A Model for Long-Term, Large-Scale Systems Research Directed Toward Agricultural Sustainability

Final Report for LS98-094

Project Type: Research and Education
Funds awarded in 1998: $256,604.00
Projected End Date: 12/31/2002
Matching Non-Federal Funds: $302,154.00
Region: Southern
State: North Carolina
Principal Investigator:
Dr. J. Paul Mueller
North Carolina State University
Expand All

Project Information


An experiment involving five systems began in September of 1998 with cover crop rye planted over the entire 81 ha site. Systems include three agroecosystems (BMP standard, organic and integrated crop/animal), a successional ecosystem and a plantation forestry system. Preliminary results indicate some benefits to no/ reduced tillage.

Project Objectives:

Our first objective at CEFS was to establish in perpetuity five very diverse ecological and agricultural systems. The next step was to initiate research projects that focused on the following: developing a set of soil quality indicators that can be used to assess management induced changes in agroecosystems, and to study energy and nutrient flows within the plant-soil systems, changes in short-and long-term carbon and nutrient inputs/exports, ecological/biological shifts and economic performance evaluations in these systems over the long-term.
We believe that during the next three years the systems will begin to diverge based on management strategies. Data collected to quantify these changes include: a) soil physical properties (bulk density, aggregate stability, soil water characteristics, pore size distribution, hydraulic conductivity), b) soil quality indices (pH, infiltration, organic matter, inorganic and total N, C02 evolution and microbial biomass), c) soil fauna (bacteria, nematodes, other soil fauna) d) weed populations (distribution, shifts, species, soil seed banks), diseases (pests, beneficials, disease-suppressive soils), insects (beneficials and pests), plant growth development and yield.
A continuing goal of the project is to seek useful methods that facilitate experimental design and analysis of large-scale agricultural and ecological field studies.


The prevailing systems of agricultural production in the United States are struggling with the appropriateness and applicability of new and current technology and its impact on farmers, rural communities, and the environment. Citizens of North Carolina and other states are currently faced with landscape-scale issues some of which include: ground and surface water contamination, soil erosion, declining soil quality, loss of wildlife habitat, and declining rural communities. Only long-term, large-scale interdisciplinary systems research can adequately address these issues.

The Center for Environmental Farming Systems (CEFS), established in 1994 at the North Carolina Department of Agriculture Cherry Farm near Goldsboro is a site where it is possible to study the interactions among management practices, soil processes, farm productivity and environmental quality. An interdisciplinary team of researchers and graduate students from North Carolina State University and North Carolina Agricultural and Technical State University are involved in ongoing research projects at CEFS. Our aim is to evaluate the impact of diverse cropping systems on soil and water quality and to develop profitable agricultural systems that also protect our environment and enhance our rural communities.

Part of any system to evaluate the sustainability of farming practices requires identification of soil attributes to serve as indicators of change and also relate to productivity. Soil quality research focuses on the interactions among soil processes rather than on soil components in isolation. The objective of our soil quality research was to monitor a set of key soil quality indicators (chemical, physical, biological) that reflect current and future systems performance potential using a long-term large scale systems research site. Soil organisms provide many valuable ecosystem services, for example nutrient cycling and biological control of agricultural pests. A common goal of IPM and sustainable agriculture is to preserve and increase the effectiveness of natural enemies. This is accomplished via environmental manipulation (conservation of natural enemies) or, when natural enemy numbers are low, to augment them through propagation and release at an inoculative or inundative rate. A research project in Entomology focused on how the various systems at CEFS affect the occurrence and relative abundance of two types of soil-inhabiting natural enemy of insects, entomopathogenic nematodes and entomopathogenic fungi. We have also examined how different management practices or vegetation types affect soil microbial biomass C and N, labile soil C pools, microbial activities and functional diversity, and N mineralization. Many projects are ongoing but some have been completed or are nearing completion. The following is a summary of results for three such projects.


