GLBRC Biofuel Cropping System Experiment – Sampling




The GLBRC Biofuel Cropping System Experiment (BCSE) was established in 2008 to examine the performance of potential bioenergy cropping systems. Identical experiments were set up at the GLBRC Intensive Site at Kellogg Biological Station, Hickory Corners MI and at the Arlington Agricultural Research Station of the University of Wisconsin, which was retired at the end of the 2017 growing season.

The BCSE is lain in a randomized complete block design that includes 5 replicated blocks (R1-R5) of up to 10 cropping systems or treatments (G1-G10), each in 30 × 40 m plots. Current cropping systems (as of 2022) include three no-till annual treatments: continuous corn with stover removal (G1) and a sorghum (photoperiod-sensitive ISU hybrid HPS5500)-soybean rotation with both points of entry (G2-G3); and seven perennial treatments, most continuous since 2008: switchgrass (G4, planted in 2021, and G5; Panicum virgatum variety Cave-in-rock), miscanthus (G6, Miscanthus x giganteus), native grasses (G7, a mix of 4 species), poplar (G8, “NM-6”, Populus nigra x Populus maximowiczii), early successional (G9, abandoned field), and restored prairie (G10).

The corn treatment (G1) has been continuous since 2008, but the other annual treatments (G2-G4) have changed over the years. In 2012, the original corn-soybean-canola cropping systems (G2-G4) were discontinued and replaced with continuous corn + cover crop (G2) and a corn-soybean rotation + cover crop (G3-G4, with each entry point). In 2018, these systems were discontinued: no-till energy sorghum (photoperiod sensitive) was established in G2 and the G3 and G4 plots were fallowed (planted to temporary cover crops). In 2019, G3 was converted to no-till energy sorghum (photoperiod insensitive) with a cover crop. In 2021, G4 was planted to switchgrass and in 2022, G2 and G3 were converted to sorghum-soybean rotations.

Each BCSE plot contains a semi-permanent set of 3 sampling stations (S1-S3) around which most within-plot sampling is performed. The location of these stations change when needed.

Subplots (may also be called microplots) within treatment plots provide a means for comparing alternative agronomic practices to the main plot, such as 1) stover removal vs. non-removal and 2) no nitrogen vs. plus nitrogen fertilizer, as described below.

In addition, since 2018, rainout shelters are deployed in the main plots (R2-R5) of the switchgrass treatment (G5), managed the same as the rainout shelters in the Marginal Land Rainfall Exclusion Experiment, and serve as a higher fertility reference site for that experiment.


Each treatment plot (except G8) consists of a 60 ft wide main plot with 15 ft wide subplots on opposite lengthwise sides. One subplot is managed the same as the main plot, the other is under alternative management. Subplots account for edge effects and the use of different harvesting equipment than in the main plot. See the current and archived plot maps for location and treatment of subplots.

Stover subplots: In G1-G4, stover removal vs. stover non-removal subplots examined the impacts of corn stover removal on ecosystem processes. In general, stover was harvested in east side subplots and was not harvested in west side subplots. The exception is that subplots in Block 1 (G2R1, G3R1, and G4R1) are lain in the opposite direction. This subplot study continues in G1 but changed in G2-G4 to a plus nitrogen vs no nitrogen fertilizer comparison in 2018-2021 as crops changed from corn-soybean rotations to sorghum-soybean rotations (G2-G3 in 2022) and switchgrass (G4 in 2021). From 2013-2017 in G2-G4, which included cover crops in their rotations, an additional study was carried out in the stover removal subplots only to examine the effects of cover crop harvesting in the main plot on ecosystem processes. Cover crop removal continued in the southern 2/3 of the subplot but ceased in the northern 1/3 of the subplot. Instead of harvesting and removal, cover crops in the northern section were killed with an herbicide and their remains were left on the field (cover crop remains, herbicided); yields from these subplot sections were recorded from 2014-2017.

