Skip to content

Groundwork · Science & research · by Saurav Das

Groundwork.

One question keeps becoming the next.

From microbes in Assam’s groundwater to soil health in Nebraska and four decades of farming systems in Pennsylvania. A life of looking beneath the surface, and connecting what I find.

Papers here
16
In the thread
5
From the trial
5
Chapters
3
Years
2012–present

01The thread

Five papers, one question at a time.

Each study answered part of a question and left the next one open. Read them in order, or jump to a step.

  1. 01Soil Health Gap2020
  2. 02Ecological units2022
  3. 03Reference sites2024
  4. 04Two-Tier framework2026
  5. 05Hierarchical targets2026 · preprint
  1. Question 01 of 052020Define improvement

    Better soil — compared with what?

    The Soil Health Gap frames soil condition as a distance from an undisturbed reference within an agroecosystem. It gives improvement a benchmark, rather than treating an isolated test result as a verdict.

    Leaves openThe reference and indicator matter; a gap is not a universal soil-health score.

    Published paper

    Soil Health Gap: A concept to establish a benchmark for soil health management

    Maharjan, Das & Acharya · Global Ecology and Conservation · 2020

    Read the paper

    But soil and climate change the potential. Which sites can we fairly compare?

  2. Question 02 of 052022Make comparison fair

    Which soils belong in the same comparison?

    The Cropland Reference Ecological Unit (CREU) groups land with comparable soil-forming and climatic conditions. It helps separate differences caused by management from differences a field inherits from its environment.

    Leaves openCREU supplies ecological context; it does not make every soil or management history identical.

    Published paper

    Cropland Reference Ecological Unit: A land classification unit for comparative soil studies

    Das & Maharjan · Ecological Indicators · 2022

    Read the paper

    A comparable area is only the start. Where, within it, should the reference come from?

  3. Question 03 of 052024Choose the benchmark

    What makes a reference defensible?

    Reference-site selection uses ecological site descriptions and state-and-transition models to identify the appropriate reference plant community. The benchmark becomes an ecological condition to establish, rather than the nearest patch that looks undisturbed.

    Leaves openReference selection still requires field evidence and knowledge of disturbance history.

    Published paper

    Reference site selection based on state-and-transition models for soil health gap evaluation within cropland reference ecological units

    Das, Hird, Maharjan, Stephenson & Kariyawasam · Soil Security · 2024

    Read the paper

    With a defensible reference in place, which measurements are useful enough to repeat?

  4. Question 04 of 052026Measure & interpret

    Can measurement be consistent and interpretation local?

    Tier 1 identifies a practical core: organic matter, pH, nitrate-N, and bulk density. Tier 2 interprets those measurements against ecologically matched reference distributions. The framework separates what to measure from what a result means in a particular soil and climate.

    Leaves openThe reference dataset comes from Nebraska; broader transfer needs regional testing and calibration.

    Published paper

    A two-tier, reference-informed framework for soil health assessment: Minimum indicator sets and ecologically dynamic thresholds

    Das, Gamage, Stephenson, Acharya, Daigh & Maharjan · Agrosystems, Geosciences & Environment · 2026

    Read the paper

    An ecological reference describes potential. What is a realistic next step for working cropland?

  5. Question 05 of 052026Find the next stepPreprint

    How do we turn potential into attainable progress?

    Hierarchical Land Classification connects ecological reference, recovery, soil-building, production, and degraded states. A synthesis of 1,059 soil organic carbon observations from 105 studies examines intermediate targets between today's condition and ecological potential.

    Leaves openPreprint. Recovery and soil-building states were not statistically distinguishable; targets are neither universal nor equally spaced.

    Preprint · not peer reviewed

    Hierarchical Land Classification for Soil Health Assessment: Linking Reference Potential, Management State, and Incremental Soil Health Targets

    Das, Bora, Panday, Maharjan, Smith, Mamo, Hird, Singh & Garrity · SSRN · 2026

    Read the preprint

    The open question: do these context-aware targets predict useful outcomes across more farms and regions?

Still open

Can a fairer benchmark lead to a better decision in the field?

