CROSS-REFERENCE TO RELATED APPLICATIONS

[001] This application claims the benefit of U.S. Provisional Application No. 63/379,436, titled “Methods for Determining Counterfactual Ecosystem Attribute Outcomes,” filed on October 13, 2022, the disclosure of which is hereby incorporated by reference in its entirety.

BACKGROUND

FIELD OF DISCLOSURE

[002] This disclosure relates generally to identifying agronomic practices that reduce ecosystem impact, and more particularly to training and implementing a model that generates counterfactual agronomic scenarios including agronomic practices that reduce an ecosystem impact relative to a baseline.

DESCRIPTION OF THE RELATED ART

[003] Agronomy models strive to support agricultural processes and enhance agronomic outcomes by incorporating scientific knowledge, technology, and a wealth of agronomic data. These models attempt to provide insights into the complex relationship between plants, soils, and the environment. While there have been advances in agronomic research, there is still room for improvement in the development of models that integrate the wealth of data from various sources (e.g., on-field and remote) and simultaneously adapt to the dynamic nature of modern agriculture.

[004] The difficulty in generating and adapting agronomy models is particularly pronounced because climate change continues to pose significant challenges in agriculture. As the landscape of ecosystem-aware agronomic practices evolves, adopting appropriate agronomic techniques to account for these advancements is hard for agriculture workers to implement and industry participants to verify and track. As such, there is a need to develop an agronomy model that allows agronomists to embrace environmentally aware agronomic practices that are appropriate for the areas and regions they manage.

SUMMARY

[005] In some aspects, the techniques described herein relate to a method including: accessing agronomic data for each of a plurality of agronomic areas; for each agronomic region of a plurality of agronomic regions: determining, for the agronomic region, a baseline agronomic scenario of a plurality of baseline agronomic scenarios, the baseline agronomic scenario including a set of agronomic practices yielding the agronomic data for a set of agronomic areas of the plurality in the agronomic region, and determining, for the agronomic region and based on the baseline agronomic scenario for the agronomic region, a baseline ecosystem outcome of a plurality of baseline ecosystem scenarios; accessing a plurality of agronomic practice modifications; and training a model using the plurality of agronomic regions, the plurality of baseline agronomic scenarios, the plurality of baseline ecosystem outcomes and the plurality of agronomic practices, the model trained to input baseline agronomic scenarios and output a counterfactual agronomic scenario including one or more agronomic practice modifications.

[006] In some aspects, the techniques described herein relate to a method, further including: receiving an agronomic area from a user; and applying the model to the agronomic area, the model generating a counterfactual agronomic scenario including an agronomic practice modification of the plurality that enhances one or more ecosystem impact in the agronomic area.

[007] In some aspects, the techniques described herein relate to a method, further including: receiving an agronomic area from a user; and applying the model to the agronomic area, the model generating a counterfactual agronomic scenario including an agronomic practice modification of the plurality that results in no change of an ecosystem attribute relative to the ecosystem outcome of the baseline agronomic scenario.

[008] In some aspects, the techniques described herein relate to a method, wherein determining the baseline agronomic scenario includes: accessing remote sensing data corresponding to the agronomic region; applying an agronomic practice identification model to the remote sensing data, the agronomic practice identification model identifying one or more agronomic practices included in the set of agronomic practices in the baseline agronomic scenario for the agronomic region; and aggregating characteristics of the one or more identified agronomic practices for the baseline agronomic scenario.

[009] In some aspects, the techniques described herein relate to a method, wherein determining the baseline agronomic scenario includes: accessing, for the agronomic region, agronomic data including in-field measurements for the set of agronomic areas in the agronomic region; applying an agronomic practice identification model to the agronomic data, the agronomic practice identification model identifying one or more agronomic practices included in the baseline agronomic scenario for the agronomic region; and aggregating characteristics of the one or more identified agronomic practices for the baseline agronomic scenario.

[0010] In some aspects, the techniques described herein relate to a method, wherein accessing agronomic data includes: accessing agronomic data representing previously implemented agronomic practices for one or more of the plurality of agronomic areas.

[0011] In some aspects, the techniques described herein relate to a method, wherein accessing agronomic data includes: accessing agronomic data representing agronomic outcomes for one or more of the plurality of agronomic areas.

[0012] In some aspects, the techniques described herein relate to a method, wherein determining the baseline ecosystem outcome for the agronomic region includes: applying an ecosystem outcome model to the set of agronomic practices in the baseline agronomic scenario for the agronomic region.

[0013] In some aspects, the techniques described herein relate to a method, wherein accessing the plurality of agronomic practice modifications includes: accessing one or more agronomy studies; and generating an agronomic practice modification based on the one or more agronomy studies.

[0014] In some aspects, the techniques described herein relate to a method, wherein accessing the plurality of agronomic practice modifications includes: accessing information from one or more agronomic experts; and generating an agronomic practice modification based on the information from the one or more agronomic experts.

[0015] In some aspects, the techniques described herein relate to a method, wherein each of the set of agronomic areas in the agronomic region share one or more characteristics.

[0016] In some aspects, the techniques described herein relate to a method, wherein the set of agronomic areas in the agronomic region represents a set of fields within a sourcing region.

[0017] In some aspects, the techniques described herein relate to a method, wherein training the model includes: simulating, for each region of the plurality, a plurality of counterfactual agronomic scenarios, each counterfactual agronomic scenario including at least one agronomic practice modifications of the plurality; and training the model based on results of the simulations.

[0018] In some aspects, the techniques described herein relate to a method including: receiving an agronomic area from a user; and applying a model to the agronomic area to output a counterfactual agronomic scenario including one or more agronomic practice modifications of a plurality of agronomic practice modifications, the model determining the counterfactual scenario by: determining an agronomic region including the agronomic area and one or more additional agronomic areas, the agronomic area and the one or more additional agronomic areas sharing at least one characteristic, determining, based on accessed agronomic data for the agronomic region, a baseline agronomic scenario for the agronomic region including a set of agronomic practices, the set of agronomic practices resulting in the agronomic data for the agronomic region; determining, for the agronomic region and based on the baseline agronomic scenario for the agronomic region, a baseline ecosystem outcome of a plurality of baseline ecosystem scenarios; accessing the plurality of agronomic practice modifications; and determining the counterfactual agronomic scenario based on the agronomic data, the agronomic region, the baseline agronomic scenario, and the plurality of agronomic practice scenarios.

[0019] In some aspects, the techniques described herein relate to a method, wherein the counterfactual agronomic scenario enhances one or more ecosystem impacts in the agronomic area.

[0020] In some aspects, the techniques described herein relate to a method, wherein the counterfactual agronomic scenario results in no change of an ecosystem attribute relative to the ecosystem outcome of the baseline agronomic scenario.

[0021] In some aspects, the techniques described herein relate to a method, wherein determining the baseline agronomic scenario includes: accessing remote sensing data corresponding to the agronomic region; applying an agronomic practice identification model to the remote sensing data, the agronomic practice identification model identifying one or more agronomic practices included in the set of agronomic practices in the baseline agronomic scenario for the agronomic region; and aggregating characteristics of the one or more identified agronomic practices for the baseline agronomic scenario.

[0022] In some aspects, the techniques described herein relate to a method, wherein determining the baseline agronomic scenario includes: accessing, for the agronomic region, agronomic data including in-field measurements for the set of agronomic areas in the agronomic region; applying an agronomic practice identification model to the agronomic data, the agronomic practice identification model identifying one or more agronomic practices included in the baseline agronomic scenario for the agronomic region; and aggregating characteristics of the one or more identified agronomic practices for the baseline agronomic scenario. [0023] In some aspects, the techniques described herein relate to a method, wherein determining the baseline ecosystem outcome for the agronomic region includes: applying an ecosystem outcome model to the set of agronomic practices in the baseline agronomic scenario for the agronomic region.

[0024] In some aspects, the techniques described herein relate to a non-transitory computer-readable storage medium store computer program instructions that, when executed by one or more processors, cause the one or more processors to: receive an agronomic area from a user; and apply a model to the agronomic area to output a counterfactual agronomic scenario including one or more agronomic practice modifications of a plurality of agronomic practice modifications, the model determining the counterfactual scenario by causing the one or more processors to: determine a region agronomic region including the agronomic area and one or more additional agronomic areas, the agronomic area and the one or more additional agronomic areas sharing at least one characteristic, determine, based on accessed agronomic data for the agronomic region, a baseline agronomic scenario for the region including a set of agronomic practices, the set of agronomic practices resulting in the agronomic data for the agronomic region; determine, for the agronomic region and based on the baseline agronomic scenario for the agronomic region, a baseline ecosystem outcome of a plurality of baseline ecosystem scenarios; access the plurality of agronomic practice modifications; and determine the counterfactual agronomic scenario based on the agronomic data, the agronomic region, the baseline agronomic scenario, and the plurality of agronomic practice scenarios.

[0025] The methods described above may be implemented by the system(s) described herein. The methods described above may be implemented by a processor executing computer program instructions stored on a non-transitory computer readable storage medium.

BRIEF DESCRIPTION OF DRAWINGS

[0026] FIG. 1 illustrates a counterfactual engine, according to an example embodiment.

[0027] FIG. 2 illustrates a first example workflow for generating counterfactual scenarios, according to an example embodiment.

[0028] FIG. 3 illustrates a second example workflow for generating counterfactual scenarios, according to an example embodiment.

[0029] FIG. 4 illustrates a third example workflow for generating counterfactual scenarios, according to an example embodiment. [0030] FIG. 5 illustrates a fourth example workflow for generating counterfactual scenarios, according to an example embodiment.

[0031] FIG. 6 illustrates regions where a counterfactual scenario including no-till was generated, according to an example embodiment.

[0032] FIG. 7 illustrates regions where a counterfactual scenario including cover crop and no-till were generated, according to an example embodiment.

[0033] FIG. 8 illustrates regions where a counterfactual scenario including new cash crops in rotation and no-till were generated, according to an example embodiment.

[0034] FIG. 9 illustrates regions where a counterfactual scenario including cover crops were generated, according to an example embodiment.

[0035] FIG. 10 illustrates regions where a counterfactual scenario including tillage impact reduction and tillage pass reduction were generated, according to an example embodiment.

[0036] FIG. 11 illustrates regions where a counterfactual scenario including fertilizer reduction and no-till were generated, according to an example embodiment.

[0037] FIG. 12 illustrates regions where a counterfactual scenario including tillage impact reduction were generated, according to an example embodiment.

[0038] FIG. 13 illustrates regions where a counterfactual scenario including fertilizer reduction was generated, according to an example embodiment.

[0039] FIG. 14 illustrates regions where a counterfactual scenario including new cash crops in rotation was generated, according to an example embodiment.

