Yolo County RCD logo
Search glass

Healthy Soils

The Yolo RCD and Yolo Land and Cattle bring the compost to the cows.

Healthy Soils Demonstration Project

This project is funded through CDFA’s healthy soils program, focusing on agricultural solutions to carbon sequestration in soils.

Over the past 150 years, carbon (C) in the atmosphere has increased by 30%. Soils contain approximately 75% of the land C pool and increasing soil C storage can significantly reduce atmospheric carbon dioxide. However, the benefits of certain agricultural practices on C sequestration remain to be ground-truthed and quantified. Our work builds on the pioneering efforts of the Marin Carbon Project to test the concept that particular agricultural conservation practices significantly reduce GHG emissions and sequester carbon beneficially in soils and vegetation. Carbon farming can enhance terrestrial carbon in plants and soils, and drives beneficial changes in other system attributes, including hydrological function, improving soil water infiltration and water-holding capacities; and biodiversity, with practices such as pollinator hedgerows and riparian plantings that provide wildlife habitat.

Our 50-acre demonstration project is being implemented at Yolo Land and Cattle, a long-time partner and advocate of conservation practices on rangeland. While compost applications on irrigated crop systems already have relatively widespread adoption, compost application on rangeland is not well understood or accepted as a practice. Even if only a fraction of the 34 million acres of grazed land in California adopted this practice, the benefits to climate and soil health are significant. This project will develop data needed for more widespread adoption and conservation incentives programming, by quantifying:

  1. reduction of net ecosystem exchange of CO2,
  2. increases in soil organic carbon storage, and
  3. increases in forage production.

You can learn more about the experimental design by clicking on the Info tab above.

Sampling Design

Within each control and treatment site we will establish one central point from which sampling locations for all metrics will be based, totaling 6 center points. Point Blue Conservation Science’s Rangeland Monitoring Network has developed a standardized point-based sampling protocol for measuring soil properties and plant communities throughout California rangelands. We will follow this methodology for those metrics, and adapt our sampling of net ecosystem exchange of CO2 (NEEC) and forage production to fit this design. Two of the treatment sites have points already established within the compost areas and have been sampled in the last 4 years for soil dynamic properties and vegetation using this protocol. The remaining 4 points will be located within the treatment and control sites such that all field sampling will occur within the designated areas and that paired sites will have similar topographic positions (slope, aspect, and position on slope).

Soil organic carbon, bulk density, and water infiltration rate

At each sampling point, we will randomly identify five subsample sites within 50 m of the location center. At these subsample sites, we will collect soil samples for lab analysis of soil organic carbon (SOC) and bulk density and measure water infiltration rate. Soil texture and water content will also be determined from each sample to help interpret the data. Each year, subsamples locations will be chosen at random again and will not correspond to previous years. Soil sampling will occur in January or February, on days when soil moisture levels are at field capacity. In 2020, the final SOC sampling will need to occur in November to collect data before final reporting is due.

For SOC measurement, soil subsamples will be taken using a step probe or by removing soil from the clean face of a pit. We will separate each subsample into 0-10 cm and 10-40 cm depths, combining the 5 subsamples for each depth category into one sample. These samples will be sent off to the University of Idaho Soil Analytic Lab, where inorganic carbonates are removed with an acid pre-treatment, and organic carbon is measured by dry combustion using an Elemental Analyzer. The same 10-40 cm depth sample will be used to determine soil texture at the University of Idaho lab using particle size analysis.

We will measure bulk density of the surface soil by inserting a ring 5.2 cm (2 inches) in diameter and 7.5 cm (3 inches) tall into the ground and removing the ring with soil contents. The contents of the ring will be sealed in plastic bags; shortly after field sampling (typically within 24 hours), the samples will be weighed to the nearest 0.1 grams. In the lab, rocks will removed by sifting samples through a 2 mm mesh sieve. Rocks will be weighed and their volume determined by displacing water in a graduated cylinder. Samples are then dried in an oven at 100° C for 24 hours and then re-weighed. We will then calculate Bulk Density (g/cm3) = oven dry weight of soil/ volume of bulk density ring - volume of rocks. We will calculate bulk density on the fine fraction (<2mm).

We will measure water infiltration rates following the protocol in the Natural Resources Conservation Service’s Soil Quality Test Kit. We will insert a 15.2 cm (6 inch) diameter PVC (polyvinyl chloride) ring with a beveled leading edge into the ground to 5 cm (2 inches) depth. We will place a plastic sheet in the ring and pour in 450 mL of water, remove the sheet and record the time it takes for the water to enter the soil. We will repeat this procedure in a second trial, except in instances where the first trial took longer than 50 minutes. If the second trial lasts more than 45 min., we will stop the timer, measure the depth of the remaining water and extrapolate the amount of time it would have taken the water to enter the soil assuming the rate measured over the 45 min. was applied to the remaining volume of water, assuming the water continues to drain at a linear rate.

Net ecosystem exchange of CO2

The same randomly-sampled locations of the 5 soil subsamples will be used in sampling net ecosystem exchange of CO2 (NEEC), though care will be taken to avoid sampling where soil has been disturbed. One 20-cm diameter soil gas PVC collar will be installed at each of the 5 subsample locations 24 hours prior to NEEC measurements, resulting in a total of 30 collars installed across the 3 paired sites (15 in treatment and 15 in control sites). At each PVC collar, CASA Systems will temporarily deploy a LI-COR model LI-8100 CO2 flux instrument in the mid-morning and again in the mid-afternoon hours on each sampling date, to approximate the diurnal pattern of CO2 emissions. The LI-8100 uses a static chamber-based gas flux system that briefly closes over the plant-soil surface, while an infrared gas analyzer measures gas concentrations from the closed chamber. Internal LI-8100 SoilFluxPro software automatically computes CO2 flux results from the rate of change in chamber gas concentration over time. This LI-8100 survey chamber system will meet all the requirements set out in the HSP Document 5: “References for Field GHG Emissions Measurements”, including the non-reactive flux chamber design, microclimate (humidity, temperature) controls, airtight pressure seals, gas mixing rate, and automated CO2 flux calculations. Standard gas samples will be used to check the calibration of the LI-8100 flux system on the day prior to soil flux measurements. Soil moisture (as volumetric water content) and temperature measurements at 5-cm depth will be recorded with LI-8100 probes for each soil sampling location at the time of CO2 flux estimation.