Where research meets building

A reading path through related ideas, not a citation network. Connections are conceptual; years are publication dates.

02The field trial

One field since 1981. Three ways to farm it.

The benchmarking thread keeps asking “compared with what?” The trial is one long, carefully kept answer.

I direct the Farming Systems Trial at Rodale Institute in Kutztown, Pennsylvania: the longest-running side-by-side comparison of organic and conventional grain cropping in North America. Same field, same weather, replicated plots, different management, for more than four decades.

  1. Conventional

    A typical grain farm

    Rotation
    Corn and soybean
    Fertility
    Synthetic nitrogen and herbicides
  2. Legume-based organic

    An organic cash-grain farm without livestock

    Rotation
    Corn, soybean, and wheat with cover crops
    Fertility
    Legume cover crops and rotation
  3. Manure-based organic

    An organic dairy or beef farm

    Rotation
    Feed grains with a perennial clover–hay phase
    Fertility
    Composted manure and legume cover crops

Since 2008, each system has also been split into full and reduced tillage.

Five readings of one record

  1. AYear-34 snapshot2025
  2. BForty-year carbon record2026 · preprint
  3. CCarbon footprint2025
  4. DSoil to grain2026
  5. EThe farm budget2026
  1. Reading A of E2025Year-34 snapshot

    After 34 years, is there more carbon under organic management?

    Soil organic carbon and total nitrogen were measured to 30 cm in the trial’s 34th year. Only the manure-based organic system had higher carbon concentrations and stocks, and only in the 0–10 and 10–20 cm layers. Compared on an equivalent-soil-mass basis to 30 cm, that advantage did not hold.

    Read with careA single year, to a single depth. A snapshot can understate a difference that builds slowly or sits in a narrow layer.

    Published paper

    Soil organic carbon and total nitrogen after 34 years under conventional and organic management practices at the Rodale Institute Farming Systems Trial

    Lorenz, Omondi, Lal, Das & Smith · Soil Science Society of America Journal · 2025

    Read the paper
  2. Reading B of E2026Forty-year carbon recordPreprint

    What does the whole record say, down to a metre?

    Seven sampling campaigns from 1984 to 2023, three of them through the full metre. Corrected to equivalent soil mass, the organic systems stored clearly more carbon than the conventional system. Rotation diversity and organic inputs mattered far more than tillage: reduced tillage, compared since 2008, had no significant overall or system-dependent effect.

    Soil organic carbon by system Mg C/ha

    Top 20 cm, equivalent soil mass

    • Conventional54.7
    • Legume-based organic65.7, +11.0 over conventional
    • Manure-based organic70.3, +15.6 over conventional

    Whole profile, 0–100 cm

    • Conventional92.9
    • Legume-based organic108.9, +16.0 over conventional
    • Manure-based organic119.0, +26.1 over conventional

    Differences are against the conventional system. One scale for both depths, so the bars compare directly.

    Source: Das, Spell, Smith & Garrity (2026), EGUsphere preprint; values as reported in the abstract.

    Read with carePreprint under review. Most of the difference is shallow: 91–94% of the extra carbon under the organic systems sat in the top 30 cm in the 2021 and 2023 profiles.

    Preprint · not peer reviewed

    Cropping-system diversification drives long-term soil carbon gains more strongly than reduced tillage

    Das, Spell, Smith & Garrity · EGUsphere · under review in SOIL · 2026

    Read the preprint
  3. Reading C of E2025Carbon footprint

    Counted as emissions, how do the systems compare?

    Carbon footprints for the three systems were built with two models, COMET-Farm (Tier 3) and the Cool Farm Tool (Tier 2). Conventional had the highest area-scaled emissions, 1.25–1.72 t CO₂-eq per hectare per year. The manure-based system was 25–37% lower and the legume-based system 52–74% lower. Maize yield-scaled emissions were also highest under conventional management.

    Area-scaled emissions relative to conventional % of conventional
    • Conventional1.25–1.72 t CO₂-eq/ha/yr
    • Manure-based organic25–37% lower
    • Legume-based organic52–74% lower

    The pale segment spans the reported range. Not shown: counting composting CO₂ raised the manure system by about 103–122%.

