The burgeoning integration of photovoltaic (PV) systems with agricultural land use, often termed agrivoltaics, presents a promising avenue for optimizing land productivity. This synergy is particularly compelling for shade-tolerant crops, such as medicinal herbs, which can thrive in the moderated microclimate beneath solar panel arrays. My investigation focused on one such valuable species, Polygonatum kingianum, a perennial herb of significant economic importance in traditional medicine. A critical, yet often overlooked, aspect of cultivating any crop—whether under open skies or the novel environment of a solar farm—is the structure and dynamics of its associated arthropod community. Understanding this community, encompassing pests, natural enemies, and neutral species, is the fundamental cornerstone for developing ecologically sound integrated pest management (IPM) strategies. Therefore, the primary objective of my research was to conduct a systematic, year-long comparative analysis of the arthropod fauna inhabiting P. kingianum plants grown under the canopy of solar panels versus those cultivated in conventional open-field conditions. I sought to quantify and compare community composition, biodiversity indices, and the population trajectories of key pest species to elucidate the ecological implications of this dual land-use system.
Materials and Methodological Framework
The study was conducted from January to December 2023 in a commercial agrivoltaic site. The site featured parallel cultivation systems: a treatment area where P. kingianum was planted directly beneath fixed-tilt photovoltaic arrays and a control area of traditional open-field cultivation at the same locality. The planting density was standardized at 16 plants per square meter in both systems.
Arthropod sampling was performed monthly using a dual-method approach to ensure comprehensive coverage of the community. First, a direct visual census was conducted. Within each cultivation system (solar panel and open-field), three fixed sampling plots were established. In each plot, 20 randomly selected P. kingianum plants were meticulously examined, and all observed arthropods on the aerial parts were recorded, identified to the lowest possible taxonomic level (typically species or genus), and counted.
Second, to capture flying and highly mobile insect species, blue sticky card traps were deployed. Five standard-sized blue cards were placed in each cultivation system, spaced 10 meters apart and suspended 30 cm above the ground. The traps were left in the field for a 24-hour period each month before being collected. Arthropods captured on the cards were later identified and counted under laboratory stereomicroscopes. This combination of methods provided robust data on both resident and mobile components of the arthropod community.
The collected data were analyzed using ecological community indices. The structure was dissected by categorizing arthropods into functional sub-communities: pests (phytophagous species), natural enemies (predators and parasitoids), and neutral insects (non-interacting species). Biodiversity was assessed using the following standard metrics, calculated for each month and for the total annual community:
Shannon-Wiener Diversity Index (H’): $$ H’ = -\sum_{i=1}^{S} (p_i \cdot \ln p_i) $$
Pielou’s Evenness Index (J’): $$ J’ = \frac{H’}{\ln S} $$
Margalef’s Richness Index (D): $$ D = \frac{S – 1}{\ln N} $$
Simpson’s Dominance Index (C): $$ C = \sum_{i=1}^{S} p_i^2 $$
where \( S \) is the total number of species, \( N \) is the total number of individuals, and \( p_i \) is the proportion of individuals belonging to the \( i \)-th species (\( p_i = N_i / N \)).
The similarity between the arthropod communities of the two cultivation systems was evaluated using the Jaccard similarity coefficient (\( q \)): $$ q = \frac{c}{a + b – c} $$ where \( a \) and \( b \) are the number of species in communities A and B, respectively, and \( c \) is the number of species common to both. A final, critical ratio was calculated: the Natural Enemy to Pest ratio (NE:P), based on individual counts, to gauge the potential for biological control in each system over time.
Results: A Comparative Ecological Portrait
Overall Arthropod Community Composition
My year-long survey revealed a rich tapestry of arthropod life associated with P. kingianum. A total of 1,090 individual arthropods, representing 27 species across 25 families and 10 orders, were documented in the solar panel environment. In contrast, the open-field system hosted a slightly higher total abundance and species count, with 2,641 individuals comprising 30 species from 28 families. The overall composition by class and order is summarized in Table 1.
| Cropping System | Class | Order | Families (#) | Species (#) | Individuals (#) |
|---|---|---|---|---|---|
| Under Solar Panel | Insecta | Orthoptera | 4 | 4 | 18 |
| Hemiptera | 4 | 4 | 19 | ||
| Thysanoptera | 1 | 1 | 27 | ||
| Coleoptera | 2 | 4 | 15 | ||
| Dermaptera | 1 | 1 | 7 | ||
| Hymenoptera | 2 | 2 | 4 | ||
| Diptera | 4 | 4 | 44 | ||
| Mantodea | 1 | 1 | 1 | ||
| Arachnida | Acarina | 1 | 1 | 899 | |
| Arachnida | Araneae | 5 | 5 | 56 | |
| Total | 25 | 27 | 1,090 | ||
| Open-Field | Insecta | Orthoptera | 5 | 5 | 38 |
| Hemiptera | 6 | 6 | 37 | ||
| Thysanoptera | 1 | 1 | 58 | ||
| Coleoptera | 2 | 4 | 13 | ||
| Dermaptera | 1 | 1 | 7 | ||
| Hymenoptera | 2 | 2 | 3 | ||
| Diptera | 4 | 4 | 40 | ||
| Mantodea | 1 | 1 | 6 | ||
| Arachnida | Acarina | 1 | 1 | 2,378 | |
| Arachnida | Araneae | 5 | 5 | 61 | |
| Total | 28 | 30 | 2,641 |
The most striking feature was the overwhelming numerical dominance of the Acarina (mite) order in both systems, constituting 82.48% and 90.04% of all individuals under the solar panel and in the open field, respectively. This single order, represented predominantly by the spider mite Tetranychus kanzawai, defined the abundance structure of the entire community.