Nitrogen fertilization subplots: In G5-G9, plus nitrogen vs. no nitrogen fertilizer subplots examine the impacts of N fertilizer on ecosystem processes. Since G5-G9 are normally N fertilized, the alternative treatment subplot has no nitrogen applied; the plus nitrogen subplot is N fertilized as in the main plot. Plus nitrogen subplots are on the east side and no nitrogen subplots are on the west side of the G5-G9 main plots. G10 is normally not fertilized, the alternative treatment subplot has nitrogen added. No nitrogen subplots are on the east side and plus nitrogen subplots are on the west side of all G10 main plots except Block 1 (G10R1) where the opposite pattern occurs.

As mentioned above, the original stover subplots in G2-G4 changed to nitrogen fertilizer subplots with the change in crops to sorghum-soybean rotations (G2-G3 in 2022) and switchgrass (G4 in 2021). For G2-G3, the subplots with stover removal became the no nitrogen subplots (on the east side of main plot in all replicates except G2r1 and G3r1). For G4, the stover non-removal subplots (with no stover removed) became the no nitrogen subplots in all replicates except G4r1, due to interference with gas sampling equipment. All no nitrogen G4 subplots are located on the west side of the main plot.

Subplots in poplar (G8) were discontinued after the 2018 harvest. From 2010-2018, the no nitrogen subplots were located on the southern edge of the main plot and the plus nitrogen fertilizer subplots on the northern edge. Note that the main plots in G8 are not fertilized every year.


Soil Sampling

Soils are sampled to measure and monitor a variety of chemical (pH, organic carbon, inorganic nitrogen, and several other nutrients) and physical (soil bulk density, texture, and moisture) properties. Both surface and deep-core soil samples are collected. Currently surface samples to 25 cm are collected on a near annual basis and analyzed for fertilizer recommendations and inorganic nitrogen. Deep-core sampling to 1 meter occurs every 5 to 10 years to primarily follow long-term changes in organic carbon and nitrogen. Soil water (leachate) was also collected from buried soil lysimeters throughout the 2010-2021 growing seasons and analyzed for an array of ions.

Surface soils:

Inorganic Nitrogen: Since 2016, surface soil cores are taken annually in the fall in the main plots of all BCSE treatments to monitor inorganic nitrogen levels. Two soil cores are taken with a sampling (push) probe (2 cm dia., 0-25 cm depth) around each of the 3 established sampling stations (S1-S3) per plot (R1-R5) within a 2 m radius of each station flag. All cores from one plot are composited in the field into the same sampling bag. For plots with row crops, an equal number of cores are taken within rows and between rows, avoiding wheel tracks between rows. For poplar plots, an equal number of cores are taken within the tree row and between tree rows. Every effort is made to sample all plots on the same day. And since 2017, subplots with alternative treatments to the main plot are also included in the annual sampling and analyses; typically, nine randomly placed cores are taken and composited from each subplot.

From 2014-2015, soil cores (to 25 cm depth) were taken 3 times per year from KBS main plots using the above procedure and analyzed for inorganic nitrogen: the timing included prior to planting of annuals (May), after fertilizer application (June), and just before harvest (October). Soil cores (to 15 cm depth) were also taken 3 times per year at Arlington from 2013-2017 and included the main plot plus the alternative treatment subplots for G1, G5-G7, and G9-G10. And from 2010-2012, soil cores were taken roughly every two weeks over the growing season from all main plots at KBS and from all main plots at Arlington plus the alternative treatment subplots in G1, G5-G7 and G9-G10 in 2012 only.

In 2009, surface soils were not sampled for inorganic nitrogen. Instead resin strips were used to monitor nitrogen availability. Resin strips were placed in the field monthly from March to November at each sampling station (S1-S3), then collected after ~4 weeks incubation and analyzed for inorganic nitrogen.

Agronomic Soil Chemistry: Sieved soil samples taken in the fall from the main plot (since 2008) and treatment subplots (since 2011), either for inorganic nitrogen as above or as additional samples collected similarly, are air dried and subsamples sent to an analytical laboratory for nutrient analysis and lime/fertilizer recommendations. Soils from annual treatments are analyzed every year; starting in 2018, soils from perennial treatments are analyzed every three years.

