The Impacts of Native Plant Diversity on Native Bee Development and Soil Health

Soil Nutrient Analysis

Soil samples were collected in the Spring (March/April) and Fall (September/October) of 2020, 2021, and 2022 in order to analyze the nutrient content and microbial communities in the soil. Soil samples were sent to the Marianna Soil Test Laboratory in Marianna, AR for nutrient analysis. The analysis included the concentrations of phosphorus, potassium, zinc, sulfate, nitrate, magnesium, and iron in the soil, as well as the pH and the estimated cation exchange capacity (CEC). The results of the pre-planting and post-planting soil nutrients were compared each year, as nutrients can fluctuate by season. Results were analyzed using a pooled t-test. The results from 2020 and 2021 are shown here, but not 2022, as the 2022 post-planting nutrient analysis reports have not yet been received from the test laboratory.

The pH was significantly different by season for both 2020 (p=0.0015) and 2021 (p=0.0102). However, pH increased, becoming closer to neutral, during the post-planting soil collection in 2020, whereas it decreased, becoming more acidic in the post-planting of 2021 (Fig. 2). pH fluctuated between pH 6 and 7.5 for all soil collection results. Phosphorus concentration in the soil decreased from pre-planting to post-planting collections, but this decrease was only significant in 2020 (p=0.0122). Potassium concentration also decreased in the post-planting collections, though this was only significant in 2021 (0.0024). Nitrate (NO₃) concentration in the post-planting soil increased slightly in 2020 and significantly in 2021 (p=0.0369). Nitrate levels were also significantly higher for both pre-planting and post-planting soil in 2021 compared to 2020 (Fig. 2). Before the buckwheat and wildflower plots were established in 2020, the area had been a frequently-mowed, but unplanted, grassy field. In 2021, the field plots had all been planted for a growing season with buckwheat or wildflowers. This could account for the overall increase in nitrate concentrations in 2021; however, soil samples taken from a grassy field adjacent to the field plots also showed higher nitrate concentrations in 2021 than in 2020. This would suggest that the planting activity was not responsible for causing the increase in nitrates from 2020 to 2021, and that instead weather conditions, natural variation, or equipment failures could account for this increase.

The post-planting results of the buckwheat, honey bee flower mix, and eastern flower mix groups were then compared to samples taken from nearby grassy fields at the research site. Post-planting samples were analyzed using a one-way ANOVA, by treatment group. In the case of the soil pH and nitrate levels in 2020, a nonparametric Kruskal-Wallis H test was performed instead. There was no significant effect of flower planting type on post-planting soil pH for either 2020 or 2021. In all plots, pH was neutral or mildly acidic, ranging from 6.1-7.4. Post-planting nitrate levels were also not significantly affected by flower planting type. In 2020, the eastern wildflower planting group and the honey bee flower planting groups had higher nitrate concentration averages than both the buckwheat and grassy field groups, though this difference was not significant. In 2021, buckwheat had a lower nitrate concentration means than the honey bee wildflowers, eastern wildflowers, and grassy field, but it was not significant.

In 2020, there was a significant effect of flower planting on post-planting potassium levels, though this was only seen in pairwise comparison between the grassy field and buckwheat groups (p=0.026, Fig. 3). The grassy field has significantly higher potassium levels than the buckwheat plot. Both the eastern flower mix and the honey bee flower mix had average potassium levels higher than the buckwheat and lower than the grassy fields, but these differences were not significant. Both of the wildflower planting groups, the eastern mix and the honey bee mix, had significantly higher post-planting phosphorus concentrations than the grassy field samples (Fig. 3). In 2021, there was no significant difference among treatment groups for either potassium or phosphorus concentrations. For both years, there was a general trend across all treatment groups of phosphorus levels decreasing from the March pre-planting samples to the September post-planting ones. The wildflower groups retained higher phosphorus levels later in the year than the buckwheat and grassy field groups, though these differences were only significant in 2020.

