A common question we hear at TSIP is: “We love your auger system, but does it give the same results as a push probe?” It’s important to ask this kind of question any time new tools are introduced to ensure consistent and accurate measurements between and across studies. We decided to expand this to three research questions to rigorously test the differences between these two soil harvesting methods:

1. Does the TSIP auger system produce the same soil carbon estimates at the field scale as a push-probe?
2. How does the variation between sampling methods compare to the spatial variation in soil carbon?
3. Is there a tradeoff between speed and accuracy with these methods?

We designed an experiment carried out at two farms located close to Skidmore College. Pitney Meadows is an organically managed produce farm in Saratoga Springs, NY and Hand Melon Farm is a conventionally managed produce farm in Greenwich, NY (Fig 1). At each farm, we collected samples at plots specifically designed to compare spatial variation to the difference between methods. 

![](/img/locatormap.png "Fig 1. Map locating plots at Pitney Meadows Farm and Hand Melon Farm in Upstate New York. ")

###### Fig 1. Map locating plots at Pitney Meadows Farm and Hand Melon Farm in Upstate New York.

In October of 2022, we took a group of Skidmore Environmental Studies and Sciences students to Hand Melon and Pitney Meadows for sample collection. Small groups of students were assigned to plots across each far and tasked with collecting paired samples, one with the probe and the other with the auger. We set up plots using a meter tape so that each plot contained 3 subplots that were 2 and 10 meters apart (Fig 2). The spatial resolution of a mobile phone GPS with good satellite connection is around 10 m so the subplots located 2 m apart would be less than the spatial resolution of a phone GPS, and therefore indistinguishable from each other, while the 10 m plots represent points that could in theory be distinguishable by phone GPS. Across the 9 plots and 2 farms, the students collected 324 samples in an afternoon (Fig 3). 

![](/img/plotdesign.png "Fig 2. Plot layout diagram demonstrating the spacing between samples within subplots and distance between subplots. Distances were measured with a meter tape at the site. ")

###### Fig 2. Plot layout diagram demonstrating the spacing between samples within subplots and distance between subplots. Distances were measured with a meter tape at the site.

![](/img/claire.jpeg "Fig 3. Claire Wologast using a TSIP auger and box to collect soil samples at Hand Melon Farm in Greenwich, NY. ")

###### Fig 3. Claire Wolgast using a TSIP auger and box to collect soil samples at Hand Melon Farm in Greenwich, NY.

Multiple pieces of data were collected for each sample point in our experiment. In addition to the soil samples, students also recorded how long it took to collect every sample sample, from the time that the point was located to getting the soil into a sample bag. Additionally, each sample was weighed to collect fresh weight to compare how much soil was collected with each tool. Finally, samples were shipped to an external lab and analyzed for total carbon by dry combustion. 

The critical finding from this study was that the difference in %C between the push probe and the auger is insignificant relative to the spatial heterogeneity of soil carbon on these farms. Although the methods didn’t produce exactly the same %C measurements, the difference between the two methods was consistently lower than the difference in %C between subplots located 10 m apart. This demonstrates that the effect the sampling method has on %C estimates is negligible relative to the natural spatial heterogeneity in %C at points that are distinguishable by a standard phone GPS and therefore can be used interchangeably without impacting farm scale C estimates. 

Another major finding was that the auger collected more soil on average than the probe. This is likely due to the fact that a plow pan at both farms prevented the probe from reaching the full 30 cm in many samples (Fig 4). Although in this study we found that the difference in %C between methods is not significant, at other farms not collecting 30 cm of soil could potentially increase the %C estimate due to the higher %C concentration in surface soils relative to deeper soils which tend to have lower %C concentrations. 

![](/img/soilprofiles.png)

###### Fig 4. Example of the auger (left panel) and probe (right panel) highlighting the ability of the auger to penetrate the plow pan which can be seen by the change in soil color. 

Finally, this study provided evidence to support our anecdotal experience that the auger is a faster sampling system than the probe. We found that the auger was 1.75-3x faster than the probe. When considering sampling large farms, this difference may provide substantial cost savings either in farmer time or technician labor. Overall, our study demonstrates that the TSIP system is a viable alternative to a push-probe, providing equivalent soil carbon estimates and significant time savings. So to answer your question, yes! You can use either soil harvesting system to collect soils, but we would definitely recommend  the TSIP auger system! 

## Citation

Jensen, K. H., Faehndrich, C. S., Colzani, E., McClure, M. L., & Covey, K. (2023). Rapid soil harvesting using a novel soil auger system for farm‐scale soil carbon estimates. *Soil Science Society of America Journal*, 88(1), 192–202.[ https://doi.org/10.1002/saj2.20603](https://doi.org/10.1002/saj2.20603)

Please email support@tsip.org if you would like a PDF.


Rapid soil harvesting using a novel soil auger system for farm-scale soil carbon estimates

## Written by: Apolonia Alvarez and Jacob Dube

Howdy y’all, it’s your NRELians Apolonia and Jacob coming to you from Golden, CO. We are both currently undergraduate students at Skidmore College interning at the National Renewable Energy Laboratory (NREL) in collaboration with The Soil Inventory Project (TSIP) for the summer under the renowned Lead Energy-Water-Land Analyst, Jordan Macknick.  

## "﻿Both organizations aim to utilize carbon data to support land management in maintaining healthy soils for agriculture and other regenerative land uses."

Renewable Energy is a sector both of us would like to further explore, hence, we were honored for the opportunity to study agrivoltaics at NREL alongside TSIP, soil sampling for total carbon, bulk density, pH, and macro and micronutrients. At NREL, we define agrivoltaics as the dual use of photovoltaics and agriculture to optimize different agricultural functions, such as crop production and grazing, while also increasing power production. With this partnership, we can build the carbon data for TSIP’s field database and contribute to NREL’s long-term analysis of solar infrastructure impact on soil health. This collaboration is especially strong as both organizations aim to utilize carbon data to support land management in maintaining healthy soils for agriculture and other regenerative land uses.  

![](/img/nrel1.jpeg)

######   Jacob and Apolonia at the National Renewable Energy Laboratory

During our time here, we have been able to soil sample at over 30 sites consisting of both agrivoltaics and standard solar in Colorado and Minnesota. We sampled sites around the Denver Metro Area, many of which were previously sampled by TSIP interns in 2022. To do this, we used a TSIP sampling kit with a standard drill to collect soil samples at 0-15 and 15-30 centimeter depths. For bulk density (BD) samples, we used a developing method of BD sampling with the potential to be cheaper and more accessible to landowners when compared to today’s options. This experimental process involved hammering a 12-inch length 3-inch diameter pipe into the ground to a depth of about 10 inches. We would then pull the core out of the ground and transfer the sample to a QR-coded gallon bag. Throughout the collection process, we quickly discovered just how hard and compact the Colorado soil can be. We broke a lot of the equipment within the first couple of sites which forced us to improvise quickly when out in the field. Luckily, we were able to effectively troubleshoot and get the job done. 

