Home Activities About Us About Biochar Resources Projects Contact Us
Conversation With AI on Biochar
Initiative News

Conversation With AI on Biochar

February 22, 2026  ·  admin
Home Posts Conversation With AI on Biochar

Q & A with Anthropic’s Claude Sonnet 4.6 on Biochar 2/22/26 and 3/7/26

By Raymond Baltar

I recently read what I consider to be an important essay on the current state of AI titled “Something Big is Happening” by an industry insider named Matt Schumer. While this is somewhat of an obvious statement to anyone who has been following AI’s most recent models and the industry’s massive investments in infrastructure, in the (long) piece he lays out his view that we have reached a point where AI is improving on itself, at an ever accelerating pace.  This is increasing both the speed for model improvements and their capabilities. The warnings by other industry insiders of likely massive job disruption—particularly in good paying, middle class jobs in law, finance, journalism, and tech—are real. While it is not time to panic, it is time to start learning how to use and leverage these increasingly capable tools in our own businesses and industries or risk being left behind by those who are.  

While I have been guilty as many others my age of using ChatGPT and Claude primarily as a better replacement for Google search, based on the Shumer essay I recently experimented with Anthropic’s Sonnet 4.6 to rebuild a website, and create a complex data logger for our pyrolysis operations at the Napa Recycling facility that can be accessed online. I was truly blown away by how well both were crafted, and with the back-and-forth interactions I had with the model to make minor style corrections or simply improve on things I wanted to add or change. This thing was downright friendly!

While these were relatively simple, basic requests, and I did not even use Anthropic’s most capable model Opus 4.6 that can handle much more complex, multi-stage tasks, I can see that we are indeed already in a world where folks with little or no training and $20 bucks a month can produce highly technical (and even creatively interesting) work.  I simply can’t imagine where we will be in 2 to 5 years.

Are there major environmental, energy, and moral issues associated with the development of AI? Absolutely! I believe that the companies developing this industry, and likely the government, need to spend much more money and effort addressing these problems. And certainly the dangers we face for AI’s use for military applications and by bad actors is very real. But that discussion will require another post, and likely on another forum.

Anyway, with all this in mind I decided to ask Sonnet 4.6 some questions about the biochar industry, and you can decide for yourself how helpful this information might be. At SBI we have a long history of trying to move the biochar agenda forward so most of my questions are focused on what this particular AI “thinks” about this topic. While many of these answers will be familiar to those that are actively involved with some aspect of the biochar ecosystem, and of course some may be subject to dispute, Overall, in my humble opinion,  the answers are both credible and insightful and may be particularly helpful for those starting on their own biochar journey.

____________________________________________

Conversation #1.   2/22/26

RB:  Looking through a financial sustainability lens, what are the five most important factors that would influence scaling the current biochar industry?

Sonnet 4.6: This is a rich question — and the answer shifts depending on whether you’re looking at it from a producer, investor, or policy standpoint. Here are the five factors I’d argue matter most, ranked by leverage:

1. Carbon Credit Market Maturity Biochar’s financial viability at scale hinges heavily on whether high-quality, durable carbon removal credits command a stable and sufficient price. Right now the voluntary carbon market is fragmented — prices for biochar credits range wildly, verification standards vary between registries (Puro.earth, Gold Standard, Verra), and corporate buyers remain skittish after high-profile greenwashing scandals in other credit categories. Until there’s a liquid, trustworthy market with prices consistently above ~$150–200/tonne CO₂e, most biochar operations can’t close their financial models on carbon revenue alone. Compliance markets — like California’s cap-and-trade or the EU ETS — are the prize, but biochar isn’t yet a recognized offset methodology in most of them.

2. Feedstock Cost and Supply Chain Reliability Biochar economics are brutally sensitive to feedstock. Forest slash, agricultural waste, and urban wood waste are theoretically abundant — but aggregating, transporting, and processing them at consistent quality is expensive and logistically complex. Operations that can co-locate with a reliable waste stream (a sawmill, a municipal green waste facility, a food processor) have a structural cost advantage. The industry can’t scale without stable, low-cost feedstock contracts, which in turn requires longer-term partnerships with feedstock generators who currently have competing disposal options (burning, landfill, composting).

3. Co-Product Revenue Diversification Biochar-only revenue models are fragile. The operations with the strongest unit economics are those that monetize multiple outputs from the same pyrolysis process — biochar for soil amendment, heat energy for drying or on-site use, syngas for electricity generation, and potentially bio-oil for industrial applications. This co-product stack is what makes pyrolysis financially competitive with incineration or composting. Investors and lenders are far more likely to finance facilities that aren’t dependent on a single revenue stream — particularly one (carbon credits) that’s still price-volatile.

4. Permitting and Regulatory Clarity The ARTi unit at American Canyon is a perfect case study here — four years of permitting for one containerized unit. That timeline is a capital killer. Air quality permits, feedstock classifications, and land use approvals vary enormously by jurisdiction, and the regulatory frameworks in most states weren’t written with small-scale distributed pyrolysis in mind. Until permitting pathways are standardized and streamlined — especially for mobile and containerized units — the time-to-revenue risk will keep many investors on the sidelines and limit the speed at which operations can replicate across sites.