Click linked name(s) to expand/collapse or show everyone's info
  • Cavell Brownie
  • Nancy Creamer
  • Mike Linker
  • Michael Wagger


Materials and methods:

Experimental Design:

The design is a randomized complete block consisting of three replications of five systems. Approximately 81 ha have been divided, based on intensive soil mapping, into 3 diverse agriculture production systems, a successional ecosystem and a plantation forestry system. Individual plots range in size from 0.81 to 4.1 ha. Blocking was done by soil type, such that all samples in a system were taken from the same diagnostic soil in a given replication. The agricultural production systems include a conventional system using all the best management practices currently used by farmers (BMP), an integrated Crop/Animal system, and an Organic system. The BMP system is split into till and no-till subplots and the Crop Animal is divided into three subplots each representing a different entry point in a 15-year rotation. Implementation of the experiment has been achieved through careful coordination and planning on the part of all researchers, graduate students, and field crew.

Systems Description:

Conventional Cash Cropping System (BMP's, short-rotation cash crop): This system represents a standard for comparison in the sense of a positive control. It is representative of one of the predominant farming systems in North Carolina and other southern states. It is characterized by using management practices commonly used by growers, annual crops, short rotations, and absence of animals. The system is represented by Conventional Till vs. No-Till and a 3 -Year Rotation: Corn - cover crop - Peanuts - cover crop - Cotton - cover crop.

Integrated Crop/Animal System (biologically diverse agroecosystem): Farming systems may benefit from synergies provided by the wise integration of crops and livestock. Perennial species and long crop rotations support a biologically diverse farming system. One feature of this system is to have every field include both cover and pasture crops in the rotation. Another feature of the system is a relatively low-cost fencing and drinking water system for livestock in order to assure that plant nutrients are efficiently recycled through the grazing animal. This system represents a 15-year rotation including - corn, soybean,/small grains (wheat, oat, triticale, barley), cotton, peanut, sweetpotato- pasture (5-years).

Organic Production System: Organic production systems employ unique approaches to nutrient availability, pest control and soil management. These approaches are profoundly different from the predominant conventional agricultural production system. Furthermore, there is a scarcity of information with respect to the long-term effects of such systems. Currently, the focus of this system is to evaluate various strategies of transitioning to organic production systems as we do so within the context of this experiment. The goal is to provide critical information that will help ease the transition of growers converting conventional land to organic production practices. The initial three-year rotation being studied includes: cover-soybean-cover-sweetpotato-cover-wheat/cabbage.

Plantation Forestry /Woodlot (commercially valued forest species): Forest species are an important component of the southern USA landscape. Forestry is not only a thriving commercial enterprise but it also is represented on many farms. Forestry systems sequester nutrients and energy in long-term perennial cycles and offer the possibility for intriguing comparisons with other farming systems. This treatment seeks to maintain the identity of the woodlot as an ecosystem while at the same time applying appropriate silvicultural practices. Species included in the experiment include: cherrybark oak (Quercus falcata var. padgodefolia Ell.), baldcypress [Taxodium distichum (L.) Rich.], green ash (Fraxinus pennsylvanica var. lanceolata (Borkh.) Sarg.) and longleaf pine (Pinus palustris Mill.).

Successional Ecosystem (old field succession): This represents one of the important standards (controls) for the comparison of the environmental impacts among farming systems. As stated by Loomis and Connor, 1992, agricultural fields are mainlined for high productivity as disclimax communities that are prevented from successional return to natural vegetation. The opportunity to study in detail the processes and biological dynamics that take place when land is released from agriculture may be of great importance in beginning to understand the complex interactions involved in agricultural sustainability. Areas selected for this treatment have been allowed to succeed naturally; influenced only by natural processes. Three 1000 m2 sampling modules have been located within each replication of the successional ecosystem where intensive measurements are taking place.

Soil samples:

We chose five stratified random sampling points in each of the plots. These points were physically marked and then geo-referenced using a Trimble backpack GPS unit. These points are stored on a GIS database and are returned to for all subsequent sampling.
At each sample site, a composite of 20, 2.54-cm diameter by 15 cm deep cores were taken adjacent to the plant row on the untracked side in a random manner near the geo-referenced point. This approach will allow for a spatial analysis of the inherent variability associated with these parameters in the most biologically active soil layer.
Baseline soil sampling was conducted in March 1999. There were three additional sampling dates over the course of the 1999 growing season, in spring shortly after planting, mid-season at peak growth, and post harvest sampling in late October. The same sampling cycle was repeated in the 2000 and 2001 growing seasons. By coordinating soil sampling among researches we are able to gain information of the effects of treatment on the interaction among physical, chemical, and biotic components and on the spatial and temporal variation of their interaction.