Bee Foraging Activity

The released bees were active and foraging for about 6 weeks in 2021 and 2 weeks in 2022. During this period they were observed 4-6 times per week during the morning (9am), midday (12pm), or late afternoon (6pm) for 2 minute intervals. The cages planted with eastern wildflower mix and honey bee wildflower mix had a significantly higher average of floral visits in 2021 (p<0.0001, Fig. 4) and higher totals of floral visitation over the whole activity period (Table 2) when compared to the buckwheat cages. In 2022, both wildflower mixes had higher averages than buckwheat, though it was only significantly higher for the honey bee wildflower mix plantings (p=0.0277, Fig. 4). Additionally, the activity period and longevity of the adult bees was longer in the wildflower groups than the buckwheat group for both years; bee foraging activity and signs of live bees ended in the buckwheat cages 1-2 weeks before that in the wildflower planted cages.

Osmia lignaria also showed strong feeding preferences for certain wildflowers when given a variety of wildflower options to feed on (Table 2). Lemon bee balm (Monarda citriodora), which was present in the eastern wildflower mix, but also growing wild at the field site and so also present in two of the honey bee mix cages, was most often visited by the bees. Several other wildflowers were commonly visited, including Rudbeckia hirta, Asclepias tuberosa, Heliopsis helianthoides, Coreopsis lanceolata, and Gaillardia aristata.

Foraging activity was also highest in the morning and midday observations, and low in the late afternoon.

Bee Nesting Activity

Female bees began visiting the nest boxes and collecting pollen provisions 7 days following their release in the cages. The first capped tubes were observed 14 days after release. Their trips to the nest box, while carrying mud and pollen, were recorded at 2 minute intervals 3-4 times per week. In 2021, bees in the cages with eastern and honey bee wildflower mix had significantly more visits to the nest boxes than those in the buckwheat planted cages (p=0.003). In 2022, overall nesting activity was low. There were no observed nest box visits for the buckwheat planted cages and only few in the wildflower planted cages (Fig. 5).

The bees were also given the option of building their nests in plastic or paper nesting tubes. The plastic tubes were put out to see if bees would nest in them. The development of the offspring through the larval instars and pupation could then be observed through the transparent tubes. The bees had a strong preference for the paper tubes, however, and would not nest in the plastic when paper was available. The plastic tubes were more slippery and retained more humidity, which might have made them less ideal for mason bee nesting.

Bee Fecundity

Bee fecundity was measured as the following endpoints: number of capped tubes, total number of nest cells, number of pollen provisions, number of larvae, and number of cocoons. Each of the nest cells could either be empty, contain a pollen provision and unhatched egg, contain a dead larvae that never pupated, or contain a cocoon. These were recorded in early 2022 for the adult bees that had been active in summer 2021. The buckwheat cages had a little nesting activity, but no successful nest cell construction, provisioning, or egg laying. Three of the honey bee wildflower mix cages and one of the eastern wildflower mix cages managed to produce live offspring. The number of capped tubes, cells, provisions, larvae, cocoons, and emerged adult offspring were all highest in the honey bee wildflower mix cages, lower in the eastern mix cages, and non-existent in the buckwheat cages, though these differences were not significant (Fig. 6), likely due to the low replication.

There were some nest parasites observed in the nest tubes, as well. The first were Dermestid beetle larvae (possibly Trogoderma sp.), which were found in 6 nest cells of the Cage E (honey bee wildflower mix) nest tubes. The beetle larvae had chewed through 3 cocoons and eaten the pupa inside. Dermestid beetles tend to be scavengers, however, rather than predators, so it is likely that the bee pupae had died prior to the beetle feeding. There were also barklice (Liposceles sp.) found in 3 cells of Cage G (eastern wildflower mix) and 6 cells of Cage I (honey bee wildflower mix). They were found in cells with mold and fungal growth, which was probably their food source. They are more likely opportunistic nest scavengers and fungal feeders, rather than posing any direct harm to the bee offspring. There were also a few jumping spiders in the tubes, but only in unoccupied tubes. There were notably no parasitoid wasps or wasp larvae found in the bee nests or parasitizing the bees. Some wasps (likely Chalcidae) were observed in Fall 2021 on the outside of the bee cages, but they were unable to pass through the screen. The aluminum screen seemed to deter parasitoid wasps of O. lignaria from reaching the nests.