The sites we sampled at ranged from less than an acre up to 30 acres, with mechanisms as simple as fixed tracking and monofacial panels to single axis tracking and semi-transparent panels. Our bigger plots were at Denver International Airport in which we worked alongside Namaste Solar to do our fieldwork. Those sites were not actively doing agrivoltaics, but we saw some growing vegetation still happening under the panels. Our other bigger plots were located in Minnesota where we sampled at 3 pollinator agrivoltaic sites. Those sites had tall grasses, flowers, and native plants beautifully thriving under the panels. We would spend the day soil sampling and taking vegetation surveys, and, in the evening, we also got to explore Minneapolis and had the chance to attend a St. Paul Saints baseball game alongside representatives from U.S. Solar. 

![](/img/nrel2.jpeg)

###### **Apolonia and Jacob at the St. Paul Saints baseball game with representatives from U.S. Solar**

 Additionally, we also worked on the agrivoltaic site at NREL, the Bifacial Agrivoltaic Research at NREL (BARN), where we helped upkeep the pollinator habitat (thistle destroyers) and assisted with vegetation phenology and harvesting. To do this, every Thursday, we would go out with the team to harvest the selected crops for that week. In doing so, we would take measurements like mass, dimensions, and production stage of the plants. This data will be compiled to hopefully show which plants do best in an agrivoltaic setup. 

There are many benefits we learned about the use of agrivoltaics on farmland such as shading for vegetation, reduced irrigation, and an increase in energy production. We became quite familiar with agrivoltaic configurations during our time at NREL. Once a week we worked at Jack’s Solar Garden, an agrivoltaics research hub in Longmont, CO consisting of 5 acres of single axis tracking, monofacial solar panels that provide shade for pollinator and crop production research plots worked on by 9 partners. This site is primarily used as a proof of concept for solar developers from across the country about the efficiency of agrivoltaics. Being able to participate and visit such an inspiring facility was truly an honor and an experience we certainly won’t forget. 

![](/img/nrel3.jpeg)

###### C﻿ollecting a soil sample at a solar site

When we weren’t in the field, we were working on the fourth floor of the Research Support Facility (RSF), as a part of the Strategic Energy Analysis Center (SEAC). Our tasks consisted of updating the newly developed data portal for the InSPIRE team, a database on research related to agrivoltaics from around the world, working with fellow researchers on analyzing agrivoltaic potential in the U.S. and creating soil sample designs for our sites using ArcGIS Online and QGIS 3.32.10. 

Although our time at NREL was fleeting, this is an experience we will not forget. We hope to continue to collaborate with the laboratory and stay in contact with our new connections. Now with our newfound knowledge, we feel confident in our aspirations to further our education in the renewable energy sector and hope to share our findings soon!

Solar and Agriculture Integration. Sampling 30 Agrovoltaic Sites with NREL

## W﻿ritten by: Maggie Carmody and Sar Lindner

Greetings! Our names are Maggie Carmody (she/her) and Sar Lindner (they/them), and we are TSIP summer 2023 interns with Stonyfield Organic. Sar just received their Bachelor of Arts in Biology, and Maggie will receive her degree in Environmental Science next spring from Skidmore College. We were both drawn to this project because of our mutual interest in regenerative agriculture and sustainable farming. Working on dairy farms as field technicians, we were excited to have an opportunity to work directly with organic farmers and their beautiful cows.

![](/img/sf1.jpg)

###### M﻿aggie (left) and Sar (right) celebrating their last soil sample of the summer.

In late May, we set out with an ambitious schedule to take almost 2,000 samples over ten weeks on fourteen organic dairy farms in Stonyfield’s supply across Vermont and New York.  This sampling effort will establish a soil carbon baseline for Stonyfield's direct milk supply farms to better understand opportunities for increasing soil carbon sequestration as a lever to reduce the company's carbon footprint. Our sampling campaign involved drilling for 0-15 cm and 15-30 cm soil carbon samples once every half-acre and using a slide hammer to take a soil core for bulk density analysis at every other sample location. We sampled a wide variety of soils, from dense forests to open pastures. On some farms, we were also helping farmers take samples for their nutrient management plans– a state program for farmers to report on their soil stewardship. We converged in Cabot, Vermont on a dirt road at the small cabin that would be our first of many accommodations. Maggie met with Dana Bourne and Maddie Stein from Stonyfield and Morgan McClure and Kelsey Jensen from TSIP while Sar graduated in Saratoga! After a rainy afternoon prepping, we headed out into the first of many pastures we would be visiting. 

Armed with two packs of gear, hundreds of bags, and plenty of water, we began to navigate to our points scattered across farms. The Green Mountains of Vermont made for constantly gorgeous scenery. Whether it was hot and buggy or cool and breezy, we spent our days trekking through wildflowers. We always enjoyed lunch with a scenic backdrop. Most days also included cows: Jersey, Holstein, Hereford, and more! We met cows of all shapes, sizes, and temperaments. Learning the personalities of different herds week to week kept us on our toes and entertained in the field. One group of young cattle was so eager to lick at our soil sampling tools that we had to take shifts distracting them to get work done. Others were a bit more skeptical of our power tools and pop music and kept their distance. Some grazed in sunny meadows, others creekside in scrubby bits of forest. Regardless, they would all raise their heads at the clang of our slide hammer. 

![](/img/sf2.jpeg)

###### Curious cows watch a bulk density sample

Past the cows themselves, we learned a great deal about organic dairy farming. The main differences between conventional dairy farming and organic dairy farming are the agricultural processes used and how the cows are cared for. Many of the farms we visited were no-till operations. In lieu of tilling, farmers relied on the natural decomposition of plant material and manure to aerate their soils. In fields that had been tilled in previous years, we could feel a hard plow pan layer of soil when drilling just below 15 cm; the soil was dense and clay-heavy. The soils in no-till fields were soft, loamy, and fragrant. Other organic farms use cover cropping to cultivate their soil; one farmer even let us harvest some kale cover crop to bring home and eat. We additionally observed the use of rotational grazing on all of the organic farms we visited. The cows were moved through paddocks periodically to give the soil and vegetation time to regenerate before the herd came back to graze. 