    Source: Graham, Ghalehgolabbehbahani, Das, Carr & Smith (2025), Renewable Agriculture and Food Systems.

    Read with careThe boundary changes the answer: the Tier 3 model ran about 27% higher, and counting CO₂ released while composting manure raised the manure system’s emissions by roughly 103–122%.

    Published paper

    Carbon footprint of organic and conventional arable crop production systems in a long-term trial

    Graham, Ghalehgolabbehbahani, Das, Carr & Smith · Renewable Agriculture and Food Systems · 2025

    Read the paper
  4. Reading D of E2026Soil to grain

    Does soil management reach the grain we eat?

    Ergothioneine is a dietary antioxidant closely tied to soil microbes. Reduced tillage raised grain ergothioneine across all three systems, consistent with more arbuscular mycorrhizal activity. The organic systems held more microbial biomass and residual soil ergothioneine, yet moved less of it into grain; microbial biomass correlated only weakly with grain levels.

    Read with careThe highest grain levels came from conventional reduced-tillage plots, where herbicide use raised oxidative stress. More biology in the soil did not automatically mean more in the food.

    Published paper

    Ergothioneine dynamics in a long-term farming systems trial: Effects of tillage, management, and oxidative stress on grain and soil ergothioneine

    Das, Omondi, Hinson, et al. · Journal of Agriculture and Food Research · 2026

    Read the paper
  5. Reading E of E2026The farm budgetPosition paper

    Can a farm afford the system?

    A review of six long-term comparisons, the Farming Systems Trial and trials in Minnesota, Wisconsin, Iowa, Maryland, and California, spanning more than four decades. Conventional systems keep a yield edge early, especially during the transition. Over the long term, organic systems frequently match or exceed conventional profitability, and rotation design decides it: organic rotations with a perennial or forage phase competed even without price premiums; short, annual-only organic rotations did not.

    Read with carePosition paper, not peer reviewed. Diversified organic systems carried lower downside risk, though not uniformly lower year-to-year variability.

    Position paper

    The Economic Performance and Profitability of Organic and Conventional Farming Systems: A Review of Long-Term Comparative Research

    Das & Smith · Rodale Institute · 2026

    Read the paper

03Connecting the dots

Where the field meets the thread.

Five patterns that only show up when the readings are laid side by side, and the ideas elsewhere in my work they meet.

  1. 01

    The trial is a benchmark built into the ground.

    Across farms, the benchmarking thread has to separate what a field inherits, its soil and climate, from what management did. Ecological units and reference sites exist for that. In the trial the separation is designed in: the systems share one field and one weather record in replicated plots, so the distance between them is management. The conventional system is a working baseline, and the gap can be read directly.

  2. 02

    How you count decides what you see.

    Measured once to 30 cm, the carbon difference shrank to the manure system’s upper layers and did not hold on an equivalent-soil-mass basis. Measured across forty years, seven campaigns, and a full metre, an 11–16 Mg C/ha gap appeared in the top 20 cm. The same systems’ emissions moved by about a quarter with the choice of model, and roughly doubled for the manure system once compost was counted. The readings differ in year, depth, and design, which is exactly the lesson: measure consistently, then interpret.

  3. 03

    Soil carbon keeps time in decades.

    Spread across the 42 years since the plots began, the topsoil gap works out to about 0.26–0.37 Mg C/ha a year. That is my arithmetic, assuming the systems started level in 1981. Real, durable, and mostly within the top 30 cm. Targets for working farms have to be paced to that clock, which is why intermediate, attainable states matter more than a distant ideal.

  4. 04

    One practice can move one outcome and not another.

    Reduced tillage did not significantly change soil carbon in the forty-year record, yet it raised grain ergothioneine in every system. The organic systems carried more microbial biomass, yet moved less ergothioneine into grain. An indicator is only as useful as its pathway to the function you care about.

  5. 05

    Rotation is the lever that keeps showing up.

    Carbon tracked rotation diversity and organic inputs more than tillage. Across six long-term trials, profit tracked rotation design more than the organic label: rotations with a perennial or forage phase held their own without premiums. In this field, the manure-based system, with compost and a perennial clover–hay phase, held the most carbon. Different currencies, one lever: what grows in the field, and for how long.

