Embedded Files
Monitoring & Quantifying Outcomes
After a carbon farm plan is completed and implementation priorities identified with a producer, it is important for planners and producers to understand different general approaches to monitoring the success and impacts of carbon farming practices that have been implemented. Monitoring how well a practice works or does not work is fundamental to learning and adapting. Furthermore, quantifying some of the GHG and other benefits of implemented practices may be important to a producer or as a reporting requirement for a funding source. These monitoring and quantification processes generally align with NRCS Conservation Evaluation and Monitoring Activities (CEMAs).
This section provides a broad overview of various approaches to monitoring and quantifying the GHG outcomes of practice implementation to complement the carbon farm planning process. It is not intended to be an in-depth or exhaustive description of monitoring and quantification methodologies. Some of the key questions to consider include:
What impact(s) are you trying to monitor or quantify? Carbon sequestered, productivity benefits, water quality benefits, etc.
Who is the monitoring or quantification for? What requirements do they have?
Over what time period do you expect to be able to see the expected impacts?
Over what area are you assessing impacts? What are the boundaries?
When monitoring or collecting data on behalf of a producer or land manager, it is critical to do so with their permission and to honor their privacy with respect to sharing data.
Observations & Monitoring
Observations & Monitoring
Field observations and assessments are indispensable monitoring tools both on their own and as a supplement to other more quantitative approaches. Over time, they help build an understanding of long-term trends and patterns, and help evaluate the outcomes of management changes.
For many agroforestry practices, the general success of a project can be directly observed over time in terms of plant establishment and growth. For many other practices, you may need to rely on a variety of observable indicators instead. For example, changes over time in plant community composition, increased plant growth and productivity, reduced signs of soil erosion, improved soil tilth, or increased earthworm numbers could all indicate the impacts of a carbon farming practice focused on building soil health and increasing soil carbon.
Observations over time are, for the most part, qualitative. And in many cases, this may be all that is needed. In other cases, when a more quantitative approach is called for, observed trends over time are still an important way of ground-truthing the results of models and direct measurements.
Modeling
Modeling
Models are used to predict and quantify changes in soil organic carbon and GHG emissions for a given agricultural system, accounting, to varying degrees, for different management practices, soil types, and climate conditions. Modeling approaches can range from being relatively simple to extremely complex.
A simple modeling approach is the use of emission factors to estimate changes in GHG emissions. For example, the Intergovernmental Panel on Climate Change (IPCC) Tier 1 emission factor for nitrous oxide in dry climates is 0.5%. Using this emission factor, if a producer changes from applying 100 lbs of nitrogen fertilizer every year to now applying 60 lbs of nitrogen fertilizer every year, we can estimate that nitrous oxide emissions have decreased by 0.5% x 40 lbs, or 0.2 lbs, of nitrous oxide each year.
Examples of a much more complex modeling approach is the use of process-based models such as the Daily CENTURY model (DayCent), the Rothamsted Carbon Model (RothC), and the DeNitrification-DeComposition model (DNDC), which simulate soil carbon and nitrogen dynamics under different conditions. These models require a high degree of user expertise and model inputs.
The COMET tools are designed to provide a more accessible interface with the DayCent model. COMET-Planner uses regional averages from Major Land Resource Areas (MLRAs) to construct a fixed baseline for a more generalized description of conservation GHG benefits. In cases where more field- or practice-specific quantification is needed, and where detailed baseline management data is available, COMET-Farm can be used. Refer back to Module 2 for more detailed information about the COMET tools.
Models are a means of estimating changes in carbon and GHG emissions, and will never perfectly reflect what is actually happening in a specific field under specific conditions. The accuracy of these estimates will depend on the accuracy of the model and the accuracy of the information put into the model.
Importantly, in addition to changes in soil organic carbon, many of these models estimate emissions of other GHGs (N2O, CH4) which are extremely difficult to measure. The use of these models therefore is also a way of accounting for changes in other GHG emissions resulting from a carbon farming practice.