## "﻿Our data represents only a small fraction of U.S. organic farmland, and the work of future soil sampling will continue to contribute to TSIP’s growing dataset and help quantify soil carbon worldwide."

As we continued to interact with the cows and learn about organic dairy production we developed a set of research questions we hope our data collection can help answer: How does carbon content impact dairy output? Are cows raised on organic farms more productive than those raised on conventional farms? How does tilling impact the carbon content of the soil? We hope the results of our data provide Stonyfield’s Direct Supply farms and TSIP with valuable information for land use management on dairy farms. Our data represents only a small fraction of U.S. organic farmland, and the work of future soil sampling will continue to contribute to TSIP’s growing dataset and help quantify soil carbon worldwide.

![](/img/sf3.jpeg)

###### C﻿ows grazing in an open pasture

This summer gave us the opportunity to go on exciting adventures while collecting our soil samples. We lived in Cabot, Barre, Middlebury, Montgomery, and Bennington, Vermont. We drove ATVs, escaped hail storms, and got up close to wild animals including deer, turkeys, and a coyote. We gained insight into what it’s like to be a dairy farmer, working outside all day surrounded by mud and manure. Despite being busy with their own hard work, everyone we met was incredibly generous and accommodating to us. We admire the passion they had for the health of their land and cattle. Although organic dairy management can be time-consuming and require initial investments, the difference shows in the healthy soil, happy cows, and delicious milk!

Measuring Soil Carbon for Organic Dairy Farmers with Stonyfield Organic

Written by Kelsey Jensen

### How do solar facilities affect soil carbon?



​​Demand for clean alternatives to non-fossil fuel derived energy has caused a dramatic increase in land needed to site onshore development of renewable energy facilities. The Biden-Harris administration has set an ambitious goal of generating at least 25 Gigawatts of electricity from wind, solar, and geothermal projects by 2025 (Energy Act of 2020). Photovoltaic (PV) solar facilities consist of large arrays of panels that convert solar energy directly to electrical current. The National Renewable Energy Lab estimates that 22,000 sq miles of land would be converted to solar arrays to power the entire country, although more recent data suggest that number could be closer to 10,000 sq miles (Ong et al 2013). This represents one of the largest land-use changes in the United States in this century. Solar facilities present an opportunity to achieve multiple environmental co-benefits when these arrays are located on degraded or marginal lands and managed with sustainable development goals in mind (Hernandez et al 2019). Locating solar arrays on former agricultural lands, for example, can allow for native plant community restoration increasing biodiversity and pollinator habitat while potentially also providing greater soil carbon sequestration and decreasing greenhouse gas emissions (Hernandez 2014a). There is great potential for environmental synergies in the deployment of renewable energy, but more research is needed to understand how to maximize these benefits. 

# The National Renewable Energy Lab estimates that 22,000 sq miles of land would be converted to solar arrays to power the entire country. This represents one of the largest land-use changes in the United States in this century. 



TSIP partners with the National Renewable Energy Laboratory and Argonne National Laboratory in field data collection and storage.



### How do solar arrays impact soil carbon?





Solar panels affect multiple aspects of the environment including ambient air temperature, evaporation, and overall water distribution (Armstrong et al 2014; Yue et al 2021). The shade created beneath a PV panel decreases temperature and evaporation in the spring and summer growing season and reduces the amount of photosynthetically active radiation reaching vegetation, which may limit plant growth. PV panels also intercept and redistribute rainfall unevenly due to the angle of the panel such that vegetation directly beneath panels receive less precipitation while vegetation on the low end of the panel (the “drip line”) receive more. These factors affect soil carbon by changing how vegetation at the site grows (how C enters the soil) and how organic matter decomposes (how C leaves the soil as CO2).

In some ecosystems, changes to the C cycle caused by solar arrays may result in a net increase of soil C while in other ecosystems there could be a net loss of soil C. For example, in desert ecosystems, the shade created by PV panels could promote plant growth and increase organic matter input to the soil while in temperate ecosystems, the reduced sunlight under panels may decrease plant growth. TSIP is working to improve our understanding of these patterns in soil C by sampling from solar facilities across the country.



### How is TSIP contributing to solar soil research? 





In the past three years, TSIP has sampled more than 20 solar facilities across Colorado, Minnesota, New York, and Florida. The facilities range encompass a wide range of soils, with total C concentrations from less than 0.1% to over 5%. When we collect these samples, we strategically collect soils from both underneath PV panels and in the alleys between panels. We also collect meta-data on the facilities such as time since establishment, methods of site preparation, current management, and land use history. We then use these data to understand drivers of soil carbon variation, and how future sampling can be conducted more efficiently to capture the greatest amount of variation with the fewest samples.





###### ![](https://lh3.googleusercontent.com/_-nO1O3TYldrZxkNSEAqwfkhCLspBtR6F38X5Twr0EZgg3hjKIi5Ng8bMNh2UO08sldwCZRrwqGBLOQsp0WH3MkCSXhocuSBhmKPE-ai-NR5_RMOOVjVK6lqaB1bBmcDwEOpkKuT0MgeOgn4Vr2Zbm8)Fig 2. Total soil carbon concentration at 0-15 cm and 15-30 cm in depth from solar facilities sampled across four states. Groups are arranged by increasing mean %C with Florida having the lowest average concentrations and Minnesota having the highest average concentrations of total carbon. The distribution of %C in our current data set will allow us to answer questions about sample design that are specific to different regions and soil characteristics.



Sample designs at solar arrays look different from agricultural lands or non-disturbed environments due to the methods used to prep sites for PV panel construction, such as removing vegetation and bulldozing (Hernandez et al 2014b; Choi et al 2020). The standard method TSIP applies in determining where to take samples relies partly on remote sensing data, such as NDVI (a measure of “greenness”), that can help us understand underlying variation in the soil. If a solar facility has removed vegetation and disturbed the top layer of soil, this metric may not be a useful reflection of the current soil conditions. To address this question, we have sampled some sites using a basic grid and others using conditioned Latin HyperCube sampling and Polaris organic matter estimates. Using these data, we will be able to determine not only which method provides the best estimates with the fewest samples, but also how this changes with variables like solar array size or soil carbon concentration. In the future, we will be able to recommend sample density and design based on the site location, soil characteristics, and land management. Providing this information will decrease the cost of soil sampling for solar utility companies and increase sampling accessibility within this industry.