The same composite soil sample was used for soil fertility analyses (inorganic N, total organic N and C, pH), detection of soil entomopathogens, and determination of microbial biomass C and N, labile soil C pools, and N mineralization. CO2 evolution was used as an indirect measure of soil microbial biomass. Measurements were made using the closed chamber method (Doran and Parkin, 1996) in May, July, and October 1999 and in June, August, and October of 2000. Data was converted to Kg C ha-1 d-1 after Sarrontonio et al. (1996). Infiltration measurements were made at the same time using the same PVC ring. Time for 2.54 cm of deionized water to infiltrate the soil surface was recorded in minutes.
Undisturbed soil cores were collected from the upper 7.5 cm in all plots on March 13, 1999, November 4, 1999 and November 11, 2000. These cores were processed in the laboratory to determine the soil water characteristic, bulk density, total porosity, and saturated hydraulic conductivity. The soil water characteristic was used to determine the soil macroporosity, microporosity and plant available water holding capacity. Measurements of aggregate stability were made on March 13, 2001 and included air-dry and field-moist samples. These samples were also collected from the upper 7.5 cm.

Research results and discussion:

Soil quality/natural resource indices:

The most striking difference at this early stage of the systems development is in the CO2 evolution data. Averaged over all sample dates CO2 evolution in the BMP no till (NT) (mean 417.4 Kg C ha-1 d-1) was 2.3, 1.8 and 1.2 times greater compared to the organic (mean = 196.4), conventional till (CT) (mean = 222.9), and successional system (mean = 247.7). However, seasonal differences (P = 0.0001) were stronger than differences between systems (P = 0.0001). Therefore LSD was used to compare systems at each sample date. There are definitive differences between the BMP subplots and results show higher microbial activity where tillage is reduced and crop residue is left on the surface. This assumption is further supported by the fact that the NT and the successional system followed a similar pattern of CO2 evolution across the growing seasons and were statistically similar on all sampling dates. The CT and organic system maintained relatively steady and lower CO2 evolution rates, and were similar on 4 of the 6 sample dates. Nitrate and ammonium concentrations adequately reflected the fertilization inputs in the systems. Organic C and N contents were very low in all systems and typical of the soil in the southeastern U.S. Both C and N contents did not differ between systems and fluctuated little over the course of the two growing seasons. There were no statistical differences between systems in soil pH but there was a rep effect. Rep A consists of very sandy, somewhat excessively drained soils with an overall average pH of 5.2. Both reps B and C are moderately drained sandy loams and loamy sands with higher clay contents and pH of 5.4 (rep C) and 5.5 (rep B).

We found infiltration rate, as measured in the field using infiltration rings to poorly reflect management-induced changes in the systems. At this stage treatment effects on infiltration time were masked by differences in soil type and soil spatial variability. Within plot infiltration times were extremely variable, therefore statistical analysis was performed on log-transformed data. Infiltration times differed by sampling date (P= 0.0001) but not between systems over the two years.

Soil physical properties:

Analysis of variance was conducted on the baseline data collected on March 13,1999. Results showed a significant replication effect for most variables and a non-significant treatment effect for all variables. The analysis of variance on data collected on all other dates is still pending. Systems with the highest soil bulk density had the lowest macroporosity and therefore lowest hydraulic conductivity. The woodlot and successional systems sometimes opposed this trend. Although having a similar bulk density to the no-till system their hydraulic conductivity was sometimes greater. This discrepancy may be explained by differences in worm and/or insect activity between these systems. Channels may have been created by the boring activity of these organisms thus increasing the ability of these systems to transmit water.