[0040] FIG. 15 shows counties in which fertilizer application was delayed, fertilizer was reduced, and no-till was generated, according to an example embodiment.

[0041] FIG. 16 shows the outlines of the 9 ASDs in Iowa, according to an example embodiment.

[0042] FIG. 17 shows counties flagged for reduced tillage as a common practice, according to an example embodiment.

[0043] FIG. 18 shows counties flagged for no-till as a common practice, according to an example embodiment.

[0044] FIG. 19 illustrates an example workflow for training a counterfactual model to generate counterfactual scenarios, according to an example embodiment.

[0045] FIG. 20 is a block diagram illustrating components of an example machine for reading and executing instructions from a machine-readable medium. [0046] The figures depict various embodiments for purposes of illustration only. One skilled in the art will readily recognize from the following discussion that alternative embodiments of the structures and methods illustrated herein may be employed without departing from the principles described herein.

DETAILED DESCRIPTION

I. INTRODUCTION

Evolution of Agronomic Management

[0047] Over the past several decades, the evolution of agronomic management has steadily shifted towards practices that are aware of their ecosystem impact (i.e., accounting for how agronomic practices during a cultivation cycle affect, e.g., emissions). Initially, traditional agricultural approaches prioritized factors such as, e.g., crop yield, pest control, and soil fertility; however, increasing awareness surrounding climate change and the role of emissions in exacerbating the issue has led to a new era of agricultural practices that aim to reduce environmental impact in agronomy. These modem methods are aimed, not only at enhancing productivity, but also reducing the sector's environmental impact and promoting long-term environmental sustainability.

[0048] The advances in environment-aware agronomic management have manifested in various forms. An example includes the adoption of conservation tillage practices, which minimize soil disturbance and carbon release associated with traditional powered tillage. Another development is implementing localized agriculture practices (e.g., on a field or subfield basis), using advanced technologies to optimize resource application and minimize waste, thereby reducing both input costs and greenhouse gas emissions. Additionally, cover crop integration into a typical cultivation cycle is proving effective at improving soil health, preserving carbon sequestration in soils, and mitigating emissions.

[0049] Despite the significant strides made in environment-aware agronomic management, individual farmers often face challenges in keeping up with the latest information, technologies, and practices. These challenges stem from factors such as a limited access to modern agronomic resources and services (e.g., academic research), a lack of comprehensive understanding of complex carbon management principles, a lack of effective digital scenario planning tools for estimating the impacts of environment-aware agronomic management across diverse climates and soil characteristics, and an insufficient technical background to identify and implement appropriate agronomic practices that would result in improved environment-aware agronomy given the wealth of information available in today’s world. Consequently, many farmers find it difficult to adopt the most up-to-date and effective environment-aware practices on their farms, which hinders widespread progress in the sector.

[0050] To address these issues, it is crucial to encourage the adoption of modem environment-aware agronomic management techniques; however, as described above, oftentimes farmers are unable to sift through the wealth of information describing modem agronomic practices to determine what is the most appropriate practice for their farm. For example, a farmer growing corn crops in Kansas may not know whether agronomic practices tailored for growing corn with a low carbon impact in Iowa, or practices tailored for growing soybeans with a low carbon impact in Kansas, would be more beneficial to implement. The farmer, in this example, is unable to determine whether the similar crop profile, or the similar geographic region, is more important in deciding which environment-aware practices to adopt for his Kansas com crop.

[0051] To that end, modem machine learning models are uniquely suited for identifying associations and indications (“indicators”) between disparate agronomic practices and their corresponding agronomic results in a way that is impossible for modern farmers to achieve at scale. Using the indicators, machine learning models can be trained to identify agronomic practices that would reduce the environmental impact in agronomic regions.

Agronomic Models

[0052] Traditionally, generating a set of “best agronomic practices” for a particular desired agronomic outcome (e.g., high yield corn) in a particular field or region (e.g., Iowa) was reserved for an agronomic manager for the field or region. The manager would adopt and implement a “best practice” for an area or region based on their personal experience, academic knowledge, history and tradition in the area or region, etc.

[0053] Within the management role, the manager may occasionally attempt to modify the best practice to improve or modify the desired outcome (e.g., lower carbon impact).

However, implementing changes is risky if the modification is outside of the manager’s knowledge set or capabilities (e.g., a new cover crop that the manager had never used before, adopting practices used in a distant region, etc.). Moreover, a manager may not have the technical skills necessary to access agronomic data from a wide array of sources, identify the pertinent indicators from the data, and determine whether particular practice modifications would correlate with their particular field or region. As such, adoption and adaptation of modern environment-aware agronomic practices has been slow, despite the wealth of agronomic data now available.

[0054] Indeed, the process of modifying an agronomic best practice would prove daunting to any manager, or any person attempting to modify agronomic best practices (e.g., a user). For example, consider a manager growing a corn crop in Iowa every cultivation cycle. That manager may implement, e.g., twelve specific agronomic practices throughout the cultivation cycle because it is the historic best practice. Now, consider that the manager wants to reduce his ecosystem impact by implementing new agronomic practices that would reduce the impact. Of course, the manager could find a few nearby farmers and ask them for advice, read up on academic studies, or do independent research on agronomic results from the surrounding farms.

[0055] However, tackling this problem on a small scale may prove fatal for the manager’s desired goal (or any person seeking to do the same). For instance, the nearby farmer may manage fields with a different soil composition so his advice may not be pertinent, or the academic research the manager leverages may be based on a more arid climate pattern and prove a contraindication, or the data from nearby farms may be in a format that he misinterprets and is largely unusable. To that end, the manager (or the person) would be better served by pulling in larger amounts of data to identify practices for appropriately modifying his best practices for environment awareness given his particular agronomic situation.

[0056] Unfortunately, a manager (or the person) cannot feasibly process the copious amount of agronomic data that may lead to better agronomic practices with his mind. This makes sense contextually. For instance, while a manager may be able to look at a handful of agronomic areas surrounding his farm to identify agronomic practices that would lead to better outcomes, he would be unable to look at, e.g., 1,000 or 10,000 fields from across the country, academic journals from across the world for the past decade, seek the advice of 100 different experts, etc. to identify new environment-aware best practices.

[0057] Fortunately, machine-learned models can perform this task, which a manager (or the person) could not feasibly accomplish with their mind. That is, a machine learned model can be trained using the vast amounts of data that would be available to a manager and which they would be unable to understand (or infer appropriate information from) at scale. The machine learned model, as described hereinbelow, may be trained to identify correlations between thousands, hundreds of thousands, and millions of field measurements, remote sensing measurements, expert recommendations, academic articles, etc. Thus, when applied by a manager, the machine learned model is performing a task that the manger would be unable to feasibly perform themselves.

Agronomy Models Improve Agronomic Management

[0058] Managers may leverage agronomy models in ways that improve the field of agronomy and agronomy management. To expand, as set forth above, improving agronomic management has traditionally relied on the practice and best judgement of a single manager iterating agronomy practices over single cultivation cycles (because the manager is unable to leverage the vast amounts of agronomic data available in today’s world). Given this reliance on a single incapable manager, any desired shifts towards environment-aware agronomic practices are slowly implemented, incrementally generated, and tailored incorrectly.

However, training and applying a machine-learned model to identify environment-aware agronomic practices using a depth of agronomic data solves many of these problems. For instance, the agronomic model may provide new environment-aware agronomic practices to a manager, or any other person interested in improved agronomic outcomes, that gives them a higher degree of certainty in quick, abrupt adoption of environment-aware agronomic practices given the amount and type of data on which the model is trained.

II. DEFINITIONS

[0059] As used herein “ecosystem attribute”, “ecosystem outcome”, or “environmental attribute”, are equivalent. Each term refers to an environmental characteristic (for example, as a result of agricultural production) that may be quantified and valued (for example, as an ecosystem credit or sustainability claim). An “ecosystem benefit” is an ecosystem outcome that has a beneficial effect on an environmental characteristic (for example, and increase in soil carbon sequestration, a decrease in water use, a decrease in fertilizer run-off, etc.) Examples of ecosystem attributes may include, without limitation: an amount of water used; an amount, type, and or timing of nitrogen use; a quantity and permanence of soil carbon sequestration; an amount of greenhouse gas emission emitted or avoided, etc. An example of a mandatory program requiring accounting of ecosystem attributes is California’s Low Carbon Fuel Standard (LCFS). Field-based agricultural management practices can be a means for reducing the carbon intensity of biofuels (e.g., biodiesel from soybeans).

[0060] An “ecosystem impact” is a change in an ecosystem attribute relative to a baseline. In various embodiments, baselines may reflect a set of regional standard practices or production (a comparative baseline), prior production practices and outcomes for a field or fanning operation (a temporal baseline), or a counterfactual alternative scenario (a counterfactual baseline). For example, a temporal baseline for determination of an ecosystem impact may be the difference between a safrinha crop production period and the safrinha crop production period of the prior year. In some embodiments, an ecosystem impact can be generated from the difference between an ecosystem attribute for the latest crop production period and a baseline ecosystem attribute averaged over a number (e.g., 2, 3, 4, 5, 6, 7, 8, 9, 10) of prior production periods.

[0061] A “counterfactual scenario” refers to, generally, what could have happened within the crop growing season in an area of land given alternative practices. For example, planting a crop on a second date rather than a first, implemented date, or spraying a second, counterfactual herbicide rather than a first, implemented herbicide, etc. In various embodiments, a counterfactual scenario is based on an approximation of sourcing region practices (for example, the most prevalent practices within a sourcing region), the most prevalent practices within a sourcing region for a field of a given characteristic (e.g. size of field, crop type, soil type, irrigation type, etc.), the most prevalent practices withing a geographic area, the most prevalent practices for fields having similar characteristics, etc.

[0062] An “ecosystem credit” is a unit of value corresponding to an ecosystem benefit or ecosystem impact, where the ecosystem attribute or ecosystem impact is measured, verified, and or registered according to a methodology. In some embodiments, an ecosystem credit may be a report of the inventory of ecosystem attributes (for example, an inventory of ecosystem attributes of a management zone, an inventory of ecosystem attributes of a field, an inventory of ecosystem attributes of a farming operation, an inventory of ecosystem attributes of a sourcing region, an inventory of ecosystem attributes of a supply chain, an inventory of a processed agricultural product, a life cycle inventory (LCI), efc.). In some embodiments, an ecosystem credit is a life-cycle assessment (LCA).