5. Agronomic Value Certification and End-Market Development Carbon credits get most of the attention, but the long-term financial sustainability of the industry may depend more on whether biochar commands a consistent premium as a soil amendment product. Right now buyers range from premium vineyards willing to pay $800–1,200/tonne for certified biochar to commodity compost operations that won’t pay much above the cost of alternatives. Building that end market requires standardized product certification (IBI and EBC standards help but aren’t universally recognized), documented agronomic ROI that farmers can bank on, and distribution infrastructure that doesn’t yet exist at scale. A robust agricultural end market would de-risk the business model significantly — it’s a revenue stream that doesn’t depend on policy or voluntary corporate commitments.


The through-line across all five is de-risking the revenue stack — the industry needs at least two or three of these factors to mature simultaneously before institutional capital flows in at the volume needed for real scale. Right now biochar sits in an awkward valley: too proven to be speculative, but not proven enough at scale to be bankable by conventional standards. Organizations like SBI that are building the demonstration record, policy relationships, and market education infrastructure are doing the foundational work that makes eventual scaling possible.

________________________________________________

RB: What are the top 5 markets for biochar in the SF Bay Area? In California? In the whole US?

Sonnet 4.6: Here’s a breakdown of the top biochar markets at each geographic level, based on current research and industry reports:


SF Bay Area — Top 5 Biochar Markets

The Bay Area is a hub for sustainability-focused buyers. The leading markets here are:

  1. Urban & Community Agriculture / Market Gardens — Small farms, urban growers, and CSAs around the Bay are increasingly using biochar for soil amendment, driven by strong consumer demand for organic produce.
  2. Landscape & Green Infrastructure — City and county projects (bioswales, urban forestry, parks) use biochar in tree planting mixes and stormwater filtration systems. SF and Oakland have been active here.
  3. Composting Operations / Compost-Biochar Blends — Bay Area composters (e.g., Recology) are exploring biochar additions to finished compost for nutrient retention and carbon sequestration credits.
  4. Carbon Credit / Voluntary Carbon Markets — The Bay Area tech sector’s heavy interest in carbon offsets makes this a significant demand driver; companies and startups purchase biochar carbon removal credits.
  5. Water/Wastewater Filtration & Remediation — Bay Area municipal utilities and environmental consultants use biochar for contaminant removal (heavy metals, nitrates) in stormwater and groundwater treatment.

California — Top 5 Biochar Markets

Research on California’s biochar market identifies soil-based agricultural applications as the most likely near-term market, followed by filtration, livestock feed, and manure management. ResearchGate More specifically:

  1. Agriculture / Soil Amendment (Central Valley) — By far the largest market. California’s Central Valley, with its massive almond and orchard industry, is a key region where converting agricultural biomass waste to biochar creates significant economic and environmental opportunity. Springer
  2. Carbon Markets / Carbon Sequestration Credits — California’s cap-and-trade program and voluntary carbon markets make it one of the most active states for biochar carbon removal purchases.
  3. Wildfire Debris & Forest Residue Management — CAL FIRE and state agencies are actively exploring converting forest slash and fire-damaged biomass into biochar to reduce fire risk and generate a revenue stream.
  4. Stormwater, Water Filtration & Soil Remediation — State regulations around water quality drive demand for biochar in municipal filtration and brownfield remediation projects.
  5. Composting / Organic Waste Diversion — California’s strict organics recycling mandates (SB 1383) are pushing compost producers to explore biochar-compost blends to improve product value and reduce emissions.

United States — Top 5 Biochar Markets

Agriculture is the dominant application nationally, accounting for over 72% of revenue share in 2025. Grand View Research The top five markets are:

  1. Agriculture / Soil Amendment — The largest and most established market, covering row crops, orchards, vineyards, and regenerative farms across the Midwest, Southeast, and West.
  2. Carbon Removal / Voluntary Carbon Credits — Favorable government initiatives and increasing funding for biochar R&D, combined with partnerships between research institutes and manufacturers, have fueled strong demand for biochar in carbon markets. IMARC
  3. Animal Feed & Livestock — The U.S. agricultural community is actively exploring biochar to improve animal health and reduce methane emissions from ruminants, integrating it into livestock feed or bedding. Inkwood Research
  4. Water & Environmental Remediation — Biochar is used in municipal stormwater systems, mine drainage treatment, and groundwater cleanup; a growing area as EPA regulations tighten.
  5. Horticulture, Turf & Specialty Growing — Golf courses, nurseries, greenhouses, and sports turf operations use biochar for water retention and root zone improvement — a niche but fast-growing segment.

Overall trend: The U.S. biochar market was valued at $543 million in 2024 and is projected to reach $1.38 billion by 2033, driven by growing demand for sustainable agriculture, carbon sequestration goals, and waste-to-resource initiatives. IMARC California, with its agricultural scale, climate policies, and tech-sector carbon demand, is positioned as one of the most important state-level markets in the country.

________________________________________________

RB: Which of these markets, if scaled, would provide the most jobs? And the most GHG emissions savings?

Sonnet 4.6: Great strategic questions. These two dimensions — jobs and GHG savings — don’t always point to the same market, so let me break them down carefully.