Functional Sub-Community Structure and Relative Abundance
Categorizing species by their ecological function provided deeper insight. The pest sub-community under the solar panel was composed of 11 species, compared to 14 species in the open field. The natural enemy and neutral insect sub-communities each contained 12 and 4 species, respectively, in both environments. The relative abundance (RA) of these groups, however, differed markedly (Table 2). The pest sub-community accounted for 88.44% of all individuals under the solar panel, a significantly lower proportion than the 95.12% observed in the open field. Consequently, the combined share of natural enemies and neutral insects was higher in the solar panel system.
| Functional Group | Order | Dominant Species Example | Relative Abundance (%) Solar Panel |
Relative Abundance (%) Open Field |
|---|---|---|---|---|
| Pests | Orthoptera | Xenocatantops humilis | 1.65 | 1.44 |
| Hemiptera | Thrips flavus (Yellow Thrips) | 1.74 | 1.40 | |
| Thysanoptera | Thrips flavus | 2.48 | 2.20 | |
| Coleoptera | Anomala spp. | 1.38 | 0.49 | |
| Acarina | Tetranychus kanzawai | 82.48 | 90.04 | |
| Pest Subtotal | 88.44 | 95.12 | ||
| Natural Enemies | Araneae | Lycosa sp. (Wolf Spider) | 5.14 | 2.31 |
| Coleoptera | Propylaea japonica (Lady beetle) | 0.64 | 0.49 | |
| Diptera | Eristalis tenax (Hoverfly) | 1.56 | 0.34 | |
| Natural Enemy Subtotal | 8.90 | 3.63 | ||
| Neutral Insects | Diptera/Hymenoptera | Chironomus sp. (Non-biting midge) | 2.66 | 1.25 |
The mite Tetranychus kanzawai was the unambiguous dominant pest in both systems. However, its RA was over 7.5 percentage points lower under the solar panel. Other species, like certain spiders and hoverflies, showed higher relative abundances in the solar panel environment.
Temporal Dynamics of Community Diversity Indices
The monthly analysis of biodiversity indices revealed distinct temporal patterns and consistent differences between the two cultivation systems. The Shannon-Wiener Diversity Index (H’) and Pielou’s Evenness Index (J’) followed a similar “V-shaped” trajectory over the growing season in both systems, initially decreasing to a mid-season low before rising sharply in the late season (August). Crucially, for most of the year, both H’ and J’ were higher in the P. kingianum crop under the solar panel compared to the open field. The lowest diversity under the panel occurred in June (H’ = 0.46), whereas the open field reached its minimum later, in July (H’ = 0.26).
The Simpson’s Dominance Index (C) exhibited the inverse pattern, increasing to a mid-season peak before declining. As expected from the higher mite dominance, the open-field system generally showed a higher dominance index than the solar panel system. The Margalef’s Richness Index (D) showed a more complex pattern: it peaked in May for both systems but was consistently higher in the open field for most months except late season, indicating a generally higher rate of species arrival in the unshaded environment, though many were rare.
Community Similarity and Natural Enemy to Pest Ratio
Despite the differences in diversity and abundance, the overall species composition between the two systems was highly similar. The Jaccard similarity coefficient was calculated at \( q = 0.84 \), indicating “very similar” communities. This high similarity was driven by the natural enemy and neutral insect sub-communities, which had a near-perfect overlap (q ≈ 1.00). The pest sub-communities were less similar (q = 0.67), reflecting the absence or rarity of certain pest species in one system or the other.
A more dynamic and ecologically significant metric is the Natural Enemy to Pest ratio (NE:P). The temporal trend of this ratio, based on individual counts, diverged between systems. Under the solar panel, the NE:P ratio was very low in early season, hitting a minimum of 2.77% in May, before undergoing a dramatic increase, surpassing the open-field ratio and reaching a peak of 96.00% in August. In the open field, the NE:P ratio started higher, fluctuated, and also peaked in August, but at a lower value of 89.47%. This suggests that while both systems saw a late-season surge in natural enemy activity relative to pests, the effect was more pronounced under the photovoltaic array.