![](https://lh5.googleusercontent.com/HYOjAXXGUZWUNQNUeUnrlqQv8RILSy4ztNkNUM3zuS9kvimQqvjJjbRNnIwPr8aTYS0Tu-xIKu-10AkiDTTd36ztPUDZK922jbYxloqCt-w5rGlzY-J4Ld6Cio1Kb1t1kKpE-ePa1cua--DJaSVTVd0)

###### Fig 3. Soil carbon concentrations at 0-15 cm depth collected from samples between or under solar panels. Preliminary results do not show significant differences in soil C relative to PV panel location, however, this may change over time as environmental effects of PV panels alter soil carbon cycling at these sites.



Future research questions include:

* How does soil C concentration vary with PV panel microsite location (under panels, between panels, or in the dripline)?
* How do different types of vegetation management, such as mowing or using animal grazing, impact soil C stocks?
* How does soil C relate to other soil health metrics at solar facilities?

### Soil Carbon Sequestration in Solar Facilities





Aggregating high resolution soil carbon data from solar facilities is a step toward sustainable solar management. TSIP is partnered with Argonne National Lab to collect and store soil data from solar facilities across the country through the Ground-Mounted Solar and Soil-Related Ecosystem Services Project (aka, Solar Soils). In addition to playing a critical role in developing data storage and security solutions for this project, TSIP will use these data to answer questions about how soil characteristics, land use history, and vegetation management affect soil carbon sequestration at a solar facility. Lands that were previously used for agricultural purposes may have high potential for soil carbon sequestration if restoration of diverse, native prairie vegetation is promoted within the facility (Perry et al 2023). Recent research on sheep grazing for vegetation management at solar facilities demonstrates the potential benefits for soil C and available nutrients (Towner et al 2022). The net effects of PV panels on greenhouse gas emissions from soil are yet to be determined but will rely on accurate measurement of soil C to inform process-based models (Armstrong et al 2014, 2016; Guoqing et al 2021). TSIP aims to provide the high resolution soil data needed to answer critical questions about where solar arrays should be located and how they should be managed as this land-use type continues to grow rapidly. 

![](/img/img_6481.jpeg)

###### Photos from Jacob Dude.

### Learn More!





There is tons of interesting research happening related to the environmental impacts of solar facilities- from soils to pollinators! Checkout these resources if you want to learn more about the research being done on this topic:

[Wild Energy Project](https://wildenergy.org/) at UC Davis

[Gemini Project](https://www.usgs.gov/centers/southwest-biological-science-center/science/gemini-solar-project#overview) at USGS Southwest Biological Research Center 

[Argonne National Lab](https://www.anl.gov/evs/article/optimizing-renewable-energy-development-and-ecosystem-services) Renewable Energy and Ecological Services 



### References



Armstrong, A., Waldron, S., Whitaker, J. & Ostle, N. J. Wind farm and solar park effects on plant–soil carbon cycling: uncertain impacts of changes in ground-level microclimate. Global Change Biology 20, 1699–1706 (2014).

Armstrong, A., Ostle, N. J. & Whitaker, J. Solar park microclimate and vegetation management effects on grassland carbon cycling. Environ. Res. Lett. 11, 074016 (2016).

Choi CS, Cagle AE, Macknick J, et al Effects of Revegetation on Soil Physical and Chemical Properties in Solar Photovoltaic Infrastructure. Frontiers in Environmental Science (2020).

Guoqing, L. et al. Ground-mounted photovoltaic solar parks promote land surface cool islands in arid ecosystems. Renewable and Sustainable Energy Transition 1, 100008 (2021).

Hernandez, R. R. et al. Environmental impacts of utility-scale solar energy. Renewable and Sustainable Energy Reviews 29, 766–779 (2014a).

Hernandez, R. R., Hoffacker, M. K. & Field, C. B. Land-Use Efficiency of Big Solar. Environmental Science & Technology 48, (2014b).

Ong, S., Campbell, C., Denholm, P., Margolis, R. & Heath, G. Land-Use Requirements for Solar Power Plants in the United States. NREL/TP-6A20-56290, 1086349 http://www.osti.gov/servlets/purl/1086349/ (2013) doi:10.2172/1086349.

Perry S, Falvo G, Mosier S, Robertson GP Long-term changes in soil carbon and nitrogen fractions in switchgrass, native grasses, and no-till corn bioenergy production systems. Soil Science Society of America Journal (2023).

Towner E, Karas T, Janski J, et al.  Managed sheep grazing can improve soil quality and carbon sequestration at solar photovoltaic sites. Oral Presentation. The American Geophysical Union Fall Meeting (2022).

Yue, S., Guo, M., Zou, P., Wu, W. & Zhou, X. Effects of photovoltaic panels on soil temperature and moisture in desert areas. Environ Sci Pollut Res 28, 17506–17518 (2021).

![]()

TSIP partnerships to generate ecological insights in solar facilities

## W﻿ritten by: Hana Sparks-Woodford and April Cook 

Hi everyone, our names are Hana and April, and we are both Class of 2024 environmental science majors at Skidmore College. During the 2022-2023 school year we were introduced to The Soil Inventory Project (TSIP) through a research project. Our role was to collect data for a global-scale meta-analysis exploring how various grazing practices influence soil organic carbon levels. Last summer, we both interned on local farms near our hometowns; Snipes Farm and Education Center in Morrisville, PA (Hana) and Stony Kill Farm Environmental Education Center in Wappingers Falls, NY (April). In our experiences as a farmer and environmental educator, respectively, we gained a passion for organic agriculture which we hoped to explore further with TSIP summer research. 

![](/img/cff1.jpeg "April (left) and Hana (right) with a bulk density sample. Photo by Dylan Fishbein, Sales & Marketing Manager at Caney Fork Farm")

###### April (left) and Hana (right) with a bulk density sample. Photo by Dylan Fishbein, Sales & Marketing Manager at Caney Fork Farms.

This summer we interned for TSIP at[ **Caney Fork Farms**](https://www.caneyforkfarms.com/) (CFF) in Carthage, Tennessee. Like TSIP, CFF values the importance of soil health: “\[Caney Fork Farms’] mission to is to inspire others with living proof that regenerative, organic farming in Tennessee benefits farmers, the community, and the environment.” The farmers are committed to improving soil health and ecological processes through their work in regenerative agriculture and agroforestry. TSIP has supported this mission through soil carbon sampling, mapping, and analysis for the past five years. As the fifth year of student interns at CFF, our work will contribute to the growing soil carbon database which reflects the effects of regenerative practices over time. This ongoing research monitors carbon during the transition from conventional to regeneration practices as well as compares carbon content between properties, practices, and land cover. This summer, we spent six weeks conducting research and four weeks working directly with the livestock and vegetable teams at CFF. In addition to being trained in farming and fieldwork, we’ve also made relationships which will last a lifetime.  