Direct Measurement
Direct Measurement
Another approach is to directly measure changes in soil carbon and biomass carbon over time. It is not possible to measure all of the soil in a field or all of the trees in a windbreak; instead, we must rely on collecting a number of representative samples for analysis, and use those to generate an estimate for what is happening across the entire field. Collecting a larger number of samples will, in general, give us more confidence in our estimates but will add cost. A direct measurement approach is often expensive and requires that you make decisions about the desired level of certainty of your estimates versus the time and cost needed.
It is important to note that even with the most intensive direct soil sampling effort to measure soil carbon changes over time, quantifying other GHG emissions (nitrous oxide and methane) will require supplementary modeling, as fluxes of these gases cannot practically be measured outside of a research setting.
Carbon Cycle Institute Field Day (Photo taken pre-COVID-19 pandemic)
Introduction to Soil Sampling
Soil sampling is a powerful tool that can help inform nutrient and other management decisions and help track changes in soil carbon and soil health over time. Collecting and analyzing a soil sample from one location provides very good information about that specific sampling location and depth at the time the sample was collected. To understand a soil at the field scale, we use a number of soil samples from multiple point locations to make inferences about the entire field. In doing so we have to think about how variable the soil is, the number of samples to collect, what soil depth(s) we are interested in, and where to collect those samples to best represent the field.
Similarly, to understand how a soil changes over time, we collect a number of samples over time to make inferences about how and to what extent the soil is changing over that time period. In doing so we have to think about how much the soil might change seasonally, when during the year to collect samples, how frequently to collect samples, and how to compare results between time points.
The specific questions we are trying to answer will inform how we sample across space and over time, what methods and protocols we use, and how we analyze the soil samples we collect.
In this section we will learn more about why it is critical that your sampling protocol, methods, and lab analyses be suited to your sampling goals.
Questions to consider before soil sampling
Questions to consider before soil sampling
What question(s) are you trying to answer?
Over what time period, and what land area are you trying to answer these questions?
What other sources of data will you need in addition to or in place of soil samples?
What sampling protocol will you use?
Do you have trained field staff or technical assistance with appropriate sampling equipment?
What lab/ facilities can provide the processing and analyses you are interested in?
Do you have funding for sampling and processing?
How will you interpret the results?
Will your sampling protocol and lab results provide data that can answer your questions with the level of confidence you need?
A variety of soil sampling tools can be used depending on the goals and resources involved. A cylindrical soil probe with a handle, pedal, and/or slide hammer can help collect samples with consistency and accuracy. Photo: Fery et al. 2018. Oregon State University Extension Service EC628.
Challenges of Soil Sampling
Soil variability
Soil variability
Soils form through the interactions of parent material, climate, topography, organisms, and time (Jenny 1941). Learn more about the Five Soil Forming Factors through the link in the Resources tab. Soils are extremely heterogeneous at the landscape scale, at the field scale, and even at much smaller scales.
Remember the ideas from Reading the Landscape and Ecological Site discussions; those same concepts are highly related to the distribution and formation of diverse soils across a landscape. A soil forming on a north-facing slope will form differently over time than a soil forming from similar parent material on a south-facing slope. Similarly, a soil forming in a slight depression in the landscape will form differently than a nearby soil in a well-drained landscape position. The complexity of these interacting factors means soils are highly variable across the landscape. Note the illustration of interacting soil types across the Colusa County, CA landscape below. Whether doing a qualitative soil assessment or a rigorous soil sampling campaign, it is important to consider the landscape context, including the distribution of soil series gleaned from sources like the Web Soil Survey.
Soil Survey of Colusa County, California. USDA NRCS 1999.
Soil map units themselves are general representations: soils are highly variable even within a soil map unit and at the sub-field scale. The figure to the right from Kiani et al. 2020 (check out the Resources tab for this link) shows soil organic carbon concentration ranging from around 2% to 0.5% over a distance of about 60 meters (~200 feet). Differences like these could be driven, for example, by slight differences in topography resulting in different erosion pressure, hydrologic conditions, or sun exposure. Soil can also be impacted by highly localized factors such as field edges, irregular vegetation such as lone shade trees, roads and trails, and livestock congregation areas such as watering and feed out areas. When assessing and sampling soils, it is critical to consider these factors and ensure that the location being assessed or sampled is representative of the conditions you are wanting to represent.