Soil pH: Soil pH of surface cores (0-25 cm) is periodically measured on sieved, field-moist soil samples taken in the fall to evaluate the need for liming. Note that soil pH is also measured in samples sent for agronomic soil chemistry analysis.

Soil Moisture: Gravimetric soil moisture is routinely measured on composite soil samples taken for inorganic nitrogen determination and on samples that accompany greenhouse gas flux measurements.

Soil Bulk Density: In 2008-2009, soil bulk density was measured on surface soils by depth intervals of 0-10 and 10-25 cm in all BCSE treatments (G1-G10). Since then, soil bulk density is routinely measured for all depth intervals of deep core samples, as described below.

Nitrogen Mineralization: Potential nitrogen mineralization and nitrification rates were measured three times over the 2013 growing season in surface (0-25) soils from main plots and alternative treatment subplots of BSCE treatments (G1-G10) using 28-day lab incubations.

Soil Total Carbon and Nitrogen: A baseline study was conducted across the KBS GLBRC Intensive Site in early spring 2008 to map the soil total carbon and nitrogen (CN) content of surface (0-10 cm) soils prior to establishment of the BCSE and the Switchgrass Nitrogen/Harvest Experiment. Since then, CN is routinely measured for all depth intervals of deep soil cores, as outlined below.

Deep soil cores:

Deep soil cores (0-1 m) are taken approximately every 5 years in all BCSE treatments (G1-G10) to primarily monitor for changes in soil bulk density and total carbon and nitrogen. Sampling occurs after harvest in late fall, except in 2008 when KBS soils were sampled in early summer. Since 2008, deep cores are taken in the vicinity of each sampling station (S1-S3) in every treatment replicate (R1-R5). In 2017, three cores were also taken from the stover non-removal subplots in G1-G4 at north, central, and south (N, C, S) sampling locations.

Extracted cores are divided into four depth intervals of 0-10, 10-25, 25-50, and 50-100 cm. Each interval is analyzed separately for soil bulk density and total carbon and nitrogen. Dried depth interval samples are then composited across sampling stations (by replicate) and a representative subsample is sent to an analytical laboratory for nutrient analysis and fertilizer recommendations. The remaining interval sample, or a portion thereof, is archived: interval samples were archived by sampling station in 2008, but since 2013 are combined by replicate (across sampling stations) and a single composited sample is archived per depth interval.

Deep soil cores taken in 2008 were also analyzed for sand, silt, and clay content (soil texture) using the hydrometer method.

Soil water:

Low tension soil water samples, also called suction lysimeters, were installed in the main plot of all BCSE treatments in replicates R1 and R2 in 2010 and in R3 and R4 in 2013. The lysimeters were buried ~1.2 meters deep and pull soil pore water (also called leachate) at a low suction. Through 2021, soil leachate samples were collected biweekly throughout the growing season (or as allowed by precipitation and soil frost) and analyzed for a variety of chemical and physical properties.

Plant Sampling

Yield: Yields are defined as the grain and/or non-grain biomass harvested in a given growing cycle and serve as the basic measurement for comparing production differences among treatments. For annual row-cropping systems of corn, soybean and canola, yields are determined as the grain biomass collected during harvest; for annual sorghum, yield is the herbaceous biomass harvested. For perennial herbaceous crops that are harvested annually, including monoculture (switchgrass and miscanthus) and polyculture (native grasses, early successional and restored prairie) systems, yield is determined as the biomass harvested at the end of the growing season. For woody cropping systems of short-rotation poplar, annual yield is determined as the biomass collected at clear-cutting divided by the number of years of growth.

Yields are measured during normal crop harvest by commercial size machine harvesters appropriate to each crop, as described in the BCSE Agronomic Protocol or KBS Aglog (search for Harvest under the Observations tab). Grain yields are reported at standard moisture and as dry matter; non-grain yields are reported as dry matter. See the Agronomic Yields protocol for procedures and calculations.