## "﻿Some of our greatest takeaways from soil carbon sampling are to always carry spare drill batteries, to take time to look up from the soil and appreciate the beautiful surrounding landscape, and always prepare a good playlist the night before."

In our six weeks of research, we collected soil samples on 3 properties; the CFF Main Farm, the satellite farm, Lock Seven, and a neighboring farm that does not practice holistic planned grazing. For the first 3 weeks of research, we collected soil samples for carbon and soil health analysis. We explored every corner of CFF’s property while using a push probe and TSIP Sampling kit including a rotary hammer to collect samples from 0-15 cm and 15-30 cm at over 200 locations. Some of our greatest takeaways from soil carbon sampling are to always carry spare drill batteries, to take time to look up from the soil and appreciate the beautiful surrounding landscape, and always prepare a good playlist the night before (your music will set the mood for the day).  

Next, we monitored Chestnut trees physiology within CFF’s agroforestry system. We monitored over 1500 Chestnut trees two times throughout the summer. In June, when the trees began to bloom, we monitored the catkins to estimate the reproductive stage of each Chestnut tree. In August, we monitored the nut set of each tree to estimate the overall nut yield, a month prior to harvest. We also took note of the pest pressure and disease presence within each individual tree. In looking at the results from our soil carbon sampling, it will be fascinating to investigate whether the presence of Chestnut trees within the agroforestry system impact carbon content!  

![](/img/cff4.jpeg "Moving animals for rotational grazing")

###### Moving animals for rotational grazing

For the last week of research, we used two methods to collect bulk density samples within the Main Farm, Lock 7, and the neighboring property. We conducted a comparative study on the efficiency of an AMS gas powered probe versus a new bulk density collector created by Professor Kris Covey. This method uses a rotary hammer to drive a steel tube into the soil. We tested this system to help develop a new low-cost, easily distributed bulk density system. We preferred the rotary hammer bulk density sampling because it does not require a time-intensive set-up, like the probe, leading us to use only this method for sampling the remaining properties. One highlight of our summer was conducting a bulk density sampling demonstration of both methods for CFF’s research team, when they visited the farm at the start of August.  

During our summer, we gained first-hand experience with the intersection between research and farming practices within a regenerative agriculture system. We each spent two weeks on the veggie team, assisting with all aspects of vegetable production including seeding, transplanting, bed-prep, weeding, harvesting, packaging, and marketing. One of our favorite days of the summer was working the CFF farm stand at the East Nashville Farmers Market. It was fantastic interacting with customers and gaining a better understanding of the Community Supported Agriculture (CSA) program. The CSA demonstrates the importance of local farms connecting community members to the land they live on. It is clear that CFF exemplifies the principles of regenerative agriculture and fills an important role within the climate justice movement, minimizing agriculture’s massive carbon footprint.  

During our two weeks working with livestock, we were responsible for monitoring cattle, sheep, and pig operations as well as managing the silvopastures. We participated in animal health monitoring, setting up electric portable fencing, and moving animals within the rotational grazing system. From our research at Skidmore, we know that rotational grazing is one of the most effective practices for improving soil carbon sequestration. It was exciting to make connections like this between the papers we read and the practices that CFF is implementing. Animal health monitoring was our favorite aspect of livestock work: we vaccinated over 100 lambs, trimmed the hooves of 75 ewes, conducted pregnancy tests on 15 cows, and treated over 40 cows for pink eye. It was a laborious and hot month, but extremely rewarding. Moving forward, our understanding of CFF land management practices which we gained from farming will help us better analyze the data we collected for TSIP.  

![](/img/cff3.jpeg)

###### Hana and April trimming the hooves of an ewe

It was an honor to be a part of a community so devoted to stewarding the land. To quote one of Caney Fork Farms core values, “*doing the hard work that matters"*, seriously does matter! The seemingly inconvenient or labor-intensive practices employed on a regenerative farm, like Caney Fork Farms, have such significant positive impacts in terms of mitigating the carbon footprint associated with conventional agriculture in the United States. We will carry this lesson with us into everything we do in the future.

Soil Research and Regenerative Farming at Caney Fork Farms

## W﻿ritten by: Dominick Catoggio and Jonathan Ramirez

I’m Dominick, an environmental science major at Skidmore College. Last semester I took Kris Covey’s Regenerative Systems class. While learning about soil science, I became aware of The Soil Inventory Project's (TSIP) internship programs. I knew this program would be a great way to get involved with fieldwork that’s closely related to my studies. I’m very grateful to Ocean Hour Farm (OHF) and TSIP for giving me this opportunity to gain hands-on experience.   

I’m Jonathan, and I’m also an environmental science major at Skidmore. Last summer, I did research with a team of students and professor Karen Kellogg. We presented our research at the National Renewable Energy Lab (NREL), and it was there where I met fellow students working with TSIP. I became interested in interning with TSIP because of my already fantastic experience doing research. Needless to say, this summer did not disappoint me!  

![](/img/ohf1.jpeg "Dominick (left) and Jonathan (right) with a herd of cows ")

###### **Dominick (left) and Jonathan (right) with a herd of cows**

We spent the majority of our internship at** [Ocean Hour Farm](https://oceanhourfarm.org/)** in Newport, Rhode Island with additional work at neighboring farm Aquidneck Farms. Our work included collecting 285 soil samples for carbon and nitrogen analysis, as well as 68 bulk density samples across the two properties. We additionally had personal research projects and farm duties that deepened our connection to the farm and staff there. We had frequent meetings with them to discuss progress with our personal projects and worked with staff from 11th Hour Racing during weekly volunteer days to help plant new crops or remove weeds.   

For the nitrogen and carbon samples, we used the TSIP sampling kit, including a battery-operated hand drill with auger attachments instead of a traditional probe. We tested a new system using a rotary hammer to collect bulk density samples. We’d collect samples by clearing away vegetation and drilling to a marked line on the attachment so we’d get a consistent amount of soil each time. Overall, the data collection went smoothly but we also had a few challenges while working in the field. The soil is shallow in many areas, and rock ledges are frequently visible from the surface. If rock gets in the way of our drill, we need to start over. This happened many times, but we were able to find clean spots almost every time. The bedrock also broke our bulk density collection device, but luckily, we were prepared for things to break. TSIP was able to swiftly get us the equipment we needed.   