Slight differences in plant available water holding capacity were observed. The crop/animal-integrated systems seemed to slightly retain more available water than other systems.
Trends in the aggregate stability data were consistent between the air-dry and field-moist runs. No differences were found between the BMP subplots or between the BMP and the organic system. Similar stabilities were also obtained between the crop/animal subplots. Aggregates in the successional system were the most stable and the lowest stability was found in the BMP and organic systems.

Soil entomopathogens:

In the Systems Experiment at CEFS we detected three endemic nematode species - Steinernema glaseri, S. carpocapsae, and Heterorhabditis bacteriophora. Even intensively cultivated soils contained entomopathogens. Entomopathogenic nematodes were detected more frequently than entomopathogenic fungi, except in the pasture rotation of the integrated crop/animal system, where we detected mostly the fungus Metarhizium anisopliae. We detected Steinernema glaseri fairly consistently among all systems - regardless of the various levels of physical and chemical disturbance. There were no differences in total numbers of nematodes detected in CT and NT in the BMP system, but H. bacteriophora was more abundant in CT, whereas S. carpocapsae was more abundant in NT. In the forestry/woodlot and successional systems, S. carpocapsae was the most abundant nematode. The detection of entomopathogenic nematodes was very low in pastures in the integrated crop/animal system. This may be a function of low numbers of insect hosts in the system or to soil compaction by cattle grazing there. Abundance of entomopathogenic nematodes was also relatively low in the organic system, where tillage is used for weed control. The survival and persistence of nematodes are probably most dependent on presence of host insects. The increased complexity of the soil environment associated with higher levels of organic matter and the presence of more vegetative cover in no-till may have been one factor determining the higher detection of S. carpocapsae in no-till. This type of environment may provide more alternate hosts for endemic nematodes. Also, under a conventional-till regime, the soil surface has higher temperatures and lower moisture compared to the rest of the soil profile. S. carpocapsae has been categorized as an “ambusher”, which is a nematode that typically remains nearly sedentary at or near the soil surface waiting for mobile surface-adapted host, which may explain its relatively higher sensitivity to tillage. On the other hand, a “cruiser” nematode, one which is typically highly mobile and seeks out relatively sedentary hosts deeper in the soil profile, may not be as affected by tillage since it can move to the more stable environment of the deeper soil profile. S. glaseri and H. bacteriophora are considered to be “cruisers”. This may, in part, explain their apparent lack of sensitivity to tillage in our study.

Changes in short-and long-term carbon and nutrient inputs/exports:

Carbon and N availability were significantly different among the five ecosystems. While available N and potential available N were consistently and significantly lower in grassland, forest and successional systems than in the two agricultural ones over the year, C availability was higher in the grassland and organic soils. High C availability and low N in grasslands significantly affected soil microbes and led to higher microbial biomass C: N ratios and lower N mineralization, suggesting a shift of microbial community composition. Grasses were most effective in reducing soluble N from the soil, tightening N cycling. Although no chemical N fertilizer was applied, organic soils sustained N supply over the growing season. Microbial activities were highest in grassland soils and preliminary data showed that organic and grassland soils had higher functional diversity than conventional agricultural soils. The functional diversity of the microbial community in conventional soils was distinct from that in other ecosystems. Microbial functional diversity was determined by examining the utilization of carbon substrates by microbial communities. Principal component analysis (PCA) of the multivariate data of microbial utilization of C substrates was conducted using the software SPSS. Principal component analysis revealed consistent differences in the functional diversity patterns among the soils. The functional diversity of the conventional soils was distinct from the rest of the soils. The functional diversity of organic soils was similar to that of successional fields and/or forest plantations, suggesting that the microbial community structure in organic soils has shifted away from that of conventional soils.