[0063] In some embodiments, an ecosystem credit may be a registry-issued credit. Optionally, an ecosystem credit is generated according to a methodology approved by an issuer. An ecosystem credit may represent a reduction or offset of an ecologically significant resource or input (e.g., carbon credits, water credits, nitrogen credits). In some embodiments, a reduction or offset is compared to a baseline of ‘business as usual’ if the ecosystem crediting or sustainability program did not exist (e.g., if one or more practice change made because of the program had not been made). [0064] In some embodiments, a reduction or offset is compared to a baseline of one or more ecosystem attributes (e.g., ecosystem attributes of one or more: field, sub-field region, county, state, region of similar environment, sourcing region, etc.) during one or more prior production period. For example, ecosystem attributes of a field in 2022 may be compared to a baseline of ecosystem attributes of the field in 2021. In some embodiments, a reduction or offset is compared to a baseline of one or more ecosystem attributes (e.g., ecosystem attributes of one or more: field, sub-field region, county, state, region of similar environment, sourcing region, etc.) during the same production period. For example, ecosystem attributes of a field may be compared to a baseline of ecosystem attributes of a sourcing region comprising the field. An ecosystem credit may represent a permit to reverse an ecosystem benefit, for example, a license to emit one metric ton of carbon dioxide. A carbon credit represents a measure (e.g., one metric ton) of carbon dioxide or other greenhouse gas emissions reduced, avoided or removed from the atmosphere. A nutrient credit, for example a water quality credit, represents pounds of a chemical removed from an environment (e.g., by installing or restoring nutrient-removal wetlands) or reduced emissions (e.g., by reducing rates of application of chemical fertilizers, managing the timing or method of chemical fertilizer application, changing type of fertilizer, etc.). Examples of nutrient credits include nitrogen credits and phosphorous credits. A water credit represents a volume (e.g., 1000 gallons) of water usage that is reduced or avoided, for example by reducing irrigation rates, managing the timing or method of irrigation, employing water conservation measures, such as reducing evaporation application.

[0065] A “sustainability claim” is a set of one or more ecosystem benefits associated with an agricultural product (for example, including ecosystem benefits associated with production of an agricultural product). Sustainability claims may or may not be associated with ecosystem credits. For example, a consumer package good entity may contract raw agricultural products from producers reducing irrigation, in order to make a sustainability claim of supporting the reduction of water demand on the final processed agricultural product. The producers reducing irrigation may or may not also participate in a water ecosystem credit program, where ecosystem credits are generated based on the quantity of water that is actually reduced compared against a baseline.

[0066] Offsets” are credits generated by third parties outside the value chain of the party with the underlying liability (e.g., oil company that generates greenhouse gases from combusting hydrocarbons purchases carbon credit from a farmer). [0067] “Insets” are ecosystem resource (e.g., carbon dioxide) reductions within the value chain of the party with the underlying liability (e.g., oil company who makes biodiesel reduces carbon intensity of biodiesel by encouraging farmers to produce the underlying soybean feedstock using sustainable farming practices). Insets are considered Scope 1 reductions.

[0068] An ecosystem credit may generally be categorized as either an inset (when associated with the value chain of production of a particular agricultural product), or an offset, but not both concurrently.

[0069] Emissions of greenhouse gases are often categorized as “Scope 1 Emissions,” “Scope 2 Emissions,” or “Scope 3 Emissions.” Scope 1 Emissions are direct greenhouse gas emissions that occur from sources that are controlled or owned by an organization. Scope 2 Emissions are indirect greenhouse gas emissions associated with purchase of electricity, steam, heating, or cooling. Scope 3 Emissions are the result of activities from assets not owned or controlled by the reporting organization, but that the organization indirectly impacts in its value chain. Scope 3 Emissions represent all emissions associated with an organization’s value chain that are not included in that organization’s Scope 1 or Scope 2 emissions. Scope 3 emissions include activities upstream of the reporting organization or downstream of the reporting organization. Upstream activities include, for example, purchased goods and services (e.g., agricultural production, such as wheat, soybeans, or corn, may be purchased inputs for production of animal feed), upstream capital goods, upstream fuel and energy, upstream transportation and distribution (e.g., transportation of raw agricultural products, such as grain from the field to a grain elevator), waste generated in upstream operations, business travel, employee commuting, or leased assets. Downstream activities include, for example, transportation and distribution other than with the vehicles of the reporting organization, processing of sold goods, use of goods sold, end-of-life treatment of goods sold, leased assets, franchises, or investments.

[0070] As used herein, a “crop growing season” may refer to a unit of grouping crop events by overlapping (for example, in the case of companion cropping) or non-overlapping periods of time. In various embodiments, harvest events are used where possible.

[0071] An “issuer” is an issuer of ecosystem credits, which may be a regulatory authority or another trusted provider of ecosystem credits. An issuer may alternatively be referred to as a “registry”. [0072] A “token” (alternatively, an “ecosystem credit token”) is a digital representation of an ecosystem benefit, ecosystem impact, or ecosystem credit. The token may include a unique identifier representing one or more ecosystem credits, ecosystem attribute, or ecosystem impact, or, in some embodiments a putative ecosystem credit, putative ecosystem attribute, or putative ecosystem impact, associated with a particular product, production location (e.g., a field), production period (e.g., cultivation cycle), and/or production zone cycle (e.g., a single management zone defined by events that occur over the duration of a single crop production season).

[0073] “Ecosystem credit metadata” are, at least, information sufficient to identify an ecosystem credit issued by an issuer of ecosystem credits. For example, the metadata may include one or more of a unique identifier of the credit, an issuer identifier, a date of issuance, identification of the algorithm used to issue the credit, or information regarding the processes or products giving rise to the credit. In some embodiments, the credit metadata may include a product identifier as defined herein. In other embodiments, the credit is not tied to a product at generation, and so there is no product identifier included in the credit metadata.

[0074] A “product” is any item of agricultural production, including crops and other agricultural products, in their raw, as-produced state (e.g., wheat grains), or as processed (e.g., oils, flours, polymers, consumer goods (e.g., crackers, cakes, plant based meats, animalbased meats (for example, beef from cattle fed a product, such as corn grown from a particular field), bioplastic containers, efc.). In addition to harvested physical products, a product may also include a benefit or service provided via use of the associated land (for example, for recreational purposes, such as a golf course), pastureland for grazing wild or domesticated animals (where domesticated animals may be raised for food or recreation).

[0075] “Product metadata” are any information regarding an underlying product, its production, and/or its transaction that may be verified by a third party and may form the basis for issuance of an ecosystem credit and/or sustainability claim. Product metadata may include, at least, a product identifier, as well as a record of entities involved in transactions. [0076] As used herein, “quality” or a “quality metric” may refer to any aspect of an agricultural product that adds value. In some embodiments, quality is a physical or chemical attribute of the crop product. For example, a quality may include, for a crop product type, one or more of: a variety; a genetic trait or lack thereof; genetic modification or lack thereof; genomic edit or lack thereof; epigenetic signature or lack thereof; moisture content; protein content; carbohydrate content; ash content; fiber content; fiber quality; fat content; oil content; color; whiteness; weight; transparency; hardness; percent chalky grains; proportion of corneous endosperm; presence of foreign matter; number or percentage of broken kernels; number or percentage of kernels with stress cracks; falling number; farinograph; adsorption of water; milling degree; immature grains; kernel size distribution; average grain length; average grain breadth; kernel volume; density; L/B ratio; wet gluten; sodium dodecyl sulfate sedimentation; toxin levels (for example, mycotoxin levels, including vomitoxin, fumonisin, ochratoxin, or aflatoxin levels); and damage levels (for example, mold, insect, heat, cold, frost, or other material damage).

[0077] In some embodiments, quality is an attribute of a production method or environment. For example, quality may include, for a crop product, one or more of: soil type; soil chemistry; climate; weather; magnitude or frequency of weather events; soil or air temperature; soil or air moisture; degree days; rain fed; irrigated or not; type of irrigation; tillage frequency; cover crop (current or historical); fallow seasons (present or historical); crop rotation; organic; shade-grown; greenhouse; level and types of fertilizer use; levels and type of chemical use; levels and types of herbicide use; pesticide-free; levels and types of pesticide use; no-till; use of organic fertilizers; manure and byproducts; minority-produced; fair- wage; geography of production (e.g., country of origin, American Viti cultural Area, mountain-grown); pollution-free production; reduced-pollution production; levels and types of greenhouse gas production; carbon-neutral production; levels and duration of soil carbon sequestration; and others. In some embodiments, quality is affected by, or may be inferred from, the timing of one or more production practices. For example, food-grade quality for crop products may be inferred from the variety of plant, damage levels, and one or more production practices used to grow the crop. In another example, one or more qualities may be inferred from the maturity or growth stage of an agricultural product, such as a plant or animal. In some embodiments, a crop product is an agricultural product.

[0078] In some embodiments, quality is an attribute of a method of storing an agricultural good (e.g., the type of storage: bin, bag, pile, in-field, box, tank, or other containerization), the environmental conditions (e.g., temperature, light, moisture, relative humidity, presence of pests, CO2 levels) during storage of the crop product, method of preserving the crop product (e.g., freezing, drying, chemically treating), or a function of the length of time of storage. In some embodiments, quality may be calculated, derived, inferred, or subjectively classified based on one or more measured or observed physical or chemical attributes of a crop product, its production, or its storage method. In some embodiments, a quality metric is a grading or certification by an organization or agency. For example, grading by the USDA, organic certification, or non-GMO certification may be associated with a crop product. In some embodiments, a quality metric is inferred from one or more measurements made of plants during growing season. For example, wheat grain protein content may be inferred from measurement of crop canopies using hyperspectral sensors and/or near-infrared (NIR) or visible spectroscopy of whole wheat grains. In some embodiments, one or more quality metrics are collected, measured, or observed during harvest. For example, dry matter content of corn may be measured using NIR spectroscopy on a combine. In some embodiments, the observed or measured value of a quality metric is compared to a reference value for the metric. In some embodiments, a reference value for a metric (for example, a quality metric or a quantity metric) is an industry standard or grade value for a quality metric of a particular agricultural good (for example, U.S. No. 3 Yellow Corn, Flint), optionally as measured in a particular tissue (for example, grain) and optionally at a particular stage of development (for example, silking). In some embodiments, a reference value is determined based on a supplier’s historical production record or the historical production record of present and/or prior marketplace participants.

[0079] A “field” is the area where agricultural production practices are being used (for example, to produce a transacted agricultural product) and/or ecosystem credits and/or sustainability claims.

[0080] As used herein, a “field boundary” may refer to a geospatial boundary of an individual field. Various methods may be used determine a field boundary. Exemplary methods are provided in U.S. Pub. No. 2022/0180526, which is hereby incorporate by reference in its entirety. Another example of methods for field boundary determination is a graphical user interface (“GUI”) comprises a map of the geographic region comprising one or more field boundary, wherein the interface is configured to return a shape corresponding to a shape drawn on a display of a user device.

[0081] As used herein, an “enrolled field boundary” may refer to the geospatial boundary of an individual field enrolled in at least one ecosystem credit or sustainability claim program on a specific date.