![](/img/ohf2.jpg "Rocky landscape at Ocean Hour Farm")

###### Rocky landscape at Ocean Hour Farm

Ocean Hour Farm serves as a center for education, scientific research, and a demonstration of regenerative agricultural practices, highlighting the critical connection between land and ocean health. We are eager to receive the data that will provide the basis of a long-term, yearly, study to determine if OHF's soil health improves as a result of their regenerative practices. A key indicator of this improvement would be an increase in soil carbon levels in previously degraded areas. In order to gain further insights, we aim to compare the data collected from OHF's soil with the data from another nearby farm, Aquidneck Farms.  

In addition to our soil sampling, we were assigned a farm duty. During sunset, we helped put the chickens back into their mobile coop. In our spare time, we also helped Ocean Hour Farm educate students about biodiversity. Dom built plant presses to preserve plants that can be presented in a book or framed on the wall. Pressed plants can be stored in an herbarium, which is a collection of plants, accompanied by their location and defining characteristics. In the future, OHF can catalog all of the plants found around the property to visualize the amount of biodiversity on display. This project can help students with plant I.D. beyond the farm.   

Jonathan helped the farm research ways to create a biodiversity index. The index aligned with their goal of being a center of research. OHF could use an index to keep track of biodiversity changes in the future as farming practices change. Without much background knowledge, the work consisted of reading papers and determining if different methods would work on the property. Jonathan also assisted the education program manager, Beth Alaimo, in creating lesson plans and activities to make learning environmental science fun for middle and high school students.  

Other than technical matters, we were also very lucky that the folks at both locations were very welcoming, and we immediately felt like a part of the team. One of our favorite community events was the celebration of the summer solstice. Everyone from OHF contributed to making a large dinner with farm fresh produce and pastured chicken. We were split into teams and gathered ingredients from the farm to create a unique side dish. The food turned out amazing and we got to see firsthand just how tight knit the community is. 

![](/img/ohf3.jpeg "Spread of food from the summer solstice dinner ")

###### Spread of food from the summer solstice dinner

We frequently visited the office (referred to as the Milk Barn) to give updates on our progress, and ask questions. This aided us in our scheduling, so we could budget our time effectively. Our schedule was always subject to change, which encouraged us to make quick decisions. For example, when our bulk density equipment broke, we quickly pivoted to a new method of data collection with the help of TSIP. The amount of collaboration that goes into each job at both Ocean Hour Farm, and TSIP is inspiring and something we will take away from this experience.

A Coastal Farm Measures Soil. Sampling at Ocean Hour Farm

## **Predicting soil organic carbon by a handheld reflectometer, dry combustion, and remotely-sensing and land databases**

If it worked, spectroscopy via a hand-held reflectometer could be relatively quick, standard, and inexpensive—but we observed substantially less predictive power than via surface satellite imagery. We collaborated with researchers at Texas A&M University, Noble Research Institute, The Yale School of the Environment, and Caesar Kleberg Wildlife Research Institute to [study the efficacy of handheld spectroscopy to predict soil carbon](https://www.mdpi.com/2571-8789/6/2/38)—running a proverbial model-off. 

## **Sampling and Machine Learning the Southern Great Plains**

Dr. Goodwin and the research team collected soil samples from 519 location on 19 beef cattle ranches in the Southern Great Plains across approximately 43,000 acres. This comprised 1738 soil samples at 6 depths in Northeastern Texas and southern Oklahoma. They used the Stratifi web application to determine sampling locations via stratified random sampling and Google Earth Engine.

The researchers dried and ground these with soil grinder, passing them through a <2 mm screen. They analyzed the samples by the Our Sci handheld reflectometer, by the conventional lab method of dry combustion, and pulled information for these locations from satellite imagery and USDA databases.

## **Results from the hand-held spectroscopy accounted for half the variability in soil organic carbon.**

The result from dry combustion showed that mean soil organic carbon ranged from .03 to 6.03%, but averaged .96%.

The team modeled the results from the reflectometer and, separately, the remotely sensed variables in Bayesian Additive Regression Trees (BART)—a best-fit, decision tree, machine learning approach. They ran each model 1000 times with different testing and training data splits.

![](/img/remote-vs-in-situ.png "From Goodwin et al. (2022)")

###### From Goodwin et al. (2022)

BART-based models on reflectometry results accounted for slightly more than half the variability in soil organic carbon values, with R2 = .54. Models based on remotely-sensed information and digital soil layers performed substantially better than those based on the handheld visNIR spectroscopy alone, These had R2 = .71 and .75, respectively.

The visNIR readings did not effectively predict soil carbon values. While other research reports high model fits of hand-held reflectometers at detecting soil organic carbon—there is high variability in predictive power.

## **Spectroscopy reads chemistry and structure**

Spectroscopy evaluates materials’ interaction with radiation. Light interacts with materials chemically and physically, absorbing or reflecting based on the content of the material it encounters.

At a very fine scale, light can identify items down to the element. Electrons change at different energy levels, emitting radiation of different colors, in the same glow discharge as neon lights.

Soil structure can also affect results. Soil particle size and shadowing, and plant residue size, texture, and shadowing, and quartz content can affect results. Texture, which creates shadows and can vary under 2mm, can change soil samples reflectances, without a change in the material’s physical or chemical properties. So, reading chemistry from spectroscopy affected by structure, can be difficult.

But, it’s also possible that carbon is not registered at all.

![](/img/soil-texture-dark.jpg "Variations in soil texture affects reflectometer readings.")

###### Variations in soil texture affects reflectometer readings.

## **How does spectroscopy read, or not read, carbon?**

[Mcbride (2021)](https://bsssjournals.onlinelibrary.wiley.com/doi/abs/10.1111/ejss.13192?casa_token=hbeM0pHHhVYAAAAA:mEZjDGhljKGa9gZpIeaDip9RXcvMR3s2gZHNrGmm4J9LMUxqsk1PccEH50BAxavJRwX01_Io25w7pG8j) beautifully and cogently explains what is happening physically, chemically, and in the spectral reflectances—in case you’re interested!

Light ranges from short wavelengths to long wavelengths. Infrared is longer than the wavelengths of visible light, and less than that of radio or microwave. Near infrared is closely to visible light, and mid infrared has substantially longer wavelengths. Visible and near infrared light together is referred to as visNIR.

Carbon can be detected by spectroscopy in the mid-infrared range of radiation. Within the range of mid infrared radiation, bonds between elements of organic matter, specifically between carbon, oxygen, hydrogen, and nitrogen, stretch, bend, compress, and twist when subjected to energy, vibrating. These organic compounds can be complex, especially when you have complex chemical constructs – such as amines and carboxylic acids and benzene rings ([Reusch, 2013](https://www2.chemistry.msu.edu/faculty/reusch/virttxtjml/Spectrpy/InfraRed/infrared.htm#ir1)). This registers as complex, but, it appears, relatively clear, signals in the mid infrared spectral signature. **Organic matter components vibrate in the mid infrared range of radiation.**

**However, the handheld reflectometer in this study did not measure mid-infrared, but instead visible to Near Infrared radiation.** The open-source Reflectometer measures reflectance at 10 wavelengths, from 370 to 940 nm, or from ultraviolet to a short-wavelength near infrared.