The whole cocoons (those not eaten by beetle larvae) were weighed, transferred to clean cages, and allowed to emerge at room temperature in the spring of 2022. Of the 24 cocoons, 16 emerged as adults, 2 females and 14 males. 7 of these managed to live beyond a week and 3 of them beyond 20 days.

Discussion

With modern agriculture, there has been a rise in more intensive growing systems, including monoculture plantings, increased use of pesticides and fertilizers, and frequent tilling. There has also been an increase in the amount of land used either as rangeland for livestock or fields for crop plantings. These changes have brought with them a great deal of investigation into how these intensive agricultural practices may effect ecosystem health and plant growth. Monoculture plantings without crop rotation, for example, have been shown to cause decreasing yields over time. This is likely due to the depletion of soil nutrients taken up by the crop plants, which over time can interfere with plant growth. Crop producers often deal with these problems either through the implementation of a crop rotation system or through the heavy use of fertilizers, though the use of fertilizers can cause further issues, such as algal blooms in local water supplies. As such, it is important to investigate how different agricultural methods can affect soil health and ways to ameliorate the negative effects of intensive agriculture in order to develop more sustainable agricultural practices.

Soil pH can vary based on time of year, region, and habitat type, and can impact nutrient absorption by plants and species richness of soil microbial communities. The highest diversity of microbial communities are generally found in soils near neutral pH, from ~6.5-7.5. Though some plants prefer highly acidic soils, below pH 5.5, these conditions can reduce  microbe diversity and interfere with calcium and magnesium absorption. In this study, pH was consistent across planting groups and close to neutral in all. Soil was slightly more acidic in March and became closer to neutral in September in 2020, though this trend was not observed in 2021.

Potassium is an essential nutrient for plant growth, involved in enzyme regulation, gas exchange, and ATP production.  Plants with sufficient potassium tend to be hardier in drought condition and more resistant to certain diseases. Potassium levels were highest in the grassy field group. The fields were mowed regularly, but otherwise left undisturbed, without any tilling or planting. After one year of planting, buckwheat had the lowest potassium levels and levels that are considered below the ideal of 131-175 ppm for this region.

Phosphorus is vital for plants for their energy storage, in the form of ATP. Insufficient phosphorus can slow or even stunt plant development. Uptake of phosphorus by plants depends on both the soil pH, with a pH between 6.5 and 7.0 being the optimum, and organic matter content of the soil, which can be affected by the variety of plants grown in the area. Phosphorus was higher in the pre-planting samples than the post-planting ones, and decreased later in the year, though both time of year averages were within the range for adequate phosphorus levels. In the post-planting analysis, the phosphorus levels in the grassy field samples were significantly lower than those in either of the wildflower planting groups. Buckwheat levels were on average lower than the wildflowers and higher than the grassy field, though not significantly. Wildflower plantings could have the potential to maintain higher phosphorus levels when compared to grassy fields and help prevent any phosphorus deficiencies in nearby crop plants, though the range of these effects are still unknown.

Nitrates are a form of usable nitrogen that can be taken up by plant roots and used for plant growth and leaf development. They are also an essential nutrient and insufficient nitrate levels can interfere with growth. Plants can differ in their nitrate requirements. Some vegetables, such as tomatoes, squash, and sweet corn take up higher levels of usable nitrogen, and over time can deplete the levels in the soil. Others, particularly legumes (Family: Fabaceae), can increase levels of available nitrogen through their association with nitrifying Rhizobium bacteria. Levels of nitrates at the research site were low across all treatment groups. They were on average higher in the two wildflower treatments, though these differences were not significant. Both wildflower mixes included members of the Fabaceae family, which could increase the nitrate content of the soil. There were, however, over 20 species of wildflower in each mix, and of those only 1 species in the honey bee mix and 3 in the eastern mix were legumes, so the nitrifying effect could be minor.