[0082] As used herein, a “management event” may refer to a grouping of data about one or more farming practices (such as tillage, harvest, efc.) that occur within a field boundary or an enrolled field boundary. A “management event” contains information about the time when the event occurred, and has a geospatial boundary defining where within the field boundary the agronomic data about the event applies. Management events are used for modeling and credit quantification, designed to facilitate grower data entry and assessment of data requirements. Each management event may have a defined management event boundary that can be all, or part of, the field area defined by the field boundary. A “management event boundary” (equivalently a “farming practice boundary”) is the geospatial boundary of an area over which farming practice action is taken or avoided. In some embodiments, if a farming practice action is an action taken or avoided at a single point, the management event boundary is point location. As used herein, a farming practice and agronomic practice are of equivalent meaning.

[0083] As used herein, a “management zone” may refer to an area within an individual field boundary defined by the combination of management event boundaries that describe the presence or absence of management events at any particular time or time window, as well as attributes of the management events (if any event occurred). A management zone may be a contiguous region or a non-contiguous region. A “management zone boundary” may refer to a geospatial boundary of a management zone. In some embodiments, a management zone is an area coextensive with a spatially and temporally unique set of one or more farming practices. In some embodiments, an initial management zone includes historic management events from one or more prior cultivation cycles (for example, at least 2, at least 3, at least 4, at least 5, or a number of prior cultivation cycles required by a methodology). In some embodiments, a management zone generated for the year following the year for which an initial management zone was created will be a combination of the initial management zone and one or more management event boundaries of the next year. A management zone or set of management zones can be a data-rich geospatial object created for each field using an algorithm that crawls through management events (e.g., all management events) and groups the management events into discrete zonal areas based on features associated with the management event(s) and/or features associated with the portion of the field in which the management event(s) occur. The creation of management zones enables the prorating of credit quantification for the area within the field boundary based on the geospatial boundaries of management events.

[0084] In some embodiments, a management zone is created by sequentially intersecting a geospatial boundary defining a region wherein management zones are being determined (for example, a field boundary), with each geospatially management event boundary occurring within that region at any particular time or time window, wherein each of the sequential intersection operations creates two branches - one with the intersection of the geometries and one with the difference. The new branches are then processed with the next management event boundary in the sequence, bifurcating whenever there is an area of intersection and an area of difference. This process is repeated for all management event boundaries that occurred in the geospatial boundary defining the region. The final set of leaf nodes in this branching process define the geospatial extent of the set of management zones within the region, wherein each management zone is non-overlapping, and each individual management zone contains a unique set of management events relative to any other management zone defined by this process.

[0085] As used herein, a “zone-cycle” may refer to a single cultivation cycle on a single management zone within a single field, considered collectively as a pair that define a foundational unit (e.g., also referred to as an “atomic unit”) of quantification for a given field in a given reporting period.

[0086] As used herein, a “field-level project start date” may refer to the start of the earliest cultivation cycle, for example, the earliest cultivation cycle where a practice change was detected and attested by a grower, or the earliest cultivation cycle where an agricultural practice is declared by a grower.

[0087] As used herein, a “historic baseline period” may refer to a period for which historic information (for example, land use history, management practice history, crops planted, yields, soil samples, etc.) is required. In various embodiments, historic data may be collected on the field directly or inferred from remote sensing data. A number of required years is specified by a methodology, based on crop rotation and management.

[0088] As used herein, a “cultivation cycle” (equivalently a “crop production period” or “production period”) may refer to the period between the first day after harvest or cutting of a prior crop (for example, cash crop) on a field or the first day after the last grazing on a field, and the last day of harvest or cutting of the subsequent crop on a field or the last day of last grazing on a field. For example, a cultivation cycle may be: a period starting with the planting date of current crop and ending with the harvest of the current crop, a period starting with the date of last field prep event in the previous year and ending with the harvest of the current crop, a period starting with the last day of crop growth in the previous year and ending with the harvest or mowing of the current crop, a period starting the first day after the harvest in the prior year and the last day of harvest of the current crop, etc. In some embodiments, cultivation cycles are approximately 365-day periods from the field-level project start date that contain completed crop growing seasons (planting to harvest/mowing, or growth start to growth stop). In some embodiments, cultivation cycles extend beyond a single 365-day period and cultivation cycles are divided into one or more cultivation cycles of approximately 365 days, optionally where each division of time includes one planting event and one harvest or mowing event. The length of the cultivation cycle may vary from year to year, depending on weather and the overall crop and management rotation schedule. A cultivation cycle may be greater or less than a calendar year, and may include multiple crop growing seasons, including cash crops growing seasons, cover crop growing seasons, perennial crop growing seasons, and fallow period crop growing seasons.

[0089] As used herein, “historic cultivation cycles” may be defined in the same way as cultivation cycles, but for the period of time in the required historic baseline period.

[0090] As used herein, a “historic segment” may refer to individual historic cultivation cycles, separated from each other in order to use to construct baseline simulations.

[0091] As used herein, “historic crop practices” may refer to crop events occurring within historic cultivation cycles.

[0092] An “indication of a geographic region” is a latitude and longitude, an address or parcel id, a geopolitical region (for example, a city, county, state), a region of similar environment (e.g., a similar soil type or similar weather), a sourcing region, a boundary file, a shape drawn on a map presented within a GUI of a user device, image of a region, an image of a region displayed on a map presented within a GUI of a user device, a user id where the user id is associated with one or more production locations (for example, one or more fields). [0093] For example, polygons representing fields may be detected from remote sensing data using computer vision methods (for example, edge detection, image segmentation, and combinations thereof) or machine learning algorithms (for example, maximum likelihood classification, random tree classification, support vector machine classification, ensemble learning algorithms, convolutional neural network, etc.).

[0094] “Ecosystem observation data” are observed or measured data describing an ecosystem, for example weather data, soil data, remote sensing data, emissions data (for example, emissions data measured by an eddy covariance flux tower), populations of organisms, plant tissue data, and genetic data. In some embodiments, ecosystem observation data are used to connect agricultural activities with ecosystem variables. Ecosystem observation data may include survey data, such as soil survey data (e.g., SSURGO). In various embodiments, the system performs scenario exploration and model forecasting, using the modeling described herein. In various embodiments, the system proposes climate-smart crop fuel feedstock CI integration with an existing model, such as the Greenhouse gases, Regulated Emissions, and Energy use in Technologies Model (GREET), which can be found online at https://greet.es.anl.gov/ (the GREET models are incorporated by reference herein). [0095] A “crop type data layer” is a data layer containing a prediction of crop type, for example USDA Cropland Data Layer provides annual predictions of crop type, and a 30m resolution land cover map is available from MapBiomas (https://mapbiomas.org/en). A crop mask may also be built from satellite-based crop type determination methods, ground observations including survey data or data collected by farm equipment, or combinations of two or more of: an agency or commercially reported crop data layer (e.g., CDL), ground observations, and satellite-based crop type determination methods.

[0096] A “vegetative index” (“VI”) is a value related to vegetation as computed from one or more spectral bands or channels of remote sensing data. Examples include simple ratio vegetation index (“RVI”), perpendicular vegetation index (“PVI”), soil adjusted vegetation index (“SAVI”), atmospherically resistant vegetation index (“AR VI”), soil adjusted atmospherically resistant VI (“SARVI”), difference vegetation index (“DVI”), normalized difference vegetation index (“ND VI”). ND VI is a measure of vegetation greenness that is particularly sensitive to minor increases in surface cover associated with cover crops.

[0097] SEP” stands for Soil Enrichment Protocol. The SEP version 1.0 and supporting documents, including requirements and guidance, (incorporated by reference herein) can be found online at https://www.climateactionreserve.org/how/protocols/soil-enri chment/. As is known in the art, SEP is an example of a carbon registry methodology, but it will be appreciated that other registries having other registry methodologies (e.g., carbon, water usage, efc.) may be used, such as the Verified Carbon Standard VM0042 Methodology for Improved Agricultural Land Management, vl.O (incorporated by reference herein), which can be found online at https://verra.org/methodology/vm0042-methodology-for-improve d- agri cultural -land-management-v 1-0/. The Verified Carbon Standard methodology quantifies the greenhouse gas (GHG) emission reductions and soil organic carbon (SOC) removals resulting from the adoption of improved agricultural land management (ALM) practices.

Such practices include, but are not limited to, reductions in fertilizer application and tillage, and improvements in water management, residue management, cash crop and cover crop planting and harvest, and grazing practices. [0098] “LRR” refers to a Land Resource Region, which is a geographical area made up of an aggregation of Major Land Resource Areas (MLRA) with similar characteristics.

[0099] DayCent is a daily time series biogeochemical model that simulates fluxes of carbon and nitrogen between the atmosphere, vegetation, and soil. It is a daily version of the CENTURY biogeochemical model. Model inputs include daily maximum/minimum air temperature and precipitation, surface soil texture class, and land cover/use data. Model outputs include daily fluxes of various N-gas species (e.g., N2O, NOx, N2); daily CO2 flux from heterotrophic soil respiration; soil organic C and N; net primary productivity; daily water and nitrate (NO3) leaching, and other ecosystem parameters. In some embodiments, model outputs may have uncertainty estimates attached to them, for example, as determined by Monte Carlo simulations.

[00100] As used herein, “sourcing region” is alternately referred to as a “supply shed”, a region in which an agricultural product is produced for a particular product or to supply a particular collection facility (for example, processing facility, a storage facility, a transportation facility, etc.).

III. COUNTERF ACTUAL ENGINE

[00101] One system an agronomist or other industry participant may employ to improve or estimate their agronomic outcomes, ecosystem outcomes, and/or ecosystem impact is a counterfactual engine 100. A counterfactual engine 100 identifies “counterfactual” agronomic practices that would produce an ecosystem impact for agronomic areas and/or regions (e.g. regions overseen by a manager, an area of interest to a user of the counterfactual engine 100, a sourcing region, etc.). To do so, the counterfactual engine 100 identifies a “baseline scenario” for an agronomic region. The baseline scenario is a set of agronomic practices that represent agronomic practices typically implemented in the region (e.g., management events, agronomic products, quality, etc.). The counterfactual engine 100 determines a baseline ecosystem outcome for the baseline scenario given the typical set of agronomic practices. The counterfactual engine 100 identifies one or more counterfactual agronomic practices that would enhance (or modify) one or more ecosystem outcome if implemented. The resulting counterfactual agronomic practices, and any of the unchanged typical agronomic practices, may be used to generate a counterfactual agronomic scenario.