Corn stover is also harvested from the main plot and stover removal subplots, as outlined in the Corn Stover protocol, which also includes procedures on grain harvest in the corn stover subplots.

Subsamples of harvested yields are collected, dried, ground and analyzed for plant total carbon and nitrogen (CN). For annual grain crops, CN is measured on the grain and corn stover harvested from each main plot and the grain harvested from each stover non-removal subplot. For perennials, CN is measured on a subsample of the “unsorted” harvest from each main plot and each no nitrogen fertilizer subplot in G5-G9 and each plus nitrogen subplot in G10. Samples are not analyzed from subplots that replicate the management of main plots.

Aboveground Biomass: Aboveground biomass is sampled from the main plots at peak biomass for estimation of aboveground net primary production (ANPP). For herbaceous crops, ANPP is determined as the sum of crop plus non-crop species and cover crops, when included. For poplar, it is determined as the sum of woody growth increment, leaf biomass (litter), understory biomass, and biomass affected by occasional agronomic practices. Each of these components are sampled under separate sampling campaigns, as outlined in the ANPP protocol.

For annual crops, ANPP samples are sorted to species (crop and non-crop “weedy” species), then oven dried and weighed as “whole” plant fractions. Grain crops of corn, soybean and canola are then threshed to determine “grain” (seed) biomass. At KBS, crop fractions—grain and the aboveground vegetative biomass termed “stover” in the datatables—are composited separately across sampling stations to form pooled replicate samples. Non-crop species are aggregated into one sample and surface litter samples—dead plant remains on the soil surface—are composited by replicate as well. Representative subsamples are then taken from these composited samples, finely ground, archived, and analyzed for plant total carbon and nitrogen (CN). At Arlington, samples were sorted as at KBS but were not pooled across sampling stations for CN analysis; samples from each sampling station were analyzed.

For perennial crops, beginning in 2018, ANPP samples are no longer sorted to species. Samples are processed as hand-harvested; they are oven dried, weighed, and cataloged as “unsorted”. After the biomass is recorded, these samples are then composited by replicate (across sampling stations) and subsamples are finely ground, archived, and analyzed for CN. From 2009-2017, all perennial crop treatments were sorted to species, and each species dried, weighed and recorded. At KBS, samples from monoculture treatments were then composited by replicate into a crop (switchgrass or miscanthus) sample and a combined sample of all additional species (“unsorted”) when encountered. Samples from mixed culture treatments (G7, G9-G10) were composited by replicate into separate samples for the top three species, by weight, and an aggregate sample of all remaining species. Representative subsamples from these composited samples were finely ground, archived, and analyzed for CN, as above. At Arlington, samples were sorted as at KBS but were not pooled across sampling stations for CN analysis; samples from each sampling station were analyzed.

For poplar (G8), aboveground woody biomass is estimated from annual stem diameter measurements and regression equations of dry stem biomass vs stem diameter. Leaf biomass is estimated from leaf litter collection throughout the leaf fall season; understory biomass is measured at peak biomass. Composite samples of leaf litter and understory biomass, plus woody biomass taken at destructive harvests, are analyzed for CN, as above.

Aboveground biomass may also be sampled at times other than peak biomass for specific purposes. For example, residue stover samples are routinely collected after corn stover harvest to determine stover harvest efficiency. And biomass samples were collected in corn (G1) and perennial (G5-G7, G9-G10) treatments just prior to harvest in 2014-2016 to determine harvest efficiency.

Species Composition: The species composition of all BCSE treatments from 2008-2017 can be ascertained from samples collected at peak biomass for ANPP and sorted to species. Beginning in 2018, ANPP samples in G2-G10 are no longer sorted to species. The line-point intercept method is used instead to provide information on the composition and cover of canopy-forming plant species in treatments G2-G10. The identity and height of the first vertically intercepted plant is recorded at ~ 1m intervals along three north-south transects (Lines 1-3) that are positioned a predetermined distance and direction from the station flags.