Net ecosystem exchange of CO2 (NEEC) will be measured in the field in two different seasons. Fall measurements will be conducted within one week of each compost application date (sometime between September and November), and will help to estimate the difference in NEEC due to the compost mineral source itself. Spring measurements will occur during the last week of April to quantify the differences in soil CO2 flux during this period of maximum (seasonal) grassland biomass accumulation and peak growth rate of forage plants. Only a spring NEEC measurement will be made prior to compost application and only a fall measurement will be completed after the last compost application in 2020 due to the constraints of the beginning and end of the grant period.

A Licor LI-8100-104C Clear Chamber will also be used to estimate net ecosystem exchange (NEE) of CO2, which is a key variable for understanding the carbon balance and long-term CO2 capture capacity of a grassland soil. NEE is defined as the gross primary production of plants minus the ecosystem respiration of CO2 (from plants and soils). The LI-104C Clear Chamber enables vegetation within the chamber to continue to take-up CO2 via photosynthesis during a soil emission measurement.

At the time of each soil CO2 flux measurement, three replicate measurements (per soil collar location) will also be made using a hand-held plant greenness (NDVI) meter. The NDVI sensor emits brief bursts of red and infrared light, and thereby measures the ratio of each type of light that is reflected back from the plant foliage. NDVI has been proven as an accurate index of herbaceous green cover density, which will be converted into seasonal net (after grazing) production of grass biomass (grams carbon per square meter) each year per sampling location.

Neither nitrous oxide nor methane GHG fluxes will be measured at our sites as part of this project, because these gases are only emitted in substantial quantities in moisture-saturated (wetland) soils or under heavily irrigated and nitrogen-fertilized cropland management.

Forage production and plant community composition

Forage production at peak standing crop will be measured inside two 8 x 8 foot exclosures placed at the ends of a 100-m transect centered on each of the 6 monitoring center-points. The bearing of the transect will be random, but remain the same each year of measurement and be permanently marked at center and each end. Within each exclosure, we will clip all above-ground plant material produced within the season at 6 randomly-placed 0.09-m2 circular plots. The exclosures will remain in the same location from year to year, but the standing plant residue will be evenly mowed down with a weed eater to a 4-inch height in the fall prior to compost applications. Approximately 75% of the mowed residue will be removed to simulate grazing and to avoid reductions in the following year’s production because of thatch build up. To the extent it is available, we will use local weather variables to predict timing of peak standing crop (George et al. 1989).

Plant community composition will be measured using the line point intercept method (Herrick et al. 2005) along the 100-m transect described above, combined with a 20-minute new species search within the 50-m radius circle around the center point. The line point intercept captures any herbaceous plant species intercepted by a pin drop at each meter along the 100-meter transect, and includes species detected on top and lower layers. If no living plant is intercepted, thatch, litter, and soil surface layers are recorded. The height of the top layer species is also recorded to gather information about vegetation structure.

Data analysis

The following response variables will be calculated for each center point sample from the data collection described above:

Response variable Sample mean (n) or Aggregate Units
Soil organic carbon, 0-10 cm Aggregate %

Soil organic carbon, 10-40 cm



Bulk density, soil surface

Mean (5)


Water infiltration rate

Mean (5)


Daily NEEC

Mean (?)


Forage production at peak biomass

Mean (12)


Cover, annual grass, legume, etc.



Cover, invasive species



Diversity, all species, native species, etc.



Diversity, forage beneficial species



Following the BACI design, we will primarily be interested in comparing the differences in the changes in the means of our response metrics before and each year after the compost applications in the control and treatment sites. We will compare these differences using simple paired or two-tailed t-tests, however with only 3 true replicated pairs, it may be difficult to detect statistically significant differences when these differences are subtle. In this case, we may rely more on simple statistical comparisons of each control-treatment pair and visual interpretation of our results to identify noteworthy impacts of the treatment.

For an advanced life cycle analysis of NEEC, estimated daily NEEC rates of grassland soils will be averaged (by treatment or control group) and plotted over time, from the presumed minimum (dry season) rates, to peak growth period (wet season) of grassland plants. Since measured NDVI and net soil CO2 fluxes will be used to estimate daily NEEC rates over several months of each year, diurnal variations in air temperature, soil temperature and moisture must be obtained from the nearest (CIMIS) weather station to compute varying daily NEEC using statistical correlation calculations with NEEC rates. Seasonal curve fitting will be applied to integrate the life cycle of NEEC gains or losses resulting from compost applications. Any and all statistically significant differences will be calculated between soil CO2 net flux rates in composted and in uncomposted plots sampled over the grassland growing season cycle.


George, M. R., W. A. Williams, N. K. McDougald, W. J. Clawson, and A. H. Murphy. 1989. Predicting peak standing crop on annual range using weather variables. Journal of Range Management 42(6): 508–513.

Herrick, J. E., J. W. Van Zee, K. M. Havstad, L. M. Burkett, W. G. Whitford. 2005. Monitoring manual for grassland, shrubland and savanna ecosystems, vol. I-II. USDA-ARS Jornada Experimental Range.