[00102] To provide a contextual example, consider again a manager growing corn crops. The manager is growing her corn crops in several agronomic areas (e.g., fields) that are within a greater agronomic region (e.g., her county in Iowa). The manager uses the counterfactual engine 100 to generate a “counterfactual scenario" that, if implemented, would enhance (or modify) one or more ecosystem outcome associated with her crops. To do so, the counterfactual engine 100 generates a baseline scenario for the manager’s agronomic region (including her agronomic areas). In this example, the baseline scenario includes agriculture practices (e.g., management events) such as, e.g., clearing the field on a first day, tilling the field on a second day, planting the field on a third day, etc. The baseline scenario may also include other characteristics associated with the region such as temperature, rainfall, soil condition, etc. Most generally, the baseline scenario represents typical agronomic practices for growing corn crops in the region during a cultivation cycle.

[00103] The counterfactual engine 100 generates a baseline ecosystem outcome for the region. The baseline ecosystem outcome is the expected ecosystem impact given the baseline scenario. In other words, the counterfactual engine 100 determines, using historic agronomic data, an ecosystem impact projection for the region given the agronomic practices in the baseline scenario. For example, clearing the field on the first day may create a first amount of expected emissions, tilling the field on the second day may generate a second amount of expected emissions, planting the field on the third day may have a third amount of expected sequestration, etc. Given these associations, the counterfactual engine 100 generates a baseline ecosystem impact for the baseline scenario.

[00104] The counterfactual engine 100 generates a counterfactual scenario for the agronomic region. The counterfactual scenario includes one or more counterfactual agronomic practices predicted enhance one or more ecosystem outcome of the region relative to the baseline ecosystem outcome. Counterfactual agronomic practices are predicted to enhance one or more ecosystem outcome based on analysis of remote sensing data and or previously obtained agronomic data. To illustrate, continuing the above example, the counterfactual agronomic practices may include, clearing the field on a different day, applying nitrogen to the field on the second day, tilling the field on the third day, etc. These counterfactual agronomic practices are not included in the baseline scenario, but if included in a new scenario, may enhance one or more ecosystem outcome in the region relative to the baseline.

III. A GENERATING COUNTERFACTUAL SCENARIOS

[00105] FIG. 1 illustrates a counterfactual engine, according to an example embodiment. The counterfactual engine 100 includes a region generation module 110, a baseline generation module 120, an ecosystem outcome module 130, a counterfactual simulation module 140 (“counterfactual module”), a training module 150, and one or more models 160. The counterfactual engine 100 may include additional or fewer elements or modules, and the functionality of one or more of its elements may be attributable to different elements. Additionally, the elements or modules may have additional or different functionality than what is described herein such that the counterfactual engine 100 is able to generate a counterfactual scenario for an agronomic region.

[00106] The region generation module 110 generates one or more agronomic regions (“regions”). Typically, a region is an aggregation of one or more agronomic areas (“areas”). Additionally, in general, at least one of the areas in the region is the area for which the counterfactual engine 100 is generating a counterfactual scenario. An area may be, e.g., a field or a portion of a field. In these cases, the areas may be delineated by field boundaries. Thus, a region may be one or more fields or portions of fields as indicated by the field boundaries. To provide an example, the region generation module 110 may generate a region that includes all the fields in a county or may generate a region that includes all the fields less than a threshold distance away from a geographic point. In various embodiments, a region may be a Land Resource Region (LRR) or USDA agricultural statistical district.

[00107] In other words, the region generation module 110 generates regions including areas that share one or more common characteristics. The example regions above included areas that shared geographic commonalities, but other commonalties can also be used. For instance, a region may include areas having similar agronomic products, weather patterns, economic indicators (e.g., crop values, etc.), agronomic practices, management events, quality factors, and/or ecosystem impacts. To this end, recall that the region generation module 110 can generate regions that have more than one shared characteristic. For example, the region generation module 110 can generate a region including one or more areas within a particular state, having a same soil composition, growing the same crops, and employing an herbicide mid cultivation cycle.

[00108] In another example, the region generation module 110 may generate regions based on upstream or downstream commonalities of its constituent areas. For example, the region generation module 110 may generate a region including a particular production region, sourcing region, or climatic region. In some embodiments, a region generation module may generate a region including all fields delivering product to a particular storage or transportation facility, or region including all fields fulfilling a contract. Similarly, the region generation module 110 may generate a region including all fields using a particular agronomic product (e.g., an herbicide) or agronomic process (e.g., cover-cropping) in generating their agronomic products, or may generate a region including all fields that could be used as an offset or inset.

[00109] The baseline generation module 120 generates a baseline scenario for the region. As described above, the baseline scenario includes one or more agronomic practices that reflect what agronomic practices managers are typically implementing in the areas of the region. The baseline generation module 120 generates a baseline scenario based on historic and current agronomic data for the areas in the region. In some examples, the typical agronomic practices may reflect a mean or median of agronomic practices for areas in the region.

[00110] As stated above, generating a baseline scenario employs both historic agronomic data and current agronomic data. Historic agronomic data may include, e.g., historic weather measurements and weather patterns, previous agronomic scenarios and practices, academic literature, historic field on-field measurements, historic remote-sensing measurements, historic agronomic outcomes (e.g., various agronomic goals of a manager such as yield, impact, quality, etc.), historic agronomic advice from experts, etc. The current agronomic data may include, e.g., a current agronomic scenario or practice, current on field measurements, current remote sensing measurements, current agronomic outcomes, agronomic advice from experts, etc. Agronomic data may also include field experiment results, regional surveys, literature data, partner data, and/or on-farm expert data.

[00111] The baseline generation module 120 may also derive commonalities between areas for use in generating regions. For instance, the baseline generation module 120 may identify one or more areas having historically high corn yields based on historic on-field sensing and remote sensing data and generate a region including the regions with high yield. Other examples are also possible.

[00112] The ecosystem outcome module 130 generates ecosystem outcomes for areas and/or regions. As described above, the ecosystem outcome is, generally, a predicted (or expected) ecosystem impact (e.g., emissions level, carbon removal, etc.) for the region given the typical agronomic practices for areas in that region. Environmental impacts can be generated for any combination of agronomic practices. Therefore, the ecosystem outcome module 130 may generate an ecosystem impact for one or more counterfactual scenarios (based on the included agronomic practices) relative to a baseline scenario. The ecosystem outcome module 130 may generate ecosystem impacts based on historic agronomic data and/or current agronomic data. When calculating ecosystem impacts, the historic agronomic data may include data relevant for those calculations such as, e.g., nitrogen or carbon emissions and usage, area or region sizes, fuel consumption rates, etc. Notably, the ecosystem outcome module 130 may generate ecosystem outcomes that are positive (e.g., beneficial) or negative (e.g., detrimental) depending on the agronomic practices included in the scenario. Because of this, each ecosystem outcome, as described below, may differently influence generation of counterfactual scenarios.

[00113] Generally, the counterfactual engine 100 preferentially employs agronomic data, including on-field measurements, when generating scenarios and determining ecosystem impacts. However, in some situations, on-field measurements are unavailable and the counterfactual engine 100 may substitute remotely sensed and/or derived agronomic information for the on-field measurements. To illustrate, in an example, a field may have not reported several months of management events (e.g., tilling dates, planting dates, etc.). In this situation, the counterfactual engine 100 may access remote data of the field (e.g., a satellite image) and derive the missing on-field measurements from the remote data (e.g., by applying computer vision models to the image to detect tilling and planting). Various methods may be used to derive the missing on-field measurements from the remote data. Exemplary methods are provided in U.S. Pub. No. 2022/0215659, U.S. Pub. No. 2023/0154183, US 11,755,966, and PCT/US2023/029167, which are hereby incorporate by reference in their entirety.

[00114] Similarly, the counterfactual engine 100 may use remote measurements to validate on-fi eld measurements (or vice versa). For instance, the counterfactual engine 100 may validate an on-field weather event using remotely sensed historic weather information. Additional details on validating on-field measurements are provided below in “III.F Additional Considerations and Configurations.”

[00115] The counterfactual simulation module 140 generates one or more counterfactual scenarios for an area or region. As described above, a counterfactual scenario is a set of agronomic practices that includes at least one counterfactual agronomic practice. Depending on the included agronomic practices, the counterfactual scenario, if it had been implemented, would yield the same or different ecosystem impact. When a counterfactual scenario would yield the same ecosystem impact, it may yield no difference in an ecosystem attribute relative to a baseline scenario. There are two possibilities when the counterfactual scenario would yield a different ecosystem impact. The counterfactual scenario may yield an enhanced ecosystem impact, indicating that there would be a beneficial difference in an ecosystem attribute relative to the ecosystem attribute of the baseline scenario. On the other hand, the counterfactual scenario may yield a reduced ecosystem impact, indicating that there would be a detrimental difference in an ecosystem attribute relative to the ecosystem attribute of the baseline scenario. Stated alternatively, in some embodiments, the counterfactual scenario, if it had been implemented, would yield one or more an enhanced ecosystem impact. Additionally, in some embodiments, the counterfactual scenario, if it had been implemented, would yield no difference in an ecosystem attribute relative to the ecosystem attribute of the baseline scenario.

[00116] As stated above, the counterfactual simulation module 140 generates a variety of counterfactual agronomic practices for counterfactual scenarios. The determined counterfactual agronomic practices may vary depending on the region and the desired agronomic outcomes. For instance, the counterfactual agronomic practices may include adding an agronomic practice that is not included in the baseline agronomic scenario, removing an agronomic practice that is included in the baseline agronomic scenario, modifying an agronomic process included in the baseline agronomic scenario, etc. This process is described in greater detail hereinbelow.

[00117] The training module 150 and the models 160 are discussed in Section III.D titled “Training a Counterfactual Engine.” In short, the counterfactual engine 100 may perform the aforementioned functionality while training a counterfactual model (e.g., models 160) to generate counterfactual scenarios. In turn, the counterfactual model may be applied to agronomic data to generate counterfactual scenarios.

III.B IMPLEMENTING A COUNTERFACTUAL ENGINE

First Example

[00118] FIG. 2 illustrates a first example workflow for generating counterfactual scenarios, according to an example embodiment. The workflow 200 may include additional or fewer elements, and/or one or more elements of the workflow 200 may be repeated. The workflow 300 may be performed by the counterfactual engine 100.

[00119] At 210, the counterfactual engine 100 receives a request from a user to generate a counterfactual scenario for an area. The request may include agronomic data for the area, desired agronomic or ecosystem outcomes, and any other pertinent information to generate the counterfactual scenario.

[00120] At 220, the counterfactual engine 100 applies a counterfactual model to the region and its associated agronomic data to determine counterfactual scenarios for the received area. [00121] At 222, the counterfactual engine 100 applies the counterfactual model to historical agronomic data and current agronomic data to determine a region including the received area. The determined region may include one or more additional areas that share one or more commonalities with the received area.