Most Jobs if Scaled

Winner: Agriculture / Soil Amendment

This market, if scaled, would generate the most jobs across the entire value chain:

  • Feedstock collection & logistics — gathering agricultural waste, wood chips, forest residues requires significant rural labor
  • Biochar production facilities — pyrolysis plant operators, technicians, QA staff
  • Application & distribution — spreading biochar on farmland is labor-intensive at scale, especially in specialty crops (orchards, vineyards)
  • Agronomic consulting — farmers need guidance on application rates and blending

Agriculture wins because it’s spatially distributed — you can’t centralize it. Hundreds or thousands of small-to-mid-scale production units near feedstock sources would be needed, multiplying jobs across many regions. In California alone, the Central Valley could support dozens of regional biochar facilities.

Close second: Wildfire / Forest Residue Management

This market is particularly compelling for job creation because it addresses two crises simultaneously — unemployment in rural timber-dependent communities and catastrophic wildfire risk. Mobile pyrolysis units deployed across national and state forests would require crews for:

  • Slash pile collection and chipping
  • Mobile unit operation and maintenance
  • Carbon monitoring and verification

These jobs would be concentrated in underserved rural communities that desperately need economic diversification.


Most GHG Savings if Scaled

Winner: Wildfire Debris & Forest Residue Management

This one isn’t close, for a few compounding reasons:

  1. Direct carbon sequestration — biochar is roughly 70–90% stable carbon that persists in soil for hundreds to thousands of years, locking away CO₂ that would otherwise return to the atmosphere
  2. Avoided emissions from decomposition — forest slash left on the ground decomposes and releases CO₂ and methane over years
  3. Avoided wildfire emissions — catastrophic wildfires are among California’s single largest annual GHG emission sources; some years California wildfires emit more CO₂ than the entire electricity sector. Converting that biomass to biochar before it burns is a massive avoided-emissions opportunity
  4. Feedstock scale — California alone has an estimated 20+ million acres of fire-prone forests with enormous volumes of burnable biomass

Close second: Carbon Markets / Carbon Removal Credits

While this is more of a financing mechanism than a sector unto itself, scaling structured carbon credit purchasing would directly fund the highest-sequestration biochar projects regardless of end use, essentially acting as a multiplier on GHG savings across all other markets.

Third: Agriculture / Soil Amendment

Agricultural biochar sequesters meaningful carbon, but the GHG math is somewhat diluted compared to forest biomass — agricultural feedstocks are smaller in scale, and some biochar applied to soil may be tilled or disturbed. Still significant at scale.


The Big Picture

Market Jobs Potential GHG Savings Potential
Agriculture / Soil Amendment ⭐⭐⭐⭐⭐ ⭐⭐⭐
Wildfire / Forest Residue ⭐⭐⭐⭐ ⭐⭐⭐⭐⭐
Carbon Markets ⭐⭐ ⭐⭐⭐⭐
Composting / Organics ⭐⭐⭐ ⭐⭐⭐
Water & Remediation ⭐⭐ ⭐⭐

The strategic sweet spot — if you’re trying to maximize both jobs and GHG savings simultaneously — is forest residue / wildfire biochar, especially in California. It hits rural job creation, fire risk reduction, carbon sequestration, and avoided wildfire emissions all at once. It’s arguably the highest-impact single biochar application available, and it’s currently underleveraged relative to its potential.

________________________________________________

RB: What are the biggest barriers to scaling biochar, and why are the best pathways to break through them?

Sonnet 4.6: Biochar has genuine potential as a carbon removal and soil amendment technology, but it faces a cluster of interconnected barriers that have kept it from scaling despite decades of interest.

The core economic problem

Biochar is expensive to produce — typically $200–600+ per tonne depending on feedstock, pyrolysis technology, and scale — but commands low prices in most markets. Agricultural buyers might pay $50–150/tonne as a soil amendment, and carbon credits have historically priced in a similar range. This means most producers can’t cover costs without subsidies or stacking multiple revenue streams simultaneously. The pathway through this is twofold: voluntary carbon markets need to mature and price biochar credits more accurately (reflecting its multi-century durability, which should command a premium over lower-permanence credits), and the co-benefits of biochar — improved water retention, reduced fertilizer needs, methane suppression in livestock — need to be better quantified and monetized.

Feedstock supply and quality

Sustainable, affordable feedstock — agricultural residues, woody biomass, sewage sludge, food waste — is geographically dispersed, inconsistent in quality, and often already claimed by competing uses like composting or direct energy. Some feedstocks introduce contaminants (heavy metals, PAHs) that limit where the biochar can be applied. Solving this requires co-locating production with large consistent feedstock sources (timber mills, food processors, wastewater treatment), developing regional aggregation networks, and creating clear regulatory standards that give buyers confidence in product quality. The EU’s EBC (European Biochar Certificate) and IBI standards are steps in the right direction, but adoption is inconsistent.

Carbon credit credibility and MRV

Unlike direct air capture, biochar’s carbon accounting is complex. How long does the carbon actually stay sequestered in a given soil and climate? What’s the net lifecycle balance accounting for transport and production emissions? Inconsistent methodologies have created a credibility problem — some early biochar credits were arguably oversold. The pathway here is rigorous, standardized MRV (monitoring, reporting, and verification) frameworks, which organizations like Puro.earth and more recently the CRCF in Europe are working toward. Better permanence modeling tied to specific biochar recalcitrance indices (like the H:Corg ratio) is making this more tractable.

