Investigating Lobster Byproducts as Soil Amendments for Disease Suppression and Soil Health Improvement in Potato Production

The lack of field research and technical application of shellfish byproducts in agriculture indicates that field research is of most use. However, because of the intrinsic challenges and large scale of field research, preliminary studies took place in the greenhouse located at The University of Maine in Orono. Following the greenhouse trial, a field trial took place. Lessons learned in the greenhouse trial aided in determining the best direction for field application, experimental design, and rates.

1. Effects of lobster shell meal (LSM) on soilborne pathogens and potato plants, Preliminary greenhouse studies. In the University of Maine greenhouse, the potato variety ‘Shepody’ was planted in a mix of garden soil and potting soil (1:1) in one-gallon pots in March and July of 2022. ‘Shepody’ was selected because it is susceptible to both V. dahliae and S. scabies. The soil was inoculated with either Verticillium dahliae or Streptomyces scabies inoculums, the causal agents of potato early dying (PED) and common scab, respectively, and non-inoculated soil was used for a non-diseased control. Prior to planting, inoculum for S. scabies was made by growing the bacteria in vermiculite mixed with liquid medium (Say’s solution: 40 g sucrose, 2.4 g asparagine, 1.2 g K2HPO4, 20 g yeast, 1 L sterile water). The bacteria and vermiculite combination was added to the soil in pots at a rate of 1:10 inoculum to soil, and mixed thoroughly. The final inoculum concentration was 105 colony forming units (CFU) per gram of soil. For V. dahliae inoculum preparation, microsclerotia, the rigid overwintering structure of V. dahliae, were used. Twenty microsclerotia per gram of soil were added to pots by suspending microsclerotia in sterile water and adding to vermiculite. Lobster shells collected from local processors were dried and ground using a grain mill to produce a fine powder, which was used as the standard LSM. Soil samples were collected and stored in a -80°C freezer until further processing to represent soil before adding potato seed pieces, pathogen inoculum, or LSM. LSM was then blended into the soil at various rates (0%, 2%, and 8% w/w). These ranges were selected based on previous studies and to help narrow in on appropriate rates for subsequent work. There were initially five pots per treatment within each pathogen inoculation category (V. dahliae, S. scabies, or uninoculated) and both planting groups. In the first planting group, the pots were immediately planted with sprouting seed tubers, and the second group remained without tubers for the first three months before planting with sprouting seed tubers. This gave an indication of how microbial communities were changing without the addition of plant tissues, but with the addition of LSM alone, and model a fall or early spring amendment timeline. This also gave time for microbial communities to respond to the addition of LSM. Pots were set up and randomly arranged in the greenhouse bench. Plants were watered as needed throughout their growth and until the plants began to naturally senesce around 3 months after planting.

            Plant heights were measured over the growing season until the natural death of the plants occurred, after which, plants were harvested. To harvest the plants, plants were cut at the soil line and all plant tissue were placed into paper bags to dry in a heated room for approximately one week, or until the plant matter was fully desiccated. Desiccated plant matter was weighed and recorded as aboveground biomass. A few days after the removal of aboveground plant tissue, tubers were removed from the soil, the surfaces were cleaned of any residual soil, and they were processed immediately. The total weight of the tubers, tuber numbers, and disease symptoms were recorded. Disease incidence described the presence or absence of symptomology and disease severity was recorded in various ways depending on the disease. To determine PED infection symptom severity, tubers were cut in half longitudinally, the percent of the vascular ring which was discolored was recorded. To determine the severity of symptoms of infection by S. scabies or Rhizoctonia solani, the causal agent black scurf, the percent of the tuber surface which was covered by scabs or sclerotia, respectively, was recorded. Indigenous populations of R. solani produced disease symptoms in tuber without inoculating the soils, and this was recorded as well.

            A follow-up greenhouse study was performed in March and July of 2023. Six treatments with five replications were established, building upon challenges associated with the first greenhouse trial. Fertilizer was added as a treatment factor and inoculum was combined in equal parts from R. solani and V. dahliae to ensure proper inoculum levels to cause disease, and is described below. Treatments were once again prepared twice to simulate a spring and fall amendment timeline as described in greenhouse trial 1. To best compare to the field trial, rates of inoculum, fertilizer, and LSM were all as similar as possible to those used in the field trial. The rate of inoculum was 20ml/foot, and as the pots were 8” in diameter, a rate of 14ml/pot was used. Virgin grain was added at the same rate to non-inoculated pots. The rate of fertilizer (14-14-14) was applied at a rate of 1100lbs/acre, typically incorporated at a depth of 12”, so each pot received 2.99grams of fertilizer. The rate of LSM was described by Coast of Maine to be applied at a rate of 5lbs/100ft2. In the field, the LSM was incorporated to a depth of approximately 4”, resulting in a rate of approximately 1.7% w/w. Data was collected in the same fashion as in the first greenhouse study, with the exception of plant height, which was not collected.