Within the visNIR range of radiation, some of these bonds, in specific compounds or arrangements, can alter in symmetry, registering as complex effects within the visNIR (Reusch, 2013 —please forgive if I’ve gotten any of this wrong) But they are composed of varied combinations of materials, and do not make a single clear mark or change in the spectra.

So, some carbon bonds change within visNIR reflectances ranges. But which bonds are present depend on the structures within which they are embedded. Is it clay? Humus?

Registering bond changes within this range may capture specific types of carbon, across a wide range of possible complex structures. A model using this information could identify a correlation between this structure, like clays, and soil organic carbon  - but it would likely not comprehensively account for all types of possible soil organic carbon. As a consequence, this soil carbon correlation might only be true for that site or soil type. We note that while this study observed lower accuracy with higher values of soil organic carbon, other researchers have observed the reverse. This may reflect the composition of the model training dataset.

A model based on this data requires this relationship to exist, and for it to be consistent in the data. It may be true within a county, or region, but not necessarily at a continent scale – or even across much closer soil types.

## **Handheld visNIR spectroscopy for carbon may not be based on comprehensive and replicable relationships**

These results suggest that data from a visNIR reflectometer may not, depending on soil types and dynamics, effectively predict soil organic carbon content.

The best use of a hand-held reflectometer might be at a small scale, where soils are less variable in texture and composition—and after accuracy testing. While data from the mid infrared range of radiation may more accurately predict soil organic carbon, these spectroscopic tools are more expensive and difficult to transport.

A model based on visNIR readings might be based on a relationship that does not exist elsewhere, and is not comprehensive. If readings are taken in the field, and samples are not dried first, reading carbon is even more difficult due to light-absorbing water.

While traditional laboratory analysis of samples takes some time between sample collection and results analysis—it is the current standard. TSIP uses dry combustion in the lab to measure soil carbon from field samples.

### Remote Sensing

The remote sensing and database information used in this study included soil chemical and physical properties from the USDA’s Soil Survey Geographic Database (SSURGO). It included NDVI and topographic indices from the USGS’ national elevation dataset. We used satellite imagery from the European Space Agency’s Sentinel 2 satellites. Importantly, this analysis did not include or try to account for management effects.

### Funding

Noble Research Institute, LLC., funded the research.

### Work Cited

[Goodwin, D. J., Kane, D. A., Dhakal, K., Covey, K. R., Bettigole, C., Hanle, J., ... & Tolleson, D. R. (2022). Can Low-Cost, Handheld Spectroscopy Tools Coupled with Remote Sensing Accurately Estimate Soil Organic Carbon in Semi-Arid Grazing Lands?. *Soil Systems*, *6*(2), 38.](https://www.mdpi.com/2571-8789/6/2/38)

McBride, M. B. (2022). Estimating soil chemical properties by diffuse reflectance spectroscopy: Promise versus reality. *European Journal of Soil Science*, *73*(1), e13192.

Reusch, W. (2013). Virtual Text of Organic Chemistry: Infrared Spectroscopy.

Should We Use Hand Held Spectroscopy to Measure Soil Carbon? The Method Did Not Effectively Predict Soil Organic Carbon in the Southern Great Plains

## **Why Soil Field Data?**

Why handle soil to know soil carbon?

Hold soil and you feel variations in texture, moisture, and material. You can tell when there is something soft or something granular. A soil in Florida composed of sand and waterlogged, fine plant material, and a highly weathered, silty soil in the Mid-Atlantic feel different in your palm. And they function different on the land.

Soil locations, compositions, and functions fill public databases and define applied chemical and mathematical relationships. From centuries of science we know soil dynamics, health, and distributions across spatial scales. We know that microbial activity changes with climate, and synthetic nitrogen can remain in groundwater for decades. Soil texture limits carbon chemical bond-formation, and roots exude enzymes at depth. Accomplished farm managers grow on and, sometimes, restore, healthy, productive, and complex land. With all this knowledge, we still need soil field data.

Soil field data allows scientists and land managers better understand soils’ dynamic conditions, make informed decisions about its use, and monitor changes over space and time (Smith et al., 2020; Bradford et al., 2021; NASEM, 2021).

There is limited soil field data for science and management for three reasons. Firstly, it comes from the subsurface. We cannot readily see or measure what is *within* the ground, unlike patterns *across* our planet. Secondly, It is physically difficult and expensive to sample soils. It takes strength and care to sample soil from depth. This can mean driving metal tubes, spades, or augers into working lands—which is hard over tens over thousands of acres. Lastly, scale, variability, and uncertainty affect results. Values from a field, or plot, are different from county or state values.

Trying to know what is happening in a field or farm can be difficult if that site differs from existing regional or continental patterns. How can you estimate what effect management has had on your property and in your county?

Through TSIP, we sample and collect more field data, and combine it with existing collections that are not currently public.

## **Why Distributed Sampling?**

A farmer in Indiana, or a project developer from Illinois, can go into a field—maybe a field that they’ve known for years—and follow pins to unbiased sampling locations. Distributed soil sampling means that people who know the land, communities, and specific locations can collect and report field data. Existing networks can manage and collect field information and work within trusted relationships. Ideally, distributed sampling avoids airline flights, costs of housing, and the challenges of people landing in new places.

Distributed data collection also allows us to streamline sampling design, data analysis, and reporting, creating and applying flexible but place- and management-based approaches. It means that we can apply a standard but multi-branched decision tree of approaches, in collaboration with our partners.

However, it is absolutely critical to train samplers so that sampling can be collected following rigorous standards set by USDA NRCS.

TSIP has also spent multiple summers sending college students to sampling sites to work with partners. They analyze the data and study the science in coursework prior to and following the sampling—contributing to the research and learning.

![](/img/points.jpg "Points on an unknown surface")

###### P﻿oint on an unknown surface

## **Models!**

Soil field data is a set of numbers in time and space. But modeling can move those numbers through both, spatially extrapolating results and forecasting future soil conditions. It places a value from a place and a day in one location into process dynamics. Soil organic matter transforms and changes differently with different conditions, depending on its origins—lignin and humus behave differently. In a systems approach, a soil value is a single expression of a soil-plant-atmosphere-and management dynamic network. Discrete pieces of ground data inform the full extent of a multidimensional projection through time.