| Month | NE:P Ratio – Solar Panel | NE:P Ratio – Open Field |
|---|---|---|
| April | 12.50% | 8.33% |
| May | 2.77% | 5.26% |
| June | 7.14% | 3.45% |
| July | 25.00% | 1.89% |
| August | 96.00% | 89.47% |
| September | 66.67% | 50.00% |
Population Dynamics of the Dominant Pest: Tetranychus kanzawai
The population trajectory of the spider mite Tetranychus kanzawai was of paramount importance due to its status as the community dominant. Both systems exhibited a clear unimodal population curve over the season. However, key differences were evident. Statistical analysis confirmed that the mean population density (individuals per plant) was significantly lower on plants growing under the solar panel compared to those in the open field (\(F = 3.553, P = 0.024\)). Furthermore, the timing of the population peak differed. Under the moderated conditions of the solar panel, the mite population peaked earlier, in May, at a mean density of 6.77 ± 2.25 per plant. In the harsh, exposed open-field environment, the population peaked later, in July, at a much higher mean density of 14.75 ± 2.37 per plant. This indicates that the solar panel environment not only suppressed the overall abundance of this key pest but also altered its seasonal phenology.
Discussion: The Ecological Influence of the Solar Panel Environment
The findings of this study demonstrate that the installation of a solar panel array over a crop of Polygonatum kingianum does not merely create shade; it fundamentally alters the ecological theater for the associated arthropod community. The observed higher diversity (H’) and evenness (J’), coupled with a lower dominance index (C) under the panels, point towards a more balanced and complex community structure. Several interrelated factors inherent to the solar panel environment likely contribute to this phenomenon.
Firstly, the physical structure of the photovoltaic system creates a heterogeneous habitat. The solar panel mounts, support frames, and the underside of the panels themselves provide additional niches for shelter, overwintering, and web-building for predators like spiders. This increased habitat complexity can support a greater variety of natural enemy species, which is reflected in their higher relative abundance under the panels. Secondly, the modified microclimate is crucial. The solar panels reduce direct solar radiation, buffer temperature extremes (lowering maximum daytime temperatures), and likely increase ambient humidity near the crop canopy. This more stable and moderated environment may be less stressful for a wider range of arthropods, including some natural enemies that are sensitive to desiccation or heat.
The most significant result pertains to the suppression of the dominant pest, Tetranychus kanzawai. Spider mites are notorious for their rapid population growth under hot, dry, and sunny conditions. The microclimate under the solar panel—characterized by reduced temperature, lower UV radiation intensity, and higher humidity—is suboptimal for their explosive reproduction. This directly explains the significantly lower population density and the earlier, less severe peak under the panels. The delayed and higher peak in the open field is classic of mite outbreaks in stressful, exposed environments. Furthermore, the enhanced natural enemy community under the solar panel, particularly spiders and predatory insects that benefit from the habitat structure, likely imposes greater biotic control pressure on the mite population, contributing to its suppression.
An intriguing aspect for future research is the potential role of polarized light pollution. Solar panels, especially when new or with specific anti-reflective coatings, can become super-stimulating sources of polarized light, attracting and trapping certain aquatic insects like mayflies or even some terrestrial species. This “ecological trap” effect, while potentially detrimental to those insect populations, may indirectly reduce the number of generalist herbivores or nuisance insects in the immediate area, contributing to a different community assemblage. The high similarity in neutral insect groups suggests this effect may not be pronounced for all flying insects in this specific system, but it remains a unique selective pressure of the photovoltaic environment.
The temporal shift in the NE:P ratio is particularly telling. The extremely low ratio in May under the solar panel coincides with the mite’s population peak there, suggesting the pest population initially outpaces natural enemy response. However, the subsequent dramatic increase in the NE:P ratio, ultimately exceeding that of the open field, indicates a strong and potentially more effective numerical or aggregative response by natural enemies later in the season within the solar panel habitat. This delayed but powerful regulatory response is a hallmark of a resilient ecosystem.
Conclusion and Implications for Agrivoltaic Management
In conclusion, this comprehensive year-long study provides clear evidence that cultivating Polygonatum kingianum under a solar panel array fosters a more diverse and evenly structured arthropod community compared to traditional open-field cultivation. The photovoltaic environment acts as an ecological modulator, mitigating the extreme abiotic conditions that favor pest outbreaks, specifically suppressing the population of the dominant spider mite, Tetranychus kanzawai. Simultaneously, it appears to enhance conditions for natural enemies, leading to a more favorable balance between pests and their predators later in the growing season.
These findings have direct practical implications for the sustainable management of agrivoltaic systems, particularly for high-value shade-tolerant crops. The reduced pest pressure under solar panels suggests a potential for lower reliance on chemical pesticides, aligning with ecological intensification goals. The solar panel infrastructure itself can be viewed not just as an energy generator but as a passive ecological engineering tool that modifies habitat and microclimate to promote beneficial agroecosystem functions. Future research should focus on quantifying the yield and quality impacts of this reduced pest load, investigating the specific mechanisms linking solar panel-induced microclimate to arthropod physiology and behavior, and exploring how different solar panel designs (e.g., bifacial panels, tracking systems) further influence agro-ecological dynamics. The integration of photovoltaic infrastructure with agriculture, therefore, offers a compelling pathway not only for dual land use but also for fostering more biodiverse and resilient farming systems.