Capital and scale-up risk

Pyrolysis plants require meaningful upfront capital ($1–20M+ depending on scale), and investors face technology risk, feedstock risk, and market risk simultaneously. Most biochar producers are small, preventing the economies of scale that would reduce costs. The breakthrough here is probably less about novel technology and more about modular, replicable plant designs that lower the cost of capital by reducing bespoke engineering, paired with long-term offtake agreements from corporate buyers committed to carbon removal portfolios (Microsoft, Stripe, Shopify, etc. have been important here).

Agronomic uptake and farmer behavior

Even where biochar is available and priced competitively, farmer adoption is slow because the benefits are highly soil- and crop-specific, often taking years to materialize, and there’s limited local agronomic knowledge. Extension services rarely cover it, and farmers are rationally skeptical of novel inputs. Long-term field trials and demonstration projects — particularly in degraded tropical soils where the benefits are largest and fastest — are necessary to build the evidence base and word-of-mouth adoption.

The most promising integrated pathway

The strongest near-term scaling model is probably pyrolysis plants co-located with industrial biomass waste streams (sawmills, rice mills, palm waste facilities) in regions with degraded agricultural soils and supportive carbon policy. Revenue stacking — carbon credits + soil amendment sales + heat/energy recovery from the pyrolysis process — is what makes the economics work. Coupling this with credible corporate carbon removal buyers who provide advance market commitments reduces investor risk enough to unlock capital. Over the medium term, if carbon markets price permanence properly, biochar’s 500–1000 year sequestration advantage over biological approaches becomes a decisive economic differentiator.

The honest summary is that biochar doesn’t face a single hard technical barrier — the technology works. It faces a coordination problem: feedstock logistics, carbon market infrastructure, agronomic knowledge diffusion, and capital formation all need to develop in parallel, which is harder to solve than any single technical challenge.

________________________________________________

RB: What about permitting for production facilities, both in local air districts and USEPA regulations?

Sonnet 4.6: Permitting is actually one of the most underappreciated barriers to biochar scaling in the US, and it can add years and significant cost to a project even when the economics otherwise work.

The fundamental regulatory ambiguity

Biochar production through pyrolysis doesn’t fit neatly into existing regulatory categories. Depending on how regulators classify the process — is it waste processing? biomass energy generation? a chemical manufacturing process? — entirely different regulatory frameworks can apply, and the classification can vary by jurisdiction, feedstock, and whether energy is being recovered. This ambiguity itself is a barrier because it creates uncertainty that chills investment before a single permit application is even filed.

Local and state air district permitting

This is often the sharpest practical obstacle. Pyrolysis produces syngas, carbon monoxide, volatile organic compounds, particulate matter, and depending on feedstock, potentially dioxins, furans, and heavy metal-bearing emissions. Local air quality management districts — particularly in California under CARB and its regional districts like the Bay Area AQMD or San Joaquin Valley APCD — apply stringent New Source Review requirements and Best Available Control Technology standards that can require extensive emissions testing, stack monitoring equipment, and afterburner or scrubbing systems that significantly increase capital costs.

The problem is that most air districts have little to no experience with pyrolysis specifically. Permitted categories like “biomass combustion” or “industrial boiler” don’t map cleanly, so permitting staff often default to the most conservative interpretation, applying requirements designed for much larger or dirtier processes. A small modular biochar unit producing a few tonnes per day can end up facing the same permitting scrutiny as a large industrial facility, eliminating the economic viability of smaller distributed systems — which are often the most logical configuration given feedstock dispersal.

Conditional use permits from local land use authorities layer on top of air permits and add another venue where neighbors, local governments, or advocacy groups can raise concerns about odor, truck traffic, and visual impacts, even for facilities with strong emissions controls.

Federal EPA dimensions

At the federal level, several regulatory programs create friction depending on facility configuration and feedstock. The most significant is the distinction between biomass and waste feedstocks under RCRA. If a facility processes materials that could be classified as solid waste — certain agricultural residues, food waste, sewage sludge — it may trigger RCRA permitting requirements, which are substantially more burdensome than air permits alone. Producers using clean woody biomass can often avoid this, but the line is not always clear and EPA regional offices interpret it differently.

The Clean Air Act’s Major Source thresholds under Title V are another concern for larger facilities. A facility emitting above certain tonnage thresholds for criteria pollutants becomes subject to full Title V operating permits, which involve federal review, public comment periods, and ongoing compliance monitoring that can take two to four years to navigate. Most biochar facilities are designed to stay below these thresholds, but it constrains facility sizing decisions in ways that work against economies of scale.

EPA’s regulations around co-firing and energy recovery from pyrolysis gas also create complications. If a facility burns syngas for heat or electricity — which is often essential to the thermal efficiency and economics of the operation — it may trigger additional New Source Performance Standards depending on how that combustion unit is classified.

Sewage sludge biochar is its own problem

Pyrolysis of sewage sludge (biosolids) has attracted significant interest because municipalities are desperate for alternatives to landfilling or land application under increasingly restrictive biosolids regulations, especially given PFAS contamination concerns. But the regulatory picture here is particularly complex. EPA’s 503 regulations govern biosolids, and there is currently no clear regulatory pathway for biosolids-derived biochar as a distinct product category. Some producers have argued that pyrolysis destroys PFAS and produces a clean product, and the science is increasingly supportive, but EPA has not formally addressed this, leaving state regulators to make inconsistent determinations.