Phenology: Phenological data on major crop life stages were collected by direct observation roughly every two weeks throughout the 2010-2017 growing seasons in Replicate 1 of all BCSE treatments. Stages include emerging growth, mature leaves, open flowers, ripe seeds, closed canopy, autumn leaf senescence, and leaf loss. Defining characteristics of these stages are given in the Phenology protocol. Since 2015, webcams have been used to record life stages. One camera is installed in each treatment plot in Replicate 1 and takes one photo at noon each day. These images are stored at KBS and are available upon request. Images from 2015-2021 are also currently available through the Phenocam network.

Leaf Area Index: Leaf area index (LAI) was measured at KBS every two weeks throughout the 2009-2017 growing seasons in all BCSE treatments (G1-G10). Measurements were made at paired locations on each of the four sides of treatment plots in Replicate 1. Note that different meters were used in 2009-2013 and 2013-2017. Dual measurements using both meters were conducted in 2013 in order to compare results.

Belowground Biomass: Different sampling methods have been used to measure the standing root biomass in the diverse cropping treatments of the BCSE. For annual crops (G1-G4), root biomass was sampled from 2008-2014 by excavating the roots of a single crop plant at each sampling station (KBS) or all roots withing a 1 m2 area (Arlington). Roots were washed, oven dried, weighed and processed for CN analysis. At KBS, roots were composited by replicate for analysis; at Arlington, roots from all stations were analyzed. For perennial crops, root biomass was sampled from 2009-2013 by taking 1-meter deep cores along a gradient from the center of the target plant (crop in monoculture treatments or dominant plant in mixed community treatments), adjacent to the target plant, to interstitial between plants. Cores were then divided into 4 depth intervals (0-10, 10-25, 25-50, and 50-100 cm) and the roots sieved out, rinsed, oven dried and weighed. These data can be used to monitor root development over time and with depth but should not be directly used to estimate areal root biomass because they are not scaled to plant density.

In-growth cores were used in all perennial treatments (G5-G10) from 2009-2014 to estimate fine root production. Two cores were buried at each sampling station at the beginning of the growing season, one was collected mid-season and the other at the end of the growing season to assess an approximate peak standing root biomass. Cores were sieved and the roots washed, oven dried, weighed, and processed for CN analysis. At KBS, roots were composited by replicate before CN analysis; at Arlington, roots from all stations were analyzed.

In 2017, the belowground plant biomass in all perennial treatments (G5-G10), except poplar (G8), was sampled shortly after machine harvest by excavating approximate 1 m3 pits in two layers (0-50 and 50-100 cm) and sieving all soil to collect the roots.

Plant Total Carbon and Nitrogen: Aboveground samples from both yield and peak biomass samples are analyzed for CN; see above sections on yield and aboveground biomass for sample collection, processing, sorting, and analysis. Belowground samples from root excavations and deep cores may also be analyzed for CN; see above section on belowground biomass for details.

Tissue Chemistry: Occasionally leaf tissue samples are collected, dried, and sent to an analytical laboratory to monitor nutrient concentrations and assess fertilizer recommendations. Samples of harvested crops were also analyzed for feedstock quality (sugars, cellulose, fatty acids, lignin content) in 2008-2012.

Greenhouse Gas (GHG) Sampling

Fluxes of the GHGs (CO2, CH4, and N2O) over the growing season have been measured in the main plots (R1-R4 at KBS, R1-R5 at Arlington through 20XX) of all BCSE treatments (G1-G10) since 2008. From 2008 -2018, only the static chamber method was used; starting in 2019 a recirculating chamber method is included as an alternative method. Chamber bases are deployed in early spring and, except for agronomic operations or repair, remain in place over the growing season. See linked protocols above for sampling and analysis specifics of each method. Extensive testing has established no measurement bias between the two chamber methods.

Gravimetric soil moisture and soil temperature are also measured during each sampling event, as described in the protocol.

Date modified: Friday, Aug 04 2023



Sign In