Practical Approaches to Microbial Community Analyses for Production Agriculture in the Southern Great Plains

Final report for OS18-115

Project Type: On-Farm Research
Funds awarded in 2018: $9,745.00
Projected End Date: 09/30/2019
Grant Recipient: Southern Plains Climate Hub
Region: Southern
State: Oklahoma
Principal Investigator:
Co-Investigators:
Expand All

Project Information

Abstract:

Soil health monitoring is an integral part of sustainable agriculture (Doran and Zeiss, 2000). Unfortunately, most current tests for determining soil health rely predominately on chemical and physical properties, excluding biological properties, thus resulting in incomplete characterizations (Cardoso et al., 2013). Phospholipid fatty acid (PLFA) analyses are a widely available measure of soil biological properties, and are unique in that they indicate not only microbial biomass, but also community structure. However, many producers in the Southern Plains have indicated that they do not fully understand the relationship between PLFA results, soil health, and management decisions and consequently are reluctant to spend money on PLFA analyses. Furthermore, it is unclear to what extent PLFA sample size affects the accuracy of the results with regard to characterization of the microbial community of an entire field. Because the sampling effort may require a substantial investment of time and money, it is important to optimize the balance between information return and resources devoted to sampling.

This study will improve our understanding of how climate, soil, and management impact soil biological communities in Southern Plains agricultural systems, and enhance producer understanding of PLFA analyses and applications in the Southern Plains.

Project Objectives:
  • Identify the best practical approach for a landowner to assess soil health via PLFA;
  • Demonstrate the utility of PLFA analyses to a range of producers across the Southern Plains, thus enabling them to assess the costs and benefits of using the tests as part of their own soil health monitoring processes.

Cooperators

Click linked name(s) to expand/collapse or show everyone's info
  • Caitlin Rottler (Researcher)
  • Clay Pope (Educator)
  • Dr. Gretchen Sassenrath (Educator and Researcher)
  • Annie Pearson (Educator)
  • Scott Robertson (Researcher)

Research

Materials and methods:

Three sites, which had been used for previous studies, were selected at Parsons, KS; Loyal, OK; and Ardmore, OK. Fields with similar soil types were selected at each location. Fields varied in size from 12.1 ha to 40.5 ha, were used to grow a variety of crops with no-till management, and had covers at the OK sites. Commercial laboratories providing PFLA analysis recommend submission of a composite sample with 10-15 subsamples per field, therefore 30 cores were obtained per field. Cores were not combined and each was analyzed separately, so that mean values for multiple combinations and sample numbers (2 to 30) could be calculated. Cores were taken to a depth of 15 cm (6in) with a 2.5 cm (1 in) diameter probe. Soil cores were freeze dried, ground, and PLFA type and abundance were determined using the method of Buyer and Sasser (Appl. Soil Ecol. 61:127-130) to identify microbial community structure based on microbial functional groups. Functional groups included in the analysis included actinomycetes, arbuscular mycorrhizal fungi (AM fungi), eukaryotes, gram negative bacteria, gram positive bacteria, anaerobes, and non-AM fungi.

To address the question of how many soil cores were sufficient to characterize the field, mean abundance of each microbial functional group at each site was determined for multiple sets of 15 randomly selected cores for each field. The relationship between sampling densities ranging from 0.74 samples/hectare to 2.48 samples/ha and variation in 15-core samples were also quantified. To better understand the relationship between the 15-core high and low values and the mean of each functional group compared to 30 cores/field, the difference between the high and low values for each 15-core sample was calculated and then expressed as a percentage of the 30-core mean for the field. To better understand the response of these proportional differences to sampling density and test for significance of the relationships, a regression analyses was run by functional type with ɑ=0.10.

Research results and discussion:

There was a significant negative relationship between proportional difference and increasing sampling density, as predicted, for Eukaryotes (p=0.04) and AM Fungi (p=0.07) (Figure 1). No sampling density at which the amount of variation among cores dropped below 40% of the average could be identified for any of the functional groups. For the functional groups with the greatest difference between highest and lowest abundances (Table 1), the number was likely below 2.7 cores/ha (1 core/acre). However, the quality of the results did increase with sampling density, which implies that the ideal sample size for many fields exceeds 10-15 cores.

Results showed that for some functional groups, even increasing the sampling density to 2.7/hectare was not sufficient to decrease the proportional difference in their abundance across cores below 40%. For some farmers, 2.7 cores/hectare may be attainable, and is likely to give a more realistic interpretation of their fields’ microbial community structure than a lower sampling density. On average, with a team of two people, collecting 30 cores required approximately 2 hours of work. To sample a ~32 ha (80 acre) field at a density of 2.7 cores/ha would take considerably longer. Furthermore, microbial community analyses are typically used in a comparative manner, rather than as stand-alone samples from a single time-point. In this context, it is most important that any variation resulting from sampling effort be maintained for subsequent sample collections, regardless of the likely accuracy of the initial collection. Results suggest that even a decrease from 2.7 cores/ha to 0.7 cores/ ha caused an increase from variation that was ~40% of the mean for the group with the highest amount of variation to a range that was over 110% of the mean for the same group (eukaryotes). It is important to note, however, that these results represent a sample size of 6 fields from southern KS to southern OK, and that past research has shown that the similarity of microbial communities across different spatial scales can depend heavily on local conditions and land use (Saetre and Bååth, Soil Biol. Biochem. 32(7): 909–917; Franklin and Mills, FEMS Microbiol. Ecol. 44(3): 335–346 and Soil Biol. Biochem. 41(9): 1833–1840; Green and Bohannan, Trends Ecol. Evol. 21(9): 501–507). This analysis should therefore be tested in other agricultural systems and with more varied sampling densities before they are assumed to hold true in other regions and systems.

Participation Summary
5 Farmers participating in research

Educational & Outreach Activities

1 Journal articles
1 Workshop field days

Participation Summary:

1 Farmers participated
2 Ag professionals participated
Education/outreach description:

Educational and outreach activities are ongoing, as the results of this research, as well as information about phospholipid fatty acids, soil microbial communities, and soil health are being incorporated into curriculum associated with the Redlands Community College soil health education trailer and rainfall simulator. These presentations are done regularly throughout the year, at trade shows and conferences. The number of individuals they reach varies widely in number, as well as in age and vocation.

The project has also been shared with producers who attended a soil health field day at Redlands Community College in September of 2019. Approximately 60 individuals were in attendance.

A manuscript for a peer-reviewed publication from the project is currently in the process of being revised and resubmitted to Agricultural and Environmental Letters, which is an open-access journal from the American Society of Agronomy.

Learning Outcomes

Key changes:
  • Learning outcomes were not formally measured as part of this project, due to the institutional ethics requirements of doing so. However, those who participated in the project and those who have participated in education and outreach programs that incorporate the project have gained knowledge, or at the very least, awareness, of soil microbial communities, their contribution to soil health, and the practical aspects of collecting reliable samples to test for them on farm fields.

Project Outcomes

Project outcomes:

Publication of project findings will help guide recommendations for sample number requirements to ensure best analytical results.   

Recommendations:

Study results indicate a need for greater testing in a range of agricultural systems to determine the number of subsamples to include for PFLA analysis of soil community structure.   

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.