Pathways through the permitting barriers

The most effective near-term strategies tend to involve a few approaches. Engaging air districts early and collaboratively — before filing formal applications — to help regulators develop familiarity with the technology pays dividends, even though it adds time upfront. Some producers have worked with districts to establish pilot operating permits that allow emissions characterization before full permit requirements are set, which benefits both parties.

Industry associations like the US Biochar Initiative have been working to develop model permit language and engage with EPA and state agencies on clearer categorical guidance. This kind of pre-competitive regulatory development is probably the highest-leverage policy intervention available, since a single clarifying guidance document from EPA or a state agency can unlock permitting for dozens of facilities simultaneously.

Siting in jurisdictions with more permitting experience or more favorable regulatory cultures matters more than many project developers initially appreciate. Some states — Colorado, Oregon, Washington — have made more progress on clear pathways than others. And federal facilities or projects on tribal lands sometimes offer permitting environments with different dynamics that can be more tractable.

Longer term, the cleanest solution is EPA developing a specific regulatory category for pyrolysis-based biochar production with tailored emissions standards that reflect actual risk rather than worst-case assumptions inherited from other industrial categories. That would require a formal rulemaking process, but the technology is mature enough that the emissions profile is well characterized, and the regulatory case for a dedicated pathway is reasonable. Whether EPA has the bandwidth and political will to prioritize it is another question.

 

___________________________________________________________________

Conversation #2.  3/7/26

RB: What is the best way to use AI models to scale the use of biochar production and use?

Sonnet5 4.6: This is a genuinely interesting question and one where I can offer some considered thinking — though I should be upfront that my knowledge cuts off in mid-2025, so some of the most recent developments in production tech or regulatory frameworks may have moved on. Scaling biochar production and use is a genuinely exciting area where AI can add real leverage across several parts of the value chain. Here’s how AI models can help most effectively:

Feedstock Optimization AI can analyze satellite imagery, agricultural data, and waste stream inventories to identify the best local feedstock sources — crop residues, forestry waste, manure — and predict seasonal availability. This is crucial because feedstock quality heavily determines biochar properties.

Pyrolysis Process Control Machine learning models can optimize temperature curves, residence times, and oxygen levels in real time during pyrolysis to maximize yield, energy efficiency, and specific biochar characteristics (surface area, pH, porosity) for a target application. This is one of the highest-impact use cases since process control is currently a bottleneck.

Biochar Characterization and Matching AI can predict biochar properties from feedstock and process parameters, then match biochar batches to the most suitable applications — soil amendment for a particular crop type, water filtration, carbon sequestration — without requiring expensive lab testing every time.

 

Carbon Accounting and MRV AI-assisted monitoring, reporting, and verification (MRV) can make biochar carbon credits more credible and cheaper to certify, which unlocks carbon market financing — one of the key levers for scaling production economically.

 

The Strategic Approach

The most practical way to start is usually to focus AI on the process control and soil-matching problems first, since those have the clearest ROI and the most available training data. Building a data infrastructure early — sensors on kilns, soil test databases, yield records — is essential because AI is only as good as the data fed into it.

Coupling AI tools with community-scale or distributed pyrolysis units (rather than waiting for large centralized facilities) is also likely the fastest path to meaningful scale, since it allows parallel deployment and learning across many contexts simultaneously.

 

Where the industry actually bottlenecks

Before jumping to AI applications, it’s worth being clear about what’s actually limiting scale, because that shapes where AI is most useful:

  • Feedstock variability — biomass quality, moisture content, and composition vary enormously by source, season, and geography, making consistent char quality hard to achieve
  • Characterization and testing — biochar properties (surface area, pH, CEC, PAH levels, stable carbon fraction) are expensive and slow to measure, creating friction in quality assurance and carbon credit verification
  • Application knowledge gaps — agronomic response to biochar is highly site- and crop-specific; farmers don’t have reliable guidance for their particular soil/climate/crop combination
  • Carbon credit integrity — methodologies for measuring, reporting, and verifying (MRV) stable carbon sequestration are contested and labor-intensive
  • Producer economics — most small and mid-scale producers are flying blind on process optimization, energy balance, and co-product recovery

Where AI can genuinely help

1. Real-time process optimization This is probably the highest near-term value. Pyrolysis and gasification systems have multiple interacting variables — temperature profiles, residence time, feedstock moisture, airflow — that determine both char yield and quality. ML models trained on sensor data could optimize these in real time, reducing variability and improving energy efficiency. This is especially valuable for the growing number of small distributed producers who lack process engineers.

2. Rapid biochar characterization Near-infrared spectroscopy (NIR) combined with ML models can predict biochar properties — stable carbon content, surface area, nutrient levels — from a quick scan rather than weeks of lab testing. This could dramatically lower the cost of quality verification and accelerate carbon credit issuance. Some early work exists here but it’s not yet at commercial scale.

3. Agronomic decision support A well-trained model incorporating soil type, climate, crop, and existing biochar research could give farmers actionable application rate and placement guidance. This is currently a major gap — the academic literature is enormous but fragmented, and farmers can’t access it meaningfully. RAG-based systems over the research corpus could help practitioners navigate it.

4. Carbon MRV AI-assisted remote sensing (satellite + field data) could help verify soil carbon changes post-application at lower cost than traditional sampling. This is one of the bigger unlocks for the voluntary carbon market — cheaper, more credible verification means more capital flowing to producers.