A statistical estimation of the spatial distribution of soil organic carbon concentration (%) at a depth of 5–10 cm across an irrigated cultivated field in Alberta, Canada (Kiani et al. 2020; doi.org/10.1139/cjss-2019-016).
Soil profile in an arid rangeland with an accumulation of carbonates. Note that a 1 meter deep soil sample collected from location A would likely return some different results than the same sample collected from location B. Photo source: USGS, https://www.usgs.gov/media/images/soil-profile-arid-rangeland.
Soil properties are highly variable on much finer scales too.
Some of this variability is visible and can be incorporated into our sampling design. For example, soil from within a row of corn will be different than soil from between rows. When collecting samples, where those samples are to be collected in relation to crop rows or individual plants should be pre-determined and followed consistently.
Other small scale variability can be harder, or even impossible, to avoid. As an example, notice the variability in the distribution of white carbonates in the soil profile to the left. These variations would not have been obvious from the soil surface and therefore could not have been avoided while collecting samples.
Other sources of error
Other sources of error
In addition to variability of the soil itself, we introduce sources of error through the process of sample collection, sample handling, and sample analysis, all of which can add noise to the final data and obscure their interpretation. Some sampling methods and analyses can be more, or less, susceptible to different sources of error. It is not uncommon for it to be difficult to distinguish between variability in the data due to noise, versus actual variability of the soil in the field. It is also possible for a source of error to introduce bias into the data, thereby skewing the results and potentially leading to inaccurate interpretation.
These sources of error can be kept to a minimum by following sampling protocols carefully and being as consistent as possible by using the same methods, tools, and analytical labs for repeat samples. Even having the same people consistently perform the same tasks can help reduce error.
Detecting a small change in a large pool
Detecting a small change in a large pool
Consider this crop field along the Sacramento River (see map below) with soils containing an average of approximately 2.8% soil organic matter or 1.4% organic carbon (assuming that soil organic matter is made up of roughly 50% carbon; Pribyl 2010) and a bulk density of 1.5 g/cm3 in the top foot. The total organic carbon contained in the top foot of one acre of this orchard can be estimated as:
(1.5 g soil per cm3) x (40,468,564 cm2 per acre) x (30.5 cm per foot) x (1 metric ton per 1,000,000 grams) = 1,851 metric tons of soil per acre in the top foot
(1,851 metric tons of soil per acre) x (1.4% organic C) = 25.9 metric tons of organic C per acre of soil in the top foot
Now consider that we want to begin planting a mixed cover crop every winter, which according to the CDFA COMET-Planner tool is estimated to sequester 0.37 metric tons of CO2e (or 0.11 metric tons of carbon) per acre per year in Sutter County, CA. That’s the equivalent of a change in soil organic matter from 2.80% to 2.81% each year, an extremely small change to detect. Even after five years, detecting an increase to 2.85% soil organic matter is difficult to do with confidence, particularly if the field in question is highly variable or if the soil sampling process introduces significant sources of error.
This simple scenario illustrates the importance of choosing sampling protocols that are appropriate for the questions being asked, carefully determining sample numbers, locations and frequency based on the questions being asked, and minimizing unnecessary sources of error by following protocols carefully and consistently.
A crop field along the Sacramento River in California. Web Soil Survey.
Collecting soil samples along a transect. Carbon Cycle Institute Field Day (Photo taken pre-COVID-19 pandemic)
Considerations for sampling to measure soil organic carbon change over time
Considerations for sampling to measure soil organic carbon change over time
There are many variations on soil sampling protocols depending on conditions, available resources, and specific goals. And while there is no single universal protocol for measuring soil organic carbon change over time, some important considerations include:
Sample size, sample distribution, and stratification
Frequency of sampling
Time of year of sampling
Sampling depth intervals
Single samples vs composite samples
Whether a baseline (pre-implementation) or control plot is needed
At a minimum, determining soil carbon content across a given area of soil to a specified depth requires a measure of soil carbon concentration and a measure of bulk density.
