfall during this period, the AV system would be expected to reduce temperatures under the panels by a couple of degrees [1], minimising heat stresses and transpiration. A dynamic hourly lettuce growth model by Van Henten (1994) [3] was used which considers air temperature, CO₂ concentration and solar irradiance (beneath the panels) as input variables. Starting in August (hour 5088), once the lettuce crop reaches the desired fresh weight per head then the crop is harvested and the cycle restarts after three days — imitating the replanting of lettuce.
Optimising the model
Three variables were considered; array height, azimuth and tilt. Firstly, as displayed below, array height can be seen to have minimal impact on both the solar generation and crop yield. However, its important to ensure it lies above the height required for workers and equipment to operate beneath yet minimises the cost of infrastructure. Secondly, the azimuth can be seen to have an impact on solar generation with the optimal angle being 173° which faces the panels south (towards the sun). Finally, the tilt angle appears to have the greatest influence on both solar and crop yield. Lowering the tilt reduces inter-row shading but eventually causes the panels to face away from the sun. Lettuce profits rise 52% by reducing the tilt angle from 50° to 1°, all due to an increase in daily average PAR. Consequently, this reduces the optimal tilt from 14° for a ground-mount system to 7° for this AV system.
Results
Annual solar generation totalled 137,466 kWh from the front-side of the panels and 22,927 kWh (~17% bifacial gain at quite a low albedo!) from the back-side of the panels. Assuming an off-grid solar PV system and appropriately sized battery this could power approximately 40 average UK homes or 500 average East African homes. This shows how a significantly smaller system or non-bifacial system could still provide the needs of a rural community. If grid connected then a proportion of the annual generation income ($6.7k) to the developer should be used to fund community development and grid-connection of homes.
Lettuce yields were at 43 t/hectare (13,194 head) and took an average of 61 days to mature after transplanting, which is a good yield despite the presence of panels; this time range is within the expected range for the growth of this type of lettuce.
Conclusions
- There is a growing portfolio of evidence demonstrating the performance of agrivoltaic systems in the Global North [4], which provides a valuable evidence base for developing agrivoltaics elsewhere. Similar evidence now needs to be generated from pilot projects in East Africa and the Global South, where environmental and socio-economic conditions mean the sustainable development and livelihood benefits are even greater.
- With 75% of the East African population dependent on rainfed agriculture [4], AV will help mitigate the expected decrease in crop yields for those subject to the harshest effects of climate change.
- Off-grid solar generation can be used as a means to develop rural communities and make them self-sufficient or act as an enabler to connect these communities to the grid.
- With consumption levels in both households and enterprises remaining very modest in East Africa even small (<10kW) AV projects could provide significant community benefits.
Feel free to share your thoughts in the comments, or message me your questions on LinkedIn where I try and post content about data science surrounding energy and climate change.
Check out the code here. Methodology will be added soon…
References
[1]: Brecht Willockx , Bert Herteleer and Jan Cappelle (June, 2020) — Combining photovoltaic modules and food crops: first agrivoltaic prototype in Belgium
[2]: Muhammad Hussnain Riaz (Department of Electrical Engineering, Lahore University) — Crop-specific Optimization of Bifacial PV Arrays for Agrivoltaic Food-Energy Production: The Light-Productivity-Factor Approach
[3] Van Henten — IMAG-DLO(1994) — Sensitivity analysis of a dynamic growth model of lettuce
[4] Richard J. Randle-Boggis (June 2021)— Agrivoltaics in East Africa: Opportunities and challenges