5. Feedstock sourcing and logistics Matching distributed biomass waste streams (agricultural residues, forestry slash, food processing waste) to nearby producers is a genuine optimization problem. ML could improve feedstock supply chain matching, predict seasonal availability, and reduce transport costs — which are often the margin killer for small operations.

6. Market and regulatory navigation The regulatory landscape for biochar — soil amendment registrations, carbon credit methodologies, food safety rules for agricultural applications — is fragmented across jurisdictions. AI-assisted compliance tools could lower the barrier for small producers to access markets they’re currently locked out of.

7. Demand Forecasting and Market Development NLP and market analysis models can identify emerging applications (construction materials, feed additives, electronics) and help producers anticipate demand shifts.


Where I’d be more cautious

AI for novel biochar applications (advanced materials, water filtration media, battery anodes) is interesting but those markets are still pre-commercial — the bottleneck there is more chemistry and engineering than data.

AI for direct carbon credit trading feels premature given how contested the underlying science still is — better to use AI to improve the science and MRV first, rather than to financialize uncertain measurements faster.


The honest bottom line

The highest-leverage AI applications are probably the unglamorous ones: process sensors + ML to reduce production variability, NIR spectroscopy + ML to cut characterization costs, and RAG systems to make the agronomic research actually accessible to the people who need it. These aren’t flashy, but they attack real bottlenecks that are currently limiting both quality and adoption.

What’s your angle on this — are you thinking about it from the production side, the carbon markets side, or agricultural application?

 

RB: I am really most interested in 3 things: 1) building business models that may need some capital infusion via grants or other public investments but that could eventually become profitable businesses and self-supporting; and 2)figuring out how to change the regulatory/permitting environment so that there is more understanding by public agencies about the benefits biochar production could bring in terms of waste management, local business applications of the biochar, and carbon sequestration, and meeting their GHG reduction targets; and 3) how to improve the technologies to reduce any pollution to zero.

Sonnet 4.6: These three goals are deeply connected — and that interdependence is actually a strategic asset if you frame it right. Let me think through each one and then how they reinforce each other.


1. Business Models That Can Start Grant-Funded and Become Self-Sustaining

The key insight here is that biochar producers are solving multiple problems simultaneously for multiple paying customers — and most current business models only capture one or two of those value streams. AI can help identify and structure the full stack.

The value stack a biochar operation can capture:

  • Tipping fees from waste generators (forestry slash, agricultural residue, food waste, green waste, biosolids) — this is often the most reliable near-term revenue and inverts the feedstock cost problem
  • Energy co-products — heat and syngas from pyrolysis can offset facility energy costs or be sold; this is often underutilized
  • Biochar product sales — to agriculture, landscaping, stormwater management, construction (biochar concrete is emerging)
  • Carbon credits — voluntary markets now, compliance markets increasingly
  • Avoided disposal costs — for municipalities, the alternative to biochar processing is often landfilling or open burning, both of which have real costs

The business model problem is that these revenue streams mature at different rates and require different customers and contracts. Grants and public investment make most sense to bridge the gap between “tipping fees cover operations” and “carbon credits + product sales make this genuinely profitable.”

Where AI specifically helps here:

  • Financial modeling under uncertainty — Monte Carlo simulations across feedstock availability, carbon price scenarios, biochar market prices, energy costs. Most small producers are using spreadsheets. Better modeling would help them make the case to investors and grant funders more credibly.
  • Feedstock aggregation optimization — identifying which waste streams in a given geography are available, at what volumes, at what tipping fee, and modeling the optimal facility size and location. This is genuinely a hard combinatorial problem that ML handles well.
  • Carbon credit methodology selection — there are now multiple methodologies (Puro.earth, American Carbon Registry, Gold Standard) with different requirements, costs, and credit prices. An AI tool that helps producers select the right methodology for their situation and prepare documentation would have real commercial value.
  • Grant matching and proposal support — there’s a significant amount of public money available (USDA, EPA, state climate programs, DOE) that small operators never access because the application burden is too high. AI-assisted grant identification and proposal drafting could be a meaningful service.

The path to self-sufficiency typically looks like:

Tipping fees → cover operations → grant-funded capital equipment → carbon credits come online → biochar product market develops → operation becomes fully self-sustaining over 5-7 years

The grant funding is most defensible when it’s framed as capital investment in infrastructure that serves a public function (waste diversion, GHG reduction) rather than as operating subsidy.


2. Changing the Regulatory and Permitting Environment

This is arguably the most important of your three goals because it’s the multiplier — better regulatory frameworks benefit every operator, not just one business. And it’s where AI can play a surprisingly powerful role.

The core problem: Regulators are applying frameworks designed for waste incineration or industrial chemical processing to what is often a relatively benign thermal conversion process. The result is permitting timelines of 2-5 years, costs of $500K-$2M, and requirements that make small-scale distributed production essentially impossible.

What regulators actually need — and don’t currently have:

  • Clear emissions data across feedstock types, process temperatures, and technology designs. The data exists but is scattered across academic literature, EPA test reports, and industry submissions. It hasn’t been synthesized in a form that’s useful for permit writers.
  • Comparative risk frameworks — how do biochar facility emissions compare to the open burning or landfilling they’re replacing? This comparison is almost never made explicit in permit applications but it’s the relevant policy question.
  • Case studies of successful regulatory approaches — some states and countries have created streamlined pathways. Oregon, for instance, has done more than most. British Columbia. The EU has biochar-specific frameworks. This knowledge isn’t reaching permit writers in California or the Southeast.