What the trial asks next

Which part of a system, its rotation, its inputs, or its disturbance, moves which outcome, and does the answer travel beyond one field in Pennsylvania?

One site, one climate: the trial is a deep record, not a wide one. That is where the benchmarking thread’s regional question comes back in.

Back to the regional question

Connections are my reading across the papers, not findings of any single one. The per-year figure is arithmetic on the reported stocks, not a measured rate.

04The journey

Water, then soil, then systems.

Three places that changed what I ask and how I work.

  1. Chapter 012012–2017

    Water

    Learning to see the invisible

    Assam · Assam Agricultural University

    What is happening in the water we cannot see?

    Doctoral work in agricultural biotechnology took me into arsenic biogeochemistry, groundwater, and microbial communities in Assam. It built a habit that still shapes my research: look beneath the visible symptom and ask which living processes are driving it.

    What I carried forward

    Follow the microbes. Pay attention to context. Start with a real problem in the field.

    GroundwaterMicrobial ecologyArsenic

    Explore the groundwater study
  2. Chapter 022018–2024

    Soil

    Learning to ask “compared with what?”

    Nebraska · University of Nebraska–Lincoln

    What does healthy soil mean in this particular field?

    Postdoctoral and research faculty work moved those questions into agricultural soils: benchmarking, carbon and nitrogen, and the effects of management. Different soils kept resisting the same scorecard. That tension runs through the Soil Health Gap, ecological units, and reference-site work.

    What I carried forward

    A measurement needs a fair comparison. A field has both a history and a potential.

    Soil healthCarbon & nitrogenReference sites

    Follow the benchmarking thread
  3. Chapter 032024–present

    Systems

    Making the knowledge useful

    Pennsylvania · Rodale Institute, Farming Systems Trial

    How does evidence become something a farmer can use?

    Directing the Farming Systems Trial put a four-decade record of organic and conventional systems, side by side in one field, at the centre of my work. I connect it with practical indicators, attainable targets, and tools that make evidence easier to use: how can careful science travel beyond a paper and become part of an everyday decision?

    What I carried forward

    Connect ecological potential, present condition, and the next useful action.

    Farming systemsLong-term evidenceResearch into tools

    Read the field trial

05Other threads in the work

Mechanisms, microbes & the bigger picture.

6 more papers around the thread: evidence, mechanisms, and where soil health claims hold up.

  1. 2026Position paper

    Soil Health in Practice: Principles, Proof, and the Path Forward

    A systems-oriented look at how soil health moved from niche concept to organizing framework, and what kinds of evidence are still needed to connect management practices to measurable outcomes.

  2. 2025Sustainability

    Soil Carbon Sequestration: A Mechanistic Perspective on Limitations and Future Possibilities

    Examines where soil carbon narratives are scientifically strong, where they become overstated, and what mechanistic understanding is required for credible long-term sequestration claims.

  3. 2024Agrosystems, Geosciences & Environment

    Soil Health Cycle

    A framework for thinking about soil health as an iterative management cycle linking practices, indicators, functions, and human decision-making rather than as a one-time score.

  4. 2024PLOS One

    Nitrogen Dynamics as a Function of Soil Types, Compaction, and Moisture

    Investigates how compaction and moisture regimes shape nitrogen transformations, leaching, volatilization, and emissions across contrasting soil textures.

  5. 2018BMC Microbiology

    Characterization of Siderophore Producing Arsenic-Resistant Staphylococcus sp. Strain TA6

    A study of arsenic-resistant bacteria from contaminated groundwater in Assam and their possible role in the local arsenic geocycle.

  6. 2017Genomics Data

    A Metagenomic Approach to Decipher the Indigenous Microbial Communities of Arsenic Contaminated Groundwater of Assam

    An early metagenomic investigation into the microbial communities inhabiting arsenic-affected groundwater systems, reflecting the foundation of my research journey.

The journey is a collaborative one

Another dot to connect?

A research question, a shared field site, or an idea worth testing. I’d like to hear it.

All publications ↗Start a conversation ↗More of the story ↗