[00122] At 224, the counterfactual engine 100 applies the counterfactual model to historical agronomic data and current agronomic data to determine a baseline scenario for the region. The baseline scenario includes the typical agronomic practices of the areas in the region. The typical agronomic practices are those practices which yield the agronomic data for areas in the region or the region itself (i.e., the agronomic data indicates the typical agronomic practices).

[00123] At 226, the counterfactual engine 100 applies the counterfactual model to historical agronomic data and current agronomic data to determine a baseline ecosystem impact for the region. The baseline ecosystem impact is a quantification of an ecosystem impact of the typical agronomic practices in the baseline scenario.

[00124] At 228, the counterfactual engine 100 applies the counterfactual model to the historical agronomic data, current agronomic data, baseline agronomic scenario, and baseline ecosystem impact to generate a counterfactual scenario for the received area. The counterfactual scenario includes one or more counterfactual agronomic practices. The counterfactual scenario may also include one or more of the previously determined typical agronomic practices. The counterfactual scenario has a lower ecosystem impact than the baseline scenario due to the included counterfactual agronomic practices.

[00125] At 230, the counterfactual engine 100 provides the counterfactual scenario for the area to the manager.

Second Example

[00126] FIG. 3 illustrates a second example workflow for generating counterfactual scenarios, according to an example embodiment. The workflow 300 may include additional or fewer elements, and/or one or more elements of the workflow 300 may be repeated. The workflow 300 may be performed by the counterfactual engine 100.

[00127] At 310, the counterfactual engine 100 accesses historic agronomic data and/or current agronomic data. The counterfactual engine 100 may access the agronomic data from one or more datastores containing said data over, e.g., a network.

[00128] At 320, the counterfactual engine generates a region including an area. The region may be generated based on one or more common characteristics between areas in the region. [00129] At 330, the counterfactual engine 100 generates a baseline scenario for the region. The baseline scenario reflects the set of typical agronomic practices of the region.

[00130] At 340, the counterfactual engine 100 may verify the baseline scenario. For instance, the counterfactual engine may verify typical agronomic practices derived from onfield agronomic data with remotely sensed agronomic data and/or replace missing on-field agronomic data with remotely sensed agronomic data.

[00131] At 350, the counterfactual engine 100 generates an ecosystem impact for the baseline scenario. One or more soil data models may be applied to the baseline scenario to quantify baseline attributes, such as ecosystem impact.

[00132] At 360, the counterfactual engine 100 generates a counterfactual scenario including one or more counterfactual agronomic practices. The counterfactual scenario may be beneficial, negative, or neutral for the region or area relative to the baseline scenario. A counterfactual scenario may consider the additive effects of one or more agronomic practices over time, for example in some circumstances a no-till practice is predicted to result in a detrimental ecosystem outcome in the first year after adoption and but generates a beneficial effect in later years. An optimized counterfactual scenario may include an initially detrimental ecosystem outcome combined with one or more agronomic practices having immediate positive effects.

Third Example

[00133] FIG. 4 illustrates a third example workflow for generating counterfactual scenarios, according to an example embodiment. The workflow 400 may include additional or fewer elements, and/or one or more elements of the workflow 400 may be repeated. The workflow 400 may be performed by the counterfactual engine 100.

[00134] At 410, the counterfactual engine 100 accesses agronomic data for areas in a region (e.g., from a database). The agronomic data include, at least, agricultural practices for each area. The agronomic data may also include various data obtained from growers regarding fields, as well as remote sensing data.

[00135] In an example, the counterfactual engine 100 may filter the agronomic data. For instance, the counterfactual engine 100 may filter the agronomic data to include grower- entered events that pass data validation and do not require any form of gap-filling (e.g., using remote sensing data). In some cases, the counterfactual engine 100 filters the agronomic data to include historic data in a manner that reduces bias towards conservation practices (e.g., from before a specified date or practice implementation that included conservation practices). [00136] In an example, the counterfactual engine 100 may be applied to a region representing an agricultural statistical district (ASD). In this context, the workflow 400 determines the average (or median) agricultural practices for the region at the scale of the agricultural statistical district.

[00137] At 420, the counterfactual engine 100 aggregates agricultural data from across the region (e.g., from the areas in the region) to determine a regional baseline scenario. As noted above, in some examples, the events are filtered prior to aggregation.

[00138] In an example, the counterfactual engine 100 employs remote sensing data to determine if an ASD has majority no-till or majority cover crops, for example, detection of common practices within a region and exclusion of effects of those common agronomic practices may be required by a methodology. If the counterfactual engine 100 determines either is true, certain management events (e.g., tillage events) are removed from the regional baseline, and other management events (e.g., cover crop events) are added to the regional baseline, respectively.

[00139] At 430, the counterfactual engine 100 simulates counterfactual scenarios for the region. The counterfactual engine 100 does so by applying a plurality of agricultural practice modifications to the regional baseline scenario. Based on the results of the simulation, the counterfactual engine 100 generates a counterfactual agronomic scenario for the region. Fourth Example

[00140] FIG. 5 illustrates a fourth example workflow for generating counterfactual scenarios, according to an example embodiment. The workflow 500 may include additional or fewer elements, and/or one or more elements of the workflow 500 may be repeated. The workflow 500 may be performed by the counterfactual engine 100.

[00141] At 510, the counterfactual engine 100 accesses a predetermined crop management scenario for a geographical region. The predetermined crop management scenario may include, e.g., the typical agronomic practices for the region. In some cases, the predetermined crop management scenario may be based on agronomic data. For instance, in some examples, the agronomic practices may be based on subject matter expert feedback, remote sensing data, and/or USDA data.

[00142] At 520, the counterfactual engine 100 aggregates agronomic data for areas in the region. The agronomic data include agronomic practices for each area.

[00143] At 530, the counterfactual engine 100 selects at least one area in the region having agronomic practices best matching the agronomic practices of the crop management scenario (e.g., a field that most accurately implements the agronomic practices in the accessed crop management scenario). The agronomic practices for this area represent the regional baseline scenario. In some embodiments, specific growing seasons are selected for those crops/regions that have entire crop growing seasons of data that closely match the pre-defined scenarios. This allows modeling of the agricultural activity as an entire unit.

[00144] At 540, the counterfactual engine 100 may filter agronomic data to grower-entered and gap-filled events that include all the required events for a given crop growing season. In various embodiments, historic data (those before a project start date) are used to reduce bias towards conservation practices.

[00145] At 550, the counterfactual engine 100 imputes missing data. Various methods may be used to impute missing data during aggregation. Exemplary methods are provided in U.S. Pub. No. 2022/0215659, U.S. Pub. No. 2023/0154183, US 11,755,966, and PCT/US2023/029167, which are hereby incorporate by reference in their entirety.

[00146] At 560, the counterfactual engine 100 simulates counterfactual ecosystem attributes (e.g., ecosystem impacts, quality factors, agronomic outcomes, etc.) using counterfactual scenarios. The counterfactual scenarios are simulated for the geographic region by applying agronomic practice changes to the baseline scenario for the region. [00147] In various embodiments, soil data are provided to the model for this simulation step. In some embodiments, soil data are obtained from actual soil samples from within an ASD. In some embodiments, soil data are obtained from an existing database. For example, in some embodiments, information about a given ASD is obtained from the Gridded SSURGO (gSSURGO) database. In various embodiments, the management practices of each field also inform the weighting of model outputs when generating an average. When averaging effect sizes, the weight of various implementation rates will inform ranges of estimates. This weighting can also include the synthetic ASD runs.

[00148] In order to determine ecosystem outcome across a range of regions, agronomic practices, etc., in some embodiments, each baseline scenario is run with all potential practices. The results may be averaged across all ASDs to get ranges for ecosystem estimates. Similarly, multiple soil/weather types may be run across a given ASD baseline/counterfactual combination and an average obtained. The baseline scenarios can serve as the spin-up management data for DayCent-CR runs, or inputs for other quantification methods including empirical models, process-based models, machine learning models, biogeochemical models, ecosystem service models, models based on remotely sensed data, life-cycle assessment and inventory models, ensemble models, food web models, population models, crop growth models, or combinations thereof. m.C EXAMPLE USE CASES

[00149] As described above, the counterfactual agronomic practices included in counterfactual scenarios may vary depending on a variety of factors including, e.g., the location of the region, agronomic outcomes, produced crops, etc.

[00150] FIGS. 6-18 illustrate some example counterfactual scenarios and counterfactual agronomic practices generated by a counterfactual engine 100. In the illustrated examples, the agronomic data are aggregated into a region at the county level (and may further be aggregated into ASD), data are filtered to only pre-practice change data (e.g., before ecosystem impact considerations), and soil samples are selected randomly from ASDs.

DayCent was used for simulations to generate the counterfactual scenarios, including counterfactual agronomic practices.

[00151] In this example, the following counterfactual agronomic practices are suggested as part of a counterfactual agronomic scenario. As a result of these counterfactual agronomic practices, the ecosystem impact of the region (e.g., measures of N2O and SOC) would be reduced.

Table 1 : Example Counterfactual Agronomic Practices

[00152] These counterfactual agronomic practices (and thereby scenarios) are plotted in FIGs. 6-18. The most frequently seen combinations of counterfactual agronomic practices are seen in FIGS. 6-10. In this example, the data is aggregated from a specific set of carbon growers in the identified states and may not be representative of the counties or states at large. Counties are filled based on the number of fields in each county with the combination of agronomic practices in question. Counties where there is no observed change in agronomic practice (e.g., maintain the baseline scenario) are filled in grey.

[00153] FIG. 6 illustrates regions where a counterfactual scenario including no-till was generated, according to an example embodiment.

[00154] FIG. 7 illustrates regions where a counterfactual scenario including cover crop and no-till were generated, according to an example embodiment.

[00155] FIG. 8 illustrates regions where a counterfactual scenario including new cash crops in rotation and no-till were generated, according to an example embodiment.

[00156] FIG. 9 illustrates regions where a counterfactual scenario including cover crops were generated, according to an example embodiment.

[00157] FIG. 10 illustrates regions where a counterfactual scenario including tillage impact reduction and tillage pass reduction were generated, according to an example embodiment.

[00158] FIG. 11 illustrates regions where a counterfactual scenario including fertilizer reduction and no-till were generated, according to an example embodiment.

[00159] FIG. 12 illustrates regions where a counterfactual scenario including tillage impact reduction were generated, according to an example embodiment.

[00160] FIG. 13 illustrates regions where a counterfactual scenario including fertilizer reduction was generated, according to an example embodiment.

[00161] FIG. 14 illustrates regions where a counterfactual scenario including new cash crops in rotation was generated, according to an example embodiment.

[00162] FIG. 15 shows counties in which fertilizer application was delayed, fertilizer was reduced, and no till was generated, according to an example embodiment.

[00163] FIG. 16 shows the outlines of the 9 ASDs in Iowa, according to an example embodiment.