Where AI can help:

  • Regulatory intelligence and comparison — an AI system that continuously monitors permit decisions, regulatory guidance, and legislation across all 50 states and relevant international jurisdictions, identifies trends, and surfaces the most useful precedents for operators trying to get permitted. This doesn’t exist and would be enormously valuable.
  • Emissions database and synthesis — aggregating all available test data on biochar facility emissions into a searchable, analyzable database that can answer questions like “what are typical PM2.5 emissions from a 500 kg/hr unit running at 500°C on woody biomass?” This would give both operators and regulators better information.
  • Permit application assistance — helping operators draft permit applications that speak the language regulators use, anticipate likely concerns, and include the comparative data that makes approval easier. Much of permit rejection is about framing and completeness, not underlying feasibility.
  • Regulatory engagement support — helping operators prepare comments on proposed rules, identify the right agency contacts, and track rulemaking proceedings that affect them. Most small operators have no capacity for this.

The strategic play is to make it easier for regulators to say yes by reducing the information burden on them. A well-organized dossier that says “here is what our emissions will be, here is how that compares to the alternatives, here are three other jurisdictions that have permitted similar facilities, here is the monitoring plan” is far more likely to succeed than a generic application.

Coalition building matters too. Waste management agencies, air quality districts, and climate offices within the same government often have conflicting incentives around biochar. AI can help map those stakeholder relationships and identify the arguments most likely to resonate with each agency. A waste management director cares about diversion rates. An air quality officer cares about PM and VOCs. A climate officer cares about GHG accounting. The same facility addresses all three — but the pitch needs to be tailored.


3. Reducing Pollution to Zero

This is the technical frontier and also the prerequisite for regulatory progress — you can’t change the permitting environment if the emissions picture is unclear or genuinely problematic.

What the actual pollution concerns are:

  • Particulate matter (PM2.5, PM10) — primarily from combustion of syngas and from char handling/transport
  • Volatile organic compounds (VOCs) and polycyclic aromatic hydrocarbons (PAHs) — a real concern at lower pyrolysis temperatures or with certain feedstocks
  • Carbon monoxide — from incomplete combustion of syngas
  • NOx — from high-temperature processes, feedstock nitrogen content
  • Biochar dust during handling, application, and transport

The good news is that none of these are fundamental to the process — they’re all engineering problems. Modern well-designed units running at appropriate temperatures with proper gas treatment can achieve very low emissions. The problem is that many units in the field, especially smaller ones, are not well-designed or well-operated.

Where AI can help technically:

  • Process parameter optimization — the relationship between pyrolysis temperature, residence time, feedstock moisture, and emissions is complex and multivariate. ML models trained on sensor + emissions data can find operating envelopes that minimize pollution while maximizing char quality and energy recovery. This is probably the single most tractable technical problem.
  • Predictive emissions monitoring — using process sensors (temperature, pressure, flow rates, feedstock characteristics) to predict emissions in real time without requiring continuous stack monitoring instruments, which are expensive. This is called PEMS (Predictive Emissions Monitoring Systems) and is an established approach in other industries that hasn’t really been applied to biochar.
  • Feedstock screening — contaminated feedstocks (treated wood, plastics-contaminated biomass, certain biosolids) can cause dramatically elevated emissions. ML models that screen feedstocks using NIR or other rapid characterization methods before they enter the reactor could prevent the worst pollution events.
  • Gas cleaning optimization — afterburners, scrubbers, and biofilters all have operating parameters that affect their efficiency. ML can optimize these in real time based on the syngas composition coming off the reactor.

The zero-emission framing is actually achievable and worth pursuing explicitly, because it changes the regulatory conversation entirely. A facility that can demonstrate verified near-zero emissions with continuous monitoring is a fundamentally different permit application than one with modeled estimates and quarterly stack tests.


How These Three Goals Reinforce Each Other

This is the strategic point worth sitting with. They’re not three separate initiatives — they form a flywheel:

Better technology → cleaner emissions data → stronger regulatory case → faster permitting → more viable business models → more capital → better technology

The intervention that accelerates the whole flywheel fastest is probably a shared data infrastructure — an open or industry-shared platform that aggregates emissions data, permit decisions, feedstock characterization, agronomic outcomes, and carbon credit methodologies. Every operator who contributes data benefits from every other operator’s data. Regulators get the synthesis they need. Investors get the risk information they need. Researchers get real-world validation data.

That kind of shared infrastructure is hard for any single company to build and maintain, but it’s exactly the kind of thing that a nonprofit, trade association, or public-private partnership can fund and govern. And it’s where AI provides compounding returns — the more data, the better the models, the better the guidance, the more operators participate.

_______________________________________________________________________________

Sonnet 4.6: Is there a particular geography or scale of operation you’re focused on? That would shape which of these threads is most actionable for you right now.

RB: We believe that since surplus biomass is a local issue—in some places more than others—and it does not make financial sense in most cases to transport low value, “waste” biomass long distances, that there is likely a sweet spot for most communities, somewhere between a small-scale 2 to 8 ton per day technology and 100 tons per day of throughput. Of course, smaller technologies could scale by installing multiple units, and maybe a modular design that could be deployed as the biochar market grows would make the most sense. So, the geography would be municipal or community in scope, perhaps determined by co-locating at an existing waste processing/composting facility, would be ideal. Having tech that could take many types of feedstock would also be important.

