Examining the utility of black soldier fly larvae composting on urban farms

Larvae Selection

Black soldier fly larvae (BSF, BSFL) that were roughly 6 days old were ordered from Fluker’s farms, Symton, the Critter Depot, and Josh’s Frogs, in staggered purchases to establish a colony in our laboratory. Larvae arrived in a moistened peat moss substrate, which was included in the first round of composting. Larvae reared for nutritional trials were stored in 4 x 9.75 x 12.25-in, 2-gallon containers; this is also where digestate was produced. Bulk digestate was produced in 18-gallon tote BSF rearing bins using food waste as the primary input with the addition of other waste streams: plant waste, spent worts from local breweries, and sheep manure.

Larval Rearing

Young larvae were stored in an environmental growth chamber with the temperature set to 25C. From there, larvae were transferred to  bins in a room within medium to large sized bug dorms to prevent escapees, underneath LED and fluorescent light. These lights are recommended for egg fertilization within indoor requirements (Nakamura et al., 2016). However, we found moving adult flies into an area with natural lighting encouraged increased mating and oviposition; this was our preferred rearing method. The adult stage of the colony was, therefore, maintained in a greenhouse.

Fluted cardboard substrates were provided to adult flies for oviposition so eggs could be collected easily, monitored, and then added to digestate production. These cardboard egg cards were placed in the environmental growth chamber at 25C until larvae reached 6 days old to ensure larval fitness. BSF adults are attracted to fluted cardboard and prefer oviposition substrates above moistened food with BSFL present to guarantee their offspring can eat after they hatch as neonates. Because of this quality we implemented ‘food traps,’ bowls with mesh lids that contain fresh food and young larvae to encourage oviposition.

Experiment larvae were fed 163 mg (about the weight of five grains of rice) of food per larva per day (mg/larva/day) (Parra Paz et al., 2015). We chose to go on the higher end of the 95 – 163 mg of food based of the heat production of BSF larvae along with feed refusal going through an aerated composting process of its own (McEachern, 2018). Due to human error, the first feeding trial for the larvae resulted in feeding rates of 23-28 mg/larvae/day. The second and third trials were closely monitored to prevent this from occurring again and were fed at the ideal 163 mg/larva/day rate.

Measurements taken from larvae include length (mm) and weight (g). Due to the small size off larvae, 10 were selected and weighed together to get an average. For length, larvae were placed in a freezer and removed 15 minutes later so there would be limited movement. These measurements were taken as proxy of how nutritional each food stream was for larvae and if any effects in development could be monitored.

Food stream acquisition

Food waste/processed food waste was obtained from produce harvested at Meig’s Purdue Agricultural Center (Throckmorton, IN) and/or from table scraps from lab members. Plant waste/organic food waste was gathered at Meig’s from weeding plots, from scrapped flats of seedlings, or from local grocery stores with bad produce. Sheep manure was collected from the Animal Sciences Research and Education Center (ASREC; West Lafayette, IN), a local Purdue associated farm that raises multiple herds of livestock. Spent worts were obtained from local breweries.

Digestate

Following BSF pupation, vacant digestate was harvested and sent to A&L Laboratories (Fort Wayne, IN) for compost analyses. These analyses looked at micro and macro nutrients, organic matter, pH, and C:N ratios. We analyzed all four BSF treatments (food waste, plant waste, spent worts, and biosolids) in addition to a common manure compost (Black Kow), a low nutrient potting mix, and a pooled sample of BSF digestate.

Field location

The field application experiment was conducted at the Meigs Horticultural Research Farm (40.28783289649686, -86.88420847116431), part of the Throckmorton Purdue Agricultural Center located in Lafayette, IN. All three focal crops were grown in a field that was planted with soybean in the previous year and had lime applied in the spring of 2020. All crops were grown in raised beds on three-foot centers with 6.5-foot spacing between the rows. Plastic drip irrigation was installed, and plastic mulch covered the beds for the tomato and cucumber crops. Beds were oriented in a north-south direction. Each crop was planted into two rows and blocked by crop type.

Crop Selection

Three focal crops were selected based on their popularity on small farms, representation of broad crop families (Solanaceae, Cucurbitaceae, Apiaceae), and representing fruit and root crops. Seeds were obtained from commercial seed suppliers and included: Tyria cv. cucumbers, Celebrity cv. tomato and Naval cv. carrot from Johnny’s Selected Seeds (Winslow, ME) and Short N’ Sweet cv. carrot from Burpee (Warminster Township, PA).  