[00164] FIG. 17 shows counties flagged for reduced tillage as a common practice, according to an example embodiment.

[00165] FIG. 18 shows counties flagged for no-till as a common practice. No clear pattern exists for reduced tillage, but ASD 70 is largely no-till common practice and will be removed from simulations for that specific practice, according to an example embodiment.

III.D TRAINING A COUNTERFACTUAL ENGINE [00166] The counterfactual engine 100 may be implemented as a counterfactual model (e.g., a model in models 160). In that regard, the counterfactual engine 100 includes a training module 150 that may train the one or more models using the functionality of the region generation module 110, the baseline generation module 120, the ecosystem outcome module 130, and the counterfactual simulation module 140. In other words, the training module 150 may generate a model 160 that leverages the functionality of the various elements of the counterfactual engine 100 to generate counterfactual scenarios.

[00167] To demonstrate, FIG. 19 illustrates an example workflow for training a counterfactual emission module to generate counterfactual scenarios, according to an example embodiment. The workflow 1900 may include additional or fewer elements, and/or one or more elements of the workflow 1900 may be repeated. The workflow 1900 may be performed by the counterfactual engine 100 employing the training module 150.

[00168] At 1910, the counterfactual engine 100 accesses agronomic data for any number of areas. The training module 150 may generate various data structures (e.g., feature vectors) representing the agronomic data for each area. For instance, a data structure may include an entry for each field in the area, and that entry may include an array of sub-entries representing various agronomic data for the field.

[00169] At 1920, the counterfactual engine 100 determines a variety of regions using the agronomic data of the areas. The counterfactual engine 100 generates a region using one or more areas having a similar characteristic. The training module 150 may generate a data structure representing the various regions and their corresponding areas, and identify correspondences between the data structures representing the areas and the regions. The training module 150 may train a model to identify areas for regions having similar characteristics using the data structures for the areas, the regions, and their identified correspondences.

[00170] At 1930, the counterfactual engine 100 determines a baseline scenario for the regions. Each baseline scenario includes one or more agronomic practices representing typical agronomic practices for fields in the region. The training module 150 may generate data structures representing the agronomic practices in a baseline scenario for each field or region. The training module 150 may train a model to generate a baseline scenario for a region based on identified correspondences between the data structures for areas, regions, accessed agronomic data and baseline scenarios. [00171] At step 1940, the counterfactual engine 100 determines a baseline ecosystem impact for the agronomic regions. The baseline ecosystem impact may be based on the agronomic data and the agronomic practices in the baseline agronomic scenarios. The training module 150 may generate a data structure representing the ecosystem impacts. The training module 150 trains a model to identify baseline ecosystem impacts based on identified correspondences between the generated data structures and the agronomic data.

[00172] At step 1950, the counterfactual engine 100 accesses agronomic practice modifications that may be implemented as counterfactual agronomic practices. The agronomic practice modifications may be accessed from, e.g., literature, experts, databases, current or historical agronomic data, etc. The counterfactual engine 100 may derive agronomic practice modifications from the accessed information (e.g., deriving a counterfactual agronomic practice from expert supplied suggestions). Each of the counterfactual agronomic practices is associated with a change in an ecosystem impact. The changes in ecosystem impact may be based on literature, experts, databased, current or historical agronomic data, etc. As such, the training module 150 may generate a data structure representing the various agronomic practice modifications and their corresponding changes to an ecosystem impact.

[00173] The counterfactual engine 100 simulates various counterfactual agronomic scenarios for regions based on the generated data structures. Each simulated counterfactual agronomic scenario includes one or more counterfactual agronomic practices, which induce a change to an ecosystem impact for the region. The training module 150 may generate data structures representing the results of the various simulations of counterfactual agronomic scenarios.

[00174] At 1950, the counterfactual module employs the training module 150 to train a model to generate counterfactual scenarios for different regions. Training the model includes accessing the various data structures created above (e.g., representing areas, regions, baseline scenarios, ecosystem impacts, and simulated counterfactual scenarios), identifying latent associations between those data structures, and creating a model that will generate appropriate counterfactual agronomic scenarios given a set of input data (e.g., an area, a region, pertinent agronomic data, etc.). The counterfactual scenarios provide counterfactual agronomic practices that reduce the ecosystem impact of an area or region relative to the ecosystem impact for its baseline scenario. [00175] In various examples, the workflow 1900 may employ additional models to generate information. For instance, a model may be used to generate a region, a baseline agronomic scenario, and/or ecosystem impacts.

IV. ADDITIONAL EXAMPLE CONFIGURATIONS

Areas and Regions

[00176] As described above, the region generation module 110 may generate areas in regions in a variety of manners. In an example, the boundaries for each field within the geographic region or area may be based on at least one image obtained from a time series of satellite imagery. That is, the region generation module 110 may apply a field recognition model to the image, and the field recognition model may output the boundaries of fields in the image.

[00177] Additionally, in some examples, the region generation module 110 may generate an area and/or region based on the crop production therein. The generated area or region may therefore be based on crop production data included in the agronomic data. The crop production data generally include, at least, a crop type and at least one crop production practice.

In this case, a remote image depicting a portion of the geographic region, additionally comprises an indication of a crop type in the area or region, and may be derived using computer vision models.

[00178] Images used in these recognition methodologies can be variable. For instance, an image may include a portion of a region, include the entire region, include a subsection of the region (e.g., an area), a subsection of an area, etc. Similarly, an image may include an entire area or field.

[00179] The region generation module 110 may generate areas and regions based on one or more of: a latitude and longitude, an address or parcel id, a city, a county, a state, a street address, a region defined by having a set of environmental attributes, a region defined by proximity to one or more grain elevators, a sourcing region, a boundary file, and a shape drawn on a map presented within a GUI of a user device. In some embodiments, an area or region is a grouping of counties by geography, climate, and cropping practices. A region or area may be a grouping of fields based one or more characteristics including soil type, terrain, elevation, mean temperature, annual precipitation, length of growing season, soil composition, and or any other factor influencing crop growth, the meet to conserve soil moisture, or the use of irrigation. [00180] The indication of a geographic region may be an area surrounding an indicated location. In this case, the area surrounding an indicated location may include one or more of: a county containing an address or latitude and longitude, a state containing an address or latitude and longitude, a user defined radius surrounding a longitude and latitude, and an area defined by connecting points representing an average drive time from a location. Graphical User Interfaces

[00181] The counterfactual engine 100 may display information to users at any point during the processes described above using a graphical user interface (“GUI”).

[00182] For instance, in an example, the counterfactual engine 100 may display an image depicting a portion of the region within a GUI of a user device. The displayed image may comprise one or more field boundaries, one or more predicted agronomic practices, and or an estimate of an ecosystem attribute. The counterfactual engine 100 may generate a GUI comprising an image depicting a portion of the geographic region. A user of the GUI may navigate between images representing a portion of a geographic region. In this case, navigating between images may display a different area or field in a region with each image. [00183] The counterfactual engine 100 may generate a GUI that displays information in various ordered or structured manners. For example, the GUI may display information in an order based on a presence of an ecosystem outcome, an absence of an ecosystem outcome, an amount of an ecosystem outcome, a permanence of an ecosystem outcome, or a difference in an ecosystem outcome. In another example, the GUI may display information in an order based on acreage of a field. In another example, the GUI may display information in an order based on one or more of a crop type or agronomic practice. For example, a crop type or agronomic practice may be associated with a featured production area in the past, present, or future (for example, a recommendation of a crop type or agronomic practice to be used in the future).

[00184] The counterfactual engine 100 may generate a GUI including an image depicting a portion of the geographic region, and the image may be overlaid with one or more of an indication of presence, absence, quantification, and permanence of the ecosystem outcome of one or more production areas, an indication of a crop type of one or more production areas, an indication of the ecosystem impact of a unique set of crop practices for one or more production areas. [00185] The counterfactual engine 100 may generate a GUI that allows a user to define a radius or draw a region by interacting with a map presented within a GUI. The distance may be used to generate various images and depictions as described above.

[00186] The counterfactual engine 100 may generate (e.g. automatically, without human interaction) a GUI comprising a map displaying field boundaries generated from remote sensing data corresponding to a region of interest. In some embodiments, a user may modify, add, or delete field boundaries with in the GUI. In some embodiments, a user may select (for example, by tapping on a touch screen of a user device) one or more of the displayed field boundaries to include in a region.

Selecting Counterfactual Agronomic Prescriptions

[00187] As described above, the counterfactual engine 100 generates counterfactual scenarios including one or more counterfactual agronomic practices for an agronomic area. Moreover, the counterfactual engine 100 sends the counterfactual scenario to a user interested in the agronomic area. The counterfactual engine 100 may take several approaches to selecting and sending counterfactual agronomic scenarios and counterfactual agronomic practices to the user.

[00188] For example, in some examples, the counterfactual agronomic scenario including most beneficial ecosystem is selected and sent to the user. The most beneficial ecosystem outcome may be the smallest amount of greenhouse gas emission, the largest amount of soil carbon sequestration, the least amount of water used, the least amount of chemical input applied, the most biodiversity, a lowest risk of reversal of a beneficial ecosystem outcome, a lowest risk of reversal of a beneficial ecosystem outcome, a greatest permanence of a beneficial ecosystem outcome, or a combination of thereof.

[00189] In another example, the most beneficial ecosystem impact is the most beneficial ecosystem outcome for a single field. Similarly, the most beneficial ecosystem outcome is based on a quantified ecosystem outcome aggregated for a group of fields associated with a single entity, all fields within the region, all of the fields within the region likely to grow a crop type, all fields within the region not known to have previously implemented the unique set of agronomic practices generating the most beneficial ecosystem outcome, etc.

[00190] Additionally, as described about, the counterfactual agronomic scenarios efficacy is measured relative to the baseline agronomic scenario for that region or area. That is, the counterfactual agronomic scenario is weighed relative to “the expected ecosystem outcome.” There are several measures of an expected ecosystem outcome. [00191] For instance, in some examples, the expected ecosystem outcome may be the ecosystem outcome for each unique set of crop practices within the field boundary weighted by the co-occurrence probability of that set of practice changes within the geographic region. In additional examples, the expected ecosystem is the ecosystem outcome for the set of crop practices implemented within the field boundary in the most recent crop season. Within these examples, the most recent crop season may be the most recent crop season for which both planting and harvest crop practice data are available, the most current crop season, the crop season immediately preceding the most current crop season.

[00192] In an example, the expected ecosystem outcome is the ecosystem outcome for the crop practices implemented within the field boundary in the most recent historical crop season. Additionally, in an example, the expected ecosystem outcome may be the average ecosystem outcome across fields within the geographic region. Within this context, the region is all fields within the region, all fields within the region having a single soil type and single weather type, or all fields within the region managed by a single entity.