Participation Summary

Educational & Outreach Activities

Participation Summary:

Education/outreach description:

Several presentations on the project have been made at meetings, grower's schools, and field days at the Center for Environmental Farming Systems. A list of some of these follows:

Desarrollo y Implementacion de un Estudio de Sistemas Agricolas al Gran Escala y Larga Duracion. I Simposio Internacional sobre Ganaderia Agroecologica, La Habana, Cuba, 6-8 December 2001. (Mueller, Barbercheck, Bell, Brownie, Creamer, Hu, King, Linker, Louws, Marra, Raczkowski, Susko, Wagger)

Investigaciones Sobre la Ecologia del Suelo en el Centro de Sistemas Agricolas Ambientales. IV Taller Internacional sobre Recursos Fitogeneticos, Sancti Spiritus, Cuba, 3-4 December, 2001. (Barbercheck, Bell, Brownie, Louws, Koenning, Raczkowski, Wagger)

Organic Agriculture Research at the Center for Environmental Farming Systems. Carolina Farm Stewardship Association 16th Annual Sustainable Agriculture Conference, Rock Hill, SC, Nov. 2-4, 2001. (Barbercheck, Bell, Brownie, Creamer, Hu, Linker, Louws, Mueller, Raczkowski, Wagger)

Sustainable Agriculture Research at the Center for Environmental Farming Systems. Carolina Farm Stewardship Association 16th Annual Sustainable Agriculture Conference, Rock Hill, SC, Nov. 2-4, 2001. (Barbercheck, Bell, Brownie, Creamer, Hu, Linker, Louws, Mueller, Raczkowski, Wagger)

Qualitative and quantitative indicators of soil quality. Carolina Farm Stewardship Association 16th Annual Sustainable Agriculture Conference, Rock Hill, SC, Nov. 2-4, 2001. (Bell, M.C., M.E. Barbercheck, F.J. Louws, and M.G. Wagger.)

Conservation of Insect Pathogens in the Soil. IOBC-ESCOP meeting, The Practice of Biological Control: Importation and management of natural enemies and agents. Aug. 2-5, 2001, Bozeman, MT (Barbercheck)

Soil Ecology at the Center for Environmental Farming Systems. Soil Ecology Society Conference, May 20-23, 2001, Calloway Gardens, GA. (Barbercheck, Bell, Brownie, Creamer, Hu, Linker, Louws, Mueller, Raczkowski, Wagger)

The Ground Crew - Soil Arthropods and Beneficial Organisms. Invited. SC Organic Grower's School. Columbia, SC, 22 September 2001. (Barbercheck, Collins, Louws)

Transition strategies to organic production systems: economic and environmental indicators. Center for Environmental Farming Systems Field Day, July 9, 2001 (Barbercheck, Creamer)

Implementation of Long-Term Cropping Systems Studies: Challenges and Opportunities
American Society of Horticultural Science, July 2000
(NG Creamer* and JP Mueller)

CEFS Organic Farming and Cropping Systems Field Day, September 23, 1999
CEFS Organic Farming and Cropping Systems Field Day, July 31, 2000
CEFS Farming Systems Field Day, July 9, 2001

Project Outcomes

Project outcomes:

In September 1999 hurricane Floyd swept over North Carolina causing severe flooding in most of the eastern part of the state. Our experimental site at CEFS was no exception. Some of our fields were covered by up to 3 meters of water. The fact that we had initiated the systems experiment in fall of 1998 and taken baseline soil samples in the Spring of 1999 gave us the unique opportunity to provide quantitative assessment on flood impacts on such parameters as microbial indicators of fecal contamination, Glomalean fungal populations, heavy metals, pesticide residues, etc. Much of this information was of great interest to state government agencies attempting to make accurate flood assessment. We anticipate that in the future our experiment will be in a unique position to contribute this kind of ago-environmental information.

Economic Analysis

A detailed economic analysis of the first three years at CEFS is presently being finalized. The systems at CEFS are unique and complex agroecosystems and therefore require more consideration than straightforward cost and return analysis. An economic analysis amendment to this report will be submitted soon.

Farmer Adoption

Since farmers are a part of our ongoing advisory structure at CEFS they have had input and a hand in the overall design of the system from the beginning. In addition each year we have held field days and work shop activities where farmers have had the opportunity to view first hand the progress of the research activities. Of special interest to the organic farmers or those contemplating organic production is the organic transition study nested within the organic system. This study is providing information on various approaches to the transition from conventional to organic production


Areas needing additional study

Incorporation of agroforestry componets and additional integration of livestock enterprises would enhance the project further.

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.