Osmia lignaria is a generalist feeder, but still showed strong floral preferences, particularly to Monarda citriodora, Rudbeckia hirta, and Asclepias tuberosa. Many bee species feed on pollen and nectar as larvae, but only on nectar as adults. Female mason bees (Osmia spp.), however, require pollen in the days following their emergence from cocoons, in order to complete the development of their ovaries and reproduce. Typically, they mate, feed on pollen and nectar for 7-10 days, and then begin provisioning and constructing nest cells. This behavior and morphology could help explain the differences in fecundity for the wildflower treatments and the buckwheat monoculture treatment. Female O. lignaria in the buckwheat cages showed minor interest in the nest tubes early in the observation period, but failed to construct any nest cells, collect any provisions, and lay any eggs. Buckwheat yields are improved by insect pollinator visitation, including that by honey bees (Apis mellifera), bumble bees (Bombus spp.), and mason bees (Osmia spp.), and in return buckwheat can provide a quality source of nectar. The pollen quality and protein content, however, may be inadequate for O. lignaria females to complete their ovary development and produce offspring.

In the wildflower planted cages, foraging activity, nesting activity, and fecundity were all higher than in the buckwheat cages. The successful production of offspring suggests that the female O. lignaria were able to find an adequate pollen source among the mixed wildflowers complete ovary development, and lay eggs. Though O. lignaria, and several other generalist feeding bees, are able to forage and survive on certain monoculture plantings, their health, longevity, and fecundity seem to improve with greater floral diversity and the presence of their more preferred flowers. Additionally, though the activity period of emerged adult mason bees is relatively short, often around 4-6 weeks, it is vital for the females to have access to floral blooms throughout the entirety of this period. Wildflower plantings, especially with staggered bloom times, can help ensure the bees will have a food source throughout their foraging and nesting period, which can help improve the number of offspring they can produce.

In the summer of 2022, the field site in Arkansas, as well as many other areas around the region, experienced an extended drought and heat wave. Wildflower diversity was poor 2022, compared to 2021, and bee health and longevity were poor overall, in all treatment groups.

Though their was a clear benefit of the wildflower plantings on bee health, activity, and fecundity when compared to the buckwheat, improvements to the husbandry of the bees could help improve the bee fecundity overall and thus better show the differences in health and reproduction among the treatment groups.  A larger cage size, such as 10'x8'x5' could increase floral availability for each bee in the cage, which could increase the amount of floral provisions and offspring. In 2022, fewer males were released, which seemed to improve female bee activity in the early days following release. Though males are vital for mating and producing female offspring, they can interfere with female bee foraging, as the males can be aggressive in their mating attempts and can knock female bees off of flowers. Mating before release, in a lab setting, and then only releasing female bees into the observation cages could also help improve female foraging and successful reproduction.

Water retention and mold growth was an issue in several of the bee nest boxes, and could have affected bee larval development. Nest cells with mold, sometimes had a pollen provision present, but no larva or developed cocoon. Similarly, nest cells with cocoons did not have any visible mold growth. Removing the nest boxes earlier in the year and storing them in a cool, dry location, such as a 4°C refrigerator, could help reduce moisture damage and mold and improve the bee offspring development into adult bees.

There was also a much higher ratio of male offspring to females. This could be due to unsuccessful mating attempts by the male bees. As with all bees, O. lignaria females are produced from a fertilized egg, while males are produced from an unfertilized egg. If the females were not mated properly by the males, then they would be able to produce male offspring, but not females. It is also possible that moisture retention affected the female nest cells more than the males, causing fewer female offspring to develop. Mason bees lay their female eggs farther back in the tubes and their males closer to the entrance, as the males emerge first from their cocoons. The mold growth was more common farther back in the tubes, and so could have disproportionately affected the female offspring over the males.