 


Sonnet 4.6:

The Community-Scale Modular Model: Why It’s Right

Your instinct about the 2–100 ton/day range is well-grounded, and the reasoning is solid: biomass has terrible energy density relative to its weight and volume, so transport economics brutally punish centralized large-scale models. The sweet spot you’re describing — call it the distributed infrastructure model — has several compounding advantages:

  • Feedstock security: a 10-20 ton/day unit drawing from a 20-mile radius is far less vulnerable to supply disruption than a 500 ton/day plant requiring regional aggregation
  • Permitting surface area: smaller facilities trigger lower permit thresholds in most jurisdictions, and community-scale operations have more political support
  • Co-location logic: siting at an existing composting or transfer station means shared infrastructure (truck access, utilities, site security, staff), pre-existing regulatory relationships, and a natural customer for the biochar-compost blend product
  • Modular scaling: starting with one unit, proving the economics, then adding units as the biochar market develops is much lower risk than betting on a large facility from day one
  • Multi-feedstock capability: at community scale, the feedstock mix changes seasonally and by availability — green waste, wood chips, agricultural residue, food waste — so technology that handles variable inputs without major process adjustments is essential

The composting co-location point deserves particular emphasis. Biochar + compost is genuinely a better product than either alone — the biochar acts as a microbial habitat and nutrient retention matrix that improves compost quality and reduces nitrogen loss. Many composting operations are already looking for ways to differentiate their product and improve margins. This is a natural partnership, and it means the biochar producer has an anchor customer from day one.


Business Model Architecture for Community-Scale Operations

Given your three goals — viable business, regulatory change, zero pollution — here’s how the financial architecture might work in practice:

Phase 1 — Grant-funded proof of concept (Years 1-3) The strongest grant case is framed as waste diversion infrastructure with climate co-benefits. USDA Rural Development, EPA Solid Waste grants, state climate bond proceeds, and increasingly FEMA’s BRIC program (which funds hazard mitigation including wildfire fuel reduction) all have relevance. The key is positioning the facility as solving a waste problem first, with carbon sequestration as a documented co-benefit. Regulators and grant funders are more comfortable with that framing than with “carbon business.”

Phase 2 — Tipping fee revenue covers operations (Years 2-4) The tipping fee model is underappreciated. If your facility takes green waste, wood chips, and agricultural residue that would otherwise go to a landfill or be open-burned, you can charge $30-80/ton depending on jurisdiction and alternative disposal cost. At 20 tons/day, that’s potentially $200K-$500K/year in revenue before you sell a pound of biochar. This is the financial foundation that makes everything else possible.

Phase 3 — Carbon credits come online (Years 3-5) The Puro.earth marketplace currently pays $100-200/tonne of CO2-equivalent for verified biochar carbon removal. A 20 ton/day facility producing biochar at roughly 30% yield from biomass is sequestering meaningful carbon. This revenue stream is relatively new and the methodology is still maturing, but it’s real and growing.

Phase 4 — Biochar product market develops (Years 4-7) This is the slowest to develop but potentially the most durable revenue. Agricultural biochar at $200-600/ton, stormwater biochar filtration media at $400-800/ton, and the emerging biochar concrete additive market all represent exit ramps to full self-sufficiency.

The modular design insight is crucial here: a facility that starts at 5 tons/day with one unit and scales to 30 tons/day with three units as markets develop carries far less financial risk than trying to build for peak capacity upfront. Each additional unit is justified by demonstrated demand, not projected demand.


Regulatory Strategy for Community-Scale Operations

The co-location with composting facilities is also strategically useful for permitting. Composting facilities already have:

  • Air quality permits (and the monitoring infrastructure to demonstrate compliance)
  • Land use entitlements for waste processing
  • Community relationships and track records
  • Truck traffic already permitted and accepted by neighbors

A biochar unit added to that footprint has a dramatically simpler permit path than a greenfield facility. The incremental air quality burden of a well-designed pyrolysis unit running at proper temperatures is genuinely small — often less than the existing composting operation’s VOC emissions from decomposition. Making that comparison explicit in permit applications is powerful.

The regulatory education agenda you described maps well onto the community scale. A 10 ton/day unit at a municipal composting facility is something a county supervisor can visit, understand, and champion. A 500 ton/day industrial facility is something that generates NIMBYism and opposition. The human scale of the community model is a political asset.

← All Posts
← Webinar to Present FInal Results:… Bi-Weekly Biochar Review →

More recent posts

Bi-Weekly Biochar Review Initiative News
February 25, 2026
Bi-Weekly Biochar Review
Biochar Bi-Weekly Review: March 1, 2026 The biochar industry faces a pivotal question in 2026: will it finally "come of…
Read More
Webinar to Present FInal Results: Emissions Testing for Flame-Cap Kilns & Conservation Burns Vs Standard Open Burn Piles Initiative News
February 20, 2026
Webinar to Present FInal Results: Emissions Testing for Flame-Cap Kilns & Conservation Burns Vs Standard Open Burn Piles
Emissions Testing WebinarDownload
Read More
News and Links Initiative News
November 28, 2025
News and Links
As many of you know who have followed our almost 5-year pyrolysis system saga we received our Authority to Construct…
Read More