If you shake a blanket, waving it in the air, and one point on the cloth is at a certain height and location, the rest of the fabric location and color flows from that point.

Field data without modeling is like knowing a few points on the blanket with the rest of the material invisible.

Preliminary findings from modeling at the Basso Lab at Michigan State University show that soil tilled minimally for 30 years, from 1991 to 2020, can contain up to 5 to 15 Mg more soil organic carbon (SOC) per hectare than that of the same soils tilled conventionally. It predicts that most areas across the state of Michigan can contain 10 to 20 Mg SOC per hectare more when under conservation tillage. 

The difference is equivalent to nearly 100,000 – 200,000 miles driven by a gasoline-powered car per year, per hectare. 

This modeling shows the possibility and scale of converting management. It suggests that if you drove past this farmland for an hour, you would pass 110 billion miles worth of emissions—20,000 round trip, coast-to-coast, drives across America.

![Modelled GHG benefits of sustainable agriculture in Michigan state. Left: Simulated SOC in Mg ha-1 for 0-26 cm in 2020 with conventional tillage. Right: Simulated SOC in Mg ha-1 for 0-26 cm in 2020 with minimum tillage. From Bruno Basso.](/img/presentation2.jpg "Modelled GHG benefits of sustainable agriculture in Michigan state. Left: Simulated SOC in Mg ha-1 for 0-26 cm in 2020 with conventional tillage. Right: Simulated SOC in Mg ha-1 for 0-26 cm in 2020 with minimum tillage. From Bruno Basso.")

###### Modelled GHG benefits of sustainable agriculture in Michigan state. Left: Simulated SOC in Mg ha-1 for 0-26 cm in 2020 with conventional tillage. Right: Simulated SOC in Mg ha-1 for 0-26 cm in 2020 with minimum tillage. From Bruno Basso.

## **Distributed field data launches models, includes managements, increases access, and informs science.**

We collect, and share systems to collect, field data for four reasons:

1. To initialize, calibrate, and validate models.
2. When management is not currently encompassed by extant models, to attribute impact to management. Some farming methods, especially traditional, small-scale, and complex approaches do not yet fit well into current modeling. That shouldn’t prevent them from being recognized for their work and impact, especially when entire landscapes are composed of varied systems!
3. To provide farmers’ access to soil and soil carbon data. Where there are barriers to data and the results of data due to limits of access, technology, analysis and science, it is often possible to remove them. If we don’t include farms and farmers, they cannot be recognized, we won’t know what contributions they make, and it can leave a barrier between investment and farm.
4. To be available for further scientific advances.

## **Making it happen with new methods.**

The Soil Inventory Project ties data to models. In the course of this work, we develop tools and methods. These include flexible sampling design, data entry on phones, quick analysis methods, machine learning emulators, and multi-model ensembles.

The central idea of pairing distributed field data and modeling is simple. But implementation calls for leveraging emerging technologies and methods to make connections between fields, data, and modeling possible—and accessible.

#### **Work Cited**

Bradford, M. A., Wood, S. A., Addicott, E. T., Fenichel, E. P., Fields, N., González-Rivero, J., et al. (2021). Quantifying microbial control of soil organic matter dynamics at macrosystem scales. Biogeochemistry, 156(1), 19-40.

National Academies of Sciences, Engineering, and Medicine. 2021. Exploring a Dynamic Soil Information System: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26170.

Smith, P., Soussana, J., Angers, D., Schipper, L., Chenu, C., Rasse, D. P., et al. (2020). How to measure, report and verify soil carbon change to realize the potential of soil carbon sequestration for atmospheric greenhouse gas removal. Global Change Biology, 26(1), 219- 241.


Distributed Soil Sampling for Modeling Increases Access—and Science

Lorem ipsum dolor sit amet consectetur. Ut nisi cursus faucibus nulla justo imperdiet ipsum quam. Varius egestas ut sapien blandit suspendisse nulla. Ac a erat volutpat faucibus ut eget congue. Justo rutrum sed elementum urna adipiscing. Purus semper morbi tellus senectus. Ullamcorper id pretium consectetur laoreet tempor hendrerit in aenean dis. Aliquam turpis lectus vehicula commodo leo auctor non quis. Amet felis nulla praesent viverra. Sem sem blandit tellus ut est at eget. Ultrices ac ut dolor et dictum at vitae cras sit.

Erat iaculis vitae lectus pulvinar aenean iaculis sollicitudin ut id. Eros sodales sed velit felis lobortis commodo nulla pellentesque. Enim tellus ridiculus est gravida cursus tellus laoreet. Duis neque risus elementum aliquet lacus. Augue dolor gravida et congue lectus ut blandit.

![Soil Carbon](/img/rectangle-34.jpg)

## Lorem ipsum dolor sit amet consectetur. 

Lorem ipsum dolor sit amet consectetur. Ut nisi cursus faucibus nulla justo imperdiet ipsum quam. Varius egestas ut sapien blandit suspendisse nulla. Ac a erat volutpat faucibus ut eget congue. Justo rutrum sed elementum urna adipiscing. Purus semper morbi tellus senectus. Ullamcorper id pretium consectetur laoreet tempor hendrerit in aenean dis. Aliquam turpis lectus vehicula commodo leo auctor non quis. Amet felis nulla praesent viverra. Sem sem blandit tellus ut est at eget. Ultrices ac ut dolor et dictum at vitae cras sit.

Erat iaculis vitae lectus pulvinar aenean iaculis sollicitudin ut id. Eros sodales sed velit felis lobortis commodo nulla pellentesque. Enim tellus ridiculus est gravida cursus tellus laoreet. Duis neque risus elementum aliquet lacus. Augue dolor gravida et congue lectus ut blandit.

Methods

Rapid soil harvesting using a novel soil auger system for farm-scale soil carbon estimates

Field Reports

Solar and Agriculture Integration. Sampling 30 Agrovoltaic Sites with NREL

Field Reports

Measuring Soil Carbon for Organic Dairy Farmers with Stonyfield Organic

Our Work

TSIP partnerships to generate ecological insights in solar facilities

Field Reports

Soil Research and Regenerative Farming at Caney Fork Farms

Field Reports

A Coastal Farm Measures Soil. Sampling at Ocean Hour Farm

Methods

Should We Use Hand Held Spectroscopy to Measure Soil Carbon? The Method Did Not Effectively Predict Soil Organic Carbon in the Southern Great Plains

Our Work

Distributed Soil Sampling for Modeling Increases Access—and Science