 Source of Amendments

Digestate produced from BSF was applied to soils shortly after transplant, with one additional application for tomatoes and cucumbers after flowering. Digestate was harvested from single food streams: food waste, plant waste, spent worts, and sheep manure which we refer to as biosolids, along with a pooled sample of these food streams. These samples were sent to A&L Great Lakes Laboratories (Fort Wayne, IN) for package C6 analyses along with a commercially available composted cow manure known as Black Kow and a low nutrient potting mix from Midwest Trading (Maple Park, IL; 80/20 mix). The C6 compost package includes solids, nitrogen, phosphorus, potassium, sulfur, calcium, magnesium, sodium, iron, aluminum, manganese, copper, zinc, organic matter, carbon, C:N ratio, pH, and soluble salts. Due to BSF rearing difficulties within the lab setting we ultimately decided on one pooled waste stream to feed to the larvae; this food stream was a mixture of food waste, plant waste, manure, and spent worts. Resulting crops had stem diameter, height, and yield measurements taken biweekly, along with pest and pathogen surveys weekly. 

Planting and treatment applications

Carrots were directly seeded into beds using a Jang Clean Seeder push seeder in single rows on May-14, 2021. Seed spacing was set to 2.5 inches, however, due to the small size of the seeds they were closer. This resulted in a row length of 16.66 ft (~5.07 meters) for the variety Short N’ Sweet and 30.66 ft (~9.34 meters) for the Naval variety (see figure 1). Soil amendments were added to subplots on Jun-22^; both treatments were applied to one third of the carrot rows with the last third being the control treatment. 3055 ± 100 grams of soil amendment was applied to the Naval carrots and 2516 ± 100 grams of soil amendment was applied to Short N’ Sweet carrots. Differences in amount of digestate applied can be attributed to seed size altering spacing and density. 

Figure 1

Tomatoes and cucumbers were seeded on Apr-29 into 72-cell trays filled with potting mix (DETAILS FROM THE BAG OF SOIL) and germinated under mist irrigation in a greenhouse at Throckmorton Purdue Agriculture Center. Seedlings were transplanted into the field on May-24 using a rainflo model 1670 water wheel transplanter. Both crops were transplanted in parallel, adjacent rows starting on the Northwest side of the plot. Seedlings were transplanted into single rows; 24-inch spacing for cucumber plants and 48-inch spacing for tomato plants. This resulted in 35 cucumber plants and 28 tomato plants. T-posts were installed along with mesh netting to create a trellis for cucumber plants to support them, tomato plants were provided with tomato cages.

179.70 grams ± 15 grams of BSF and BK soil amendments were added to cucumber blocks C1, C2, C6, C7, and tomato blocks T1, T2, T5, and T6 on May-25. The remaining cucumber and tomato blocks (C3, C4, C5, T3, and T4) received soil amendment applications on Jun-4 (figure 1). All blocks received an additional application of soil amendment treatments applied to tomatoes and cucumbers on Jul-14 after plants had flowered and produced small fruits of roughly the same size as the first addition of soil amendments (179.7 grams). One block was made up of 5 plants within one of three treatments, only 3 plants were harvested from each treatment where available.

Data Collection

Pest and pathogen surveys were completed weekly during the summer of 2021. Subplots including 3 out 5 plants in each block were surveyed by examining the whole plant for pests and checking three leaves for aphids. Plant measurements were taken biweekly after the first three weeks. We measured stem diameter in millimeters using an automatic caliper. Plant height was measured as well in centimeters, plant height being defined as from the stem base to the apical meristem. Fruits were harvested and weighed from cucumber plants twice; Jul-14 and Jul-23. Tomato plant fruits were harvested and weighed on Aug-13, Aug-27, and Sept-11.

 Cucumber plants were harvested on Jul-30 in 2021, to examine dried above and below ground biomass. Fresh weight was not taken for these plants due to bacterial wilt drying some parts of the plants out. Carrots were harvested on Aug-4, combined fresh weights of roots and shoots were collected along with dried above and below ground biomass. Tomato plants were harvested on Aug-17 and -24; fresh weight was taken in kilograms on Aug-17 and in pounds on the Aug-24 due to access of scales. Plant material was dried using a drying oven for 1 to 2 weeks at 75C. To ensure plant material was completely void of moisture, 5 bags were weighed randomly, dried for an additional 24 hours, and weighed again. When a stable weight was recorded during the 24-hr interval plants were deemed dry, removed from the oven and weighed.

Soil samples were collected on Oct-13 in 2021, from the B1 plot at the Meig’s Horticultural Research Farm. Samples were taken using a soil corer and sent to A&L Great Lakes Laboratories (Fort Wayne, IN) for soil analysis. We ordered the analysis packages S1 and S3 which includes organic matter, available phosphorus, exchangeable potassium, magnesium, calcium, soil pH, buffer pH, cation exchange capacity, percent base saturation of cation elements along with sulfur, zinc, manganese, iron, copper, and boron. A total of 27 soil samples were sent for analyses taken from the plots outlined in figure 2.

Figure 2

Data Analysis

Compost and soil data were both analyzed using a principal component analysis (PCA) to illuminate any latent variables that could not be measured or identified. Bartlett’s tests were run as well to confirm PCAs were an appropriate statistical analysis. Crop measurements, including above and below ground biomass, stem diameter, and growth, were measured for significant interactions using an ANOVA in addition to Tukey HSD to confirm significance. The statistical software used to analyze data generated from these methods was JMP Pro 16.0.