Soil Carbon Concentration
Soil Carbon Concentration
Soil carbon concentration is typically reported as a percentage or, in other words, how many grams of carbon are contained in 100 grams of dry soil. There are three different analyses for determining the concentration of carbon in a soil sample:
Organic Matter
Organic Matter
Typically measured using Loss on Ignition (LOI), this is the percentage of all combustible material in the soil sample, which is assumed to be all of the organic matter. Organic matter is made up of approximately 50% carbon, although a range of values has been reported (Pribyl 2010). The LOI analysis, while inexpensive and widely available, is considered to have low accuracy in determining actual soil carbon concentrations.
Total Carbon
Total Carbon
Not all soil analysis labs perform this analysis. This is a direct measurement of the total carbon contained in a sample of soil using a dry combustion method. This is the most accurate method for measuring all of the carbon contained in a sample of soil. For soils that do not contain notable quantities of Inorganic Carbon, Total Carbon is considered to be equal to Total Organic Carbon. However, soils containing inorganic carbon, such as alkaline soils or carbonate-amended soils, require pre-treatment (see below).
Total Organic Carbon
Total Organic Carbon
This is the same dry combustion analysis as for Total Carbon, except that soil samples are pre-treated with acid to remove the inorganic carbon fraction prior to dry combustion analysis.
Bulk Density
Bulk Density
Bulk density is the mass of dry soil divided by the volume, typically reported in units of grams per cubic centimeter. It can be used as a measure of soil compaction; a highly compacted soil will have a higher bulk density than a well-structured porous version of the same soil. For the purposes of soil organic carbon estimation, bulk density is a key parameter for converting a soil carbon concentration (e.g. percent) into a quantity of soil carbon in a given area. Let’s say a soil has 2% organic carbon concentration and a bulk density of 1.6 g/cm3 in the top 15 cm of soil:
(1.6 g soil per cm3) x (40,468,564 cm2 per acre) x (15 cm) x (1 metric ton per 1,000,000 grams) = 971 metric tons of soil per acre in the top 15 cm
(971 metric tons of soil per acre) x (2% organic C) = 19.4 metric tons of organic C per acre of soil in the top 15 cm
Bulk density is measured by collecting a sample of soil of a known volume, drying and weighing the soil, and calculating the dry mass divided by the known volume. Care must be taken to not compact the bulk density sample or lose any amount of the sample, as these will impact the bulk density measurement.
Collecting bulk density samples from a soil pit. C
Other Supporting Analyses
Other Supporting Analyses
Because of the effort involved in collecting soil samples, it often makes sense to have the lab perform other analyses on the samples, either to provide to the producer or as supporting information for the carbon data. It is recommended that you ask the producer or landowner if they are interested in any additional analyses of their soil. A rangeland manager may never have had their soils tested for nutrients and might be interested in getting this information. Or a producer might be interested in a specific micronutrient and ask that you add that analysis onto your sample submission to the lab. Soil texture data is often collected because it plays an important role in soil carbon dynamics and may help explain patterns in soil carbon accumulation.
Additional and support analyses may include:
Soil texture
Nutrients
Cation Exchange Capacity
pH
Aggregate stability
Various soil biological analyses
Additional Resources
Additional Resources
If the following URLs do not open immediately when selecting the link, right click and select “open in a new tab” to view the resource.
❓Check your knowledge!
❓Check your knowledge!
True/False: It is important to choose a soil sampling protocol and analyses that are appropriate for the questions I am trying to answer.
True/False: Qualitative observations of the soil have no role to play in soil health assessment.
True/False: There is one universal soil sampling protocol to follow for measuring changes in soil carbon over time.
Changes in soil organic carbon can be difficult to detect because: (select two)
A. Soil is highly variable
B. Soil carbon does not change year to year
C. Changes are small relative to the pool of carbon that already exists in the soil
Select the two measurements that are necessary to estimate the quantity of soil carbon in a given area:
A. Aggregate stability
B. Soil carbon concentration
C. Bulk density
D. Cation exchange capacity
Return to
Discussion board: If you have any questions throughout Module 3, please use the discussion board below to post.
Return to
Case Studies
Page updated
Google Sites
Report abuse