2. Field Studies. Following the greenhouse studies, two field trials took place in Presque Isle at the Aroostook Research Farm, Maine. Potato ‘Russet Burbank’ was planted in three rows within 25-foot-long plots with four replicates. Soil treatments included unamended soil, unamended soil inoculated with V. dahliae, LSM amendment alone, LSM amendment with V. dahliae inoculum added, soil fumigation with metam sodium at 50 gal/A amended with LSM and inoculated with V. dahliae, and soil fumigation with metam sodium at 50 gal/A inoculated with V. dahliae. These additional treatments gave a comparison to typical potato agricultural practices. The fumigation was performed in the fall of 2022 under favorable weather conditions. These trials took place in adjacent fields which have not had exposure to LSM previously and had been planted in oats for at least one year prior to these trials. Lobster shell amendments were applied in the fall of 2022 before planting seed tubers in the spring. The rate of LSM as an amendment was informed by industry standards provided by Coast of Maine, a compost and soil amendment company. The rate of LSM for the field trial was consequently 5lbs per 100 ft² (approximately 0.35-0.7% of the soil % w/w). LSM was also procured from Coastal Chitin, LLC, a local processor, which aided in the industry collaborations and simplified the process of preparing large volumes of LSM. This prepared LSM was received pre-dried, pre-ground, and ready to use. The field was maintained during the growing season using standard operational practices, including fertilizer application and pest control. The fields were irrigated with rainwater only. This field evaluation also served as a demonstration plot, because of its location at the research farm, and was shared with farmers at the annual summer meeting. A second field trial was established in the spring of 2023 to simulate a spring amendment timeline. However, because of the difficulty of fumigation in the spring, this factor was not included in this trial. Consequently, the experimental design was simpler and consisted of four treatments: unamended and not inoculated, unamended and inoculated, amended with LSM only, and amended with LSM and inoculated. Soil was sampled at planting, mid-season, and before harvest.

     At harvest, potato yield, disease incidence and severity of PED were evaluated to determine the impacts of these treatments. Additionally, incidences of hollow heart, a physiological condition, were noted if tubers had hollowing in their core. Data was analyzed using similar methods for both greenhouse field studies. Statistical analyses were performed using the R package Agricolae. Effects of treatments on disease incidence, disease severity, and plant biomass were determined using analysis of variance and Fisher’s Least Significant Differences tests.

3. Soil microbial community dynamics analysis. In the trials described in objective 1, field soil samples were collected in the fall-amended trial after fumigation, then in both before planting, in the middle of the growing season, and before harvest. Each soil sample consisted of a minimum of ten cores randomly selected in the plot which were homogenized in a gallon size Ziploc bag to give a representative composite sample of the plot. This helped us catalog changes to the microbial community over time. Soil samples were sieved and stored at -80°C until processed. DNA was extracted using 0.25 grams of soil and a DNEasy PowerSoil Pro Kit according to the provided instruction. DNA was quantified, and purity was ensured before sending for sequencing analysis at The University of Michigan Genomic Sequencing Facility. Samples were processed using NextGen sequencing on an Illumina MiSeq to return sequencing reads which were used to identify fungal and bacterial community members. Amplicons of the samples were generated using the V3-V4 hypervariable regions of 16S rRNA region to determine the presence of various bacteria present in the soil samples. For fungal populations, the ITS2 rDNA region was amplified (van Horn et al. 2021). 

     Raw data of Illumina sequencing was processed to filter sequences, remove chimeras and other unwanted contaminating sequences, and clustered into amplicon sequence variants (ASVs) at 97% sequence similarity, which was analyzed using the DADA2 pipeline. R packages such as Phyloseq, DeSeq2, and Microbiome were used to quantify and visualize community differences between samples and treatments. Microbial diversity was measured by a series of ASV-based analyses of alpha and beta diversity (Rosenzweig et al. 2012; Rajendhran et al. 2011). Community differences were visualized using ordinations of Bray-Curtis and Jaccard dissimilarity (van Horn et al. 2021; Callahan et al. 2016). Core members which were unique to treatments were also called and described, should they match a previously described organism to genus and species (Callahan et al. 2016). Differential abundance of microbes was performed using DeSeq2 to determine if specific ASVs or phyla were enhanced or reduced following our treatments (Love et. al 2016).