Aggregation and Filtering

[00193] As described above, the counterfactual engine 100 may aggregate agronomic data when generating baselines scenarios. Additionally, when aggregating data, the data may be filtered according to several factors. Table 2 illustrates several parameters about which agronomic data may be aggregated and how that agronomic data may be aggregated.

Table 2: Aggregation Parameters and Details

[00194] Additionally, the counterfactual engine 100 may perform specific actions if certain remote sensing data are identified. For instance, if remote sensing data are used to determine if an ASD as a whole has majority no-till or majority cover crops, the counterfactual engine may filter or manage that data. For instance, in this example, if either of the above conditions are determined to be true, the counterfactual engine 100 removes tillage events and/ or adds cover crop events, respectively.

[00195] Other filtering and aggregation processes are also possible. Generating Environmental Footprints and Counterfactual Simulations [00196] Notably, the counterfactual engine 100 aggregates data when generating a region including an area, and then generates counterfactual scenarios for that region based on the aggregated data. However, in other embodiments, the counterfactual engine 100 may generate counterfactual scenarios for each area in a region, and then aggregate the counterfactual scenarios for the areas to generate the counterfactual scenario for the region. All of the various processes described herein may be applied to each methodology.

[00197] Additionally, in order to provide a comprehensive ecosystem impact assessment (i.e., determine an ecosystem impact of a scenario), each baseline scenario is run with all potential practice changes (e.g., all possible counterfactual agronomic practices). The results of the simulations may be averaged across all ASDs to get ranges for impact assessments. Similarly, multiple soil/weather types may be run across a given ASD baseline/practice combination and an average obtained. The baseline scenarios serve as the spin-up management data for DayCent-CR runs, or inputs for other quantification methods including empirical models, process-based models, machine learning models, biogeochemical models, ecosystem service models, models based on remotely sensed data, life-cycle assessment and inventory models, ensemble models, food web models, population models, crop growth models, or combinations thereof.

[00198] When performing these calculations and simulations, in some examples, the agronomic data may include soil data. The soil data may be obtained from actual soil samples from within an ASD, or may be obtained from an existing database. For example, in some embodiments, information about a given ASD is obtained from the Gridded SSURGO (gSSURGO) database.

[00199] Additionally, in some examples, generating impact assessments (e.g., ecosystem impacts and counterfactual scenarios) may include employing one or more biogeochemical models, empirical models, process-based models, machine learning models, ecosystem service models, models based on remotely sensed data, life-cycle assessment and inventory models, ensemble models, food web models, population models, direct measurement and statistical sample designs, crop growth models, or combinations thereof are applied to the inputs to obtain the baseline and counterfactual scenarios.

[00200] For example, the counterfactual engine 100 may employ DayCent to calculate changes in N2O emissions from direct and indirect sources based on changing agronomic practices. DayCent provides changes in SOC sequestration and direct N2O emissions. The counterfactual engine 100 also includes both direct and indirect N2O emissions from default equations for comparison and total emissions quantification, which may be used to identify changes in ecosystem impacts when using counterfactual agronomic practices. Typically, these results are associated with an uncertainty which may further be used to train a counterfactual model.

[00201] In another example, the counterfactual engine 100 may employ Life Cycle Impact Assessment (LCIA) to calculate changes in various ecosystem attributes (which may be used in generating baseline and counterfactual scenarios, ecosystem impacts, etc.). To get LCIA results, the counterfactual engine 100 accesses a database comprising activities and associated ecosystem attributes (e.g., GHG emissions) resulting from agronomic practices. This allows the counterfactual engine to determine mappings between various agronomic practices and ecosystem impacts, which, in turn, may be used by the counterfactual engine 100 to train a counterfactual model.

[00202] There are several types of agronomic data in the LCIA which may prove useful to the counterfactual engine 100.

[00203] For instance, the LCIA includes agronomic data regarding fertilizers. The agronomic data reflects the ecosystem impact scales with rate and depends on fertilizer type. The production of the fertilizer and the diesel used to apply are all accounted for. The diesel used scales with field area. The LCIA may include agronomic data regarding tillage. The agronomic data reflects that ecosystem impact scales with area tilled due to the fuel used for tillage passes. The LCIA may include agronomic data regarding organic amendments. The agronomic data reflects that ecosystem impact depends on the fuel used to apply the amendments and therefore scales with area. It does not account for production by animals. The LCIA may include agronomic data regarding irrigation. The agronomic data reflects that ecosystem impact depends on the electricity for electricity used to run irrigation and, therefore, scales with field area.

Validating and Imputing Agronomic Data

[00204] Various remote sensing algorithms may be used to validate grower-supplied data. Such algorithms may be used to determine geographical boundaries of one or more fields, the presence or absence of one or more management events, evidential data related to one or more management events, or any of a variety of agriculturally relevant attributes. In some embodiments, the prevalence of one or more crop types, and one or more management events, may be determined for each crop type, each crop production cycle and each sourcing region from which an ingredient within a processed product was procured. In various embodiments, a sourcing region or supply chain from which an ingredient within a processed product was procured may be documented by purchase records documenting the entry of a product into a sourcing region (for example a scale ticket documenting delivery of a product to an elevator) or supply chain (for example, a delivery confirmation documenting delivery of a product to a processing facility of buyer). In various embodiments, a sourcing region or supply chain from which an ingredient within a processed product was procured may be estimated based on historical or regional practice, for example, in a relevant time period a buyer purchased a type of agricultural commodity from a limited number of known producers, or from a known set of aggregators or intermediate processors each sourcing a known percentage or amount of a type of agricultural commodity from a number of known producers, or from a certain region (for example, one or more counties, or one or more states). Various methods may be used to impute missing data during aggregation. Exemplary methods are provided in U.S. Pub. No. 2022/0215659, which is hereby incorporate by reference in its entirety.

[00205] In various embodiments, a method is provided, comprising: receiving an indication of a geographic region, wherein the geographic region contains at least one production area; receiving a time series of satellite imagery, the time series of satellite imagery covering at least the geographic region; generate boundaries for each field within the geographic region based on at least one image obtained from the time series of satellite imagery; for each field boundary within the geographic region, detect crop production data based on the time series of satellite imagery, where the crop production data includes at least one crop production practice; for each unique set of crop practices detected within the geographic region, generate an ecosystem outcome for one or more of the field boundaries within the geographic region.

[00206] In various embodiments, a method is provided, comprising: receiving an indication of a geographic region containing a plurality of fields; receiving a time series of satellite imagery, the time series of satellite imagery covering at least the geographic region; accessing time series of historical crop production data for one or more field within the geographic region, wherein at least one crop field is missing crop production data for at least one period within the time series; generate estimated values for missing crop production data based on the time series of satellite imagery; generate an ecosystem outcome for each field within the geographic region; displaying within a GUI of a user device an image depicting a portion of the geographic region, wherein the image is overlaid with one or more of an indication of presence, absence, quantification, and permanence of the generated ecosystem outcome of one or more production areas.

V. EXAMPLE COMPUTER SYSTEM

[00207] FIG. 20 is a block diagram illustrating components of an example machine for reading and executing instructions from a machine-readable medium. Specifically, FIG. 20 shows a diagrammatic representation of, e.g., a counterfactual engine 100 in the example form of a computer system 2000. The computer system 2000 can be used to execute instructions 2024 (e.g., program code or software) for causing the machine to perform any one or more of the methodologies (or processes) described herein. In alternative embodiments, the machine operates as a standalone device or a connected (e.g., networked) device that connects to other machines. In a networked deployment, the machine may operate in the capacity of a server machine or a client machine in a server-client system environment (e.g., a counterfactual engine connected to one or more computing devices), or as a peer machine in a peer-to-peer (or distributed) system environment.

[00208] The machine may be a server computer, a client computer, a personal computer (PC), a tablet PC, a set-top box (STB), a smartphone, an internet of things (loT) appliance, a network router, switch or bridge, or any machine capable of executing instructions 2024 (sequential or otherwise) that specify actions to be taken by that machine. Further, while only a single machine is illustrated, the term “machine” shall also be taken to include any collection of machines that individually or jointly execute instructions 2024 to perform any one or more of the methodologies discussed herein.

[00209] The example computer system 2000 includes one or more processing units (generally processor 2002). The processor 2002 is, for example, a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), a controller, a state machine, one or more application specific integrated circuits (ASICs), one or more radio-frequency integrated circuits (RFICs), or any combination of these. The computer system 2000 also includes a main memory 2004. The computer system may include a storage unit 2016. The processor 2002, memory 2004, and the storage unit 2016 communicate via a bus 2008.

[00210] In addition, the computer system 2000 can include a static memory 2006, a graphics display 2010 (e.g., to drive a plasma display panel (PDP), a liquid crystal display (LCD), or a projector). The computer system 2000 may also include alphanumeric input device 2012 (e.g., a keyboard), a cursor control device 2014 (e.g., a mouse, a trackball, a joystick, a motion sensor, or other pointing instrument), a signal generation device 2018 (e.g., a speaker), and a network interface device 2020, which also are configured to communicate via the bus 2008.

[00211] The storage unit 2016 includes a machine-readable medium 2022 on which is stored instructions 2024 (e.g., software) embodying any one or more of the methodologies or functions described herein. For example, the instructions 2024 may include the functionalities of modules of the system described in FIG. 1. The instructions 2024 may also reside, completely or at least partially, within the main memory 2004 or within the processor 2002 (e.g., within a processor’s cache memory) during execution thereof by the computer system 2000, the main memory 2004 and the processor 2002 also constituting machine-readable media. The instructions 2024 may be transmitted or received over a network 2026 via the network interface device 2020.

[00212] While machine-readable medium 2022 is shown in an example embodiment to be a single medium, the term “machine-readable medium” should be taken to include a single medium or multiple media (e.g., a centralized or distributed database, or associated caches and servers) able to store the instructions 2024. The term “machine-readable medium” shall also be taken to include any medium that is capable of storing instructions 2024 for execution by the machine and that cause the machine to perform any one or more of the methodologies disclosed herein. The term “machine-readable medium” includes, but not be limited to, data repositories in the form of solid-state memories, optical media, and magnetic media.

VI. ADDITIONAL CONSIDERATIONS

[00213] In the description above, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the illustrated system and its operations. It will be apparent, however, to one skilled in the art that the system may be operated without these specific details. In other instances, structures and devices are shown in block diagram form in order to avoid obscuring the system.

[00214] Reference in the specification to “one embodiment,” “an embodiment,” “one configuration,” or “a configuration” means that a particular feature, structure, or characteristic described in connection with the embodiment or configuration is included in at least one embodiment or configuration of the system. The appearances of the phrase “in one embodiment” or “in a configuration” in various places in the specification are not necessarily all referring to the same embodiment.

[00215] Some portions of the detailed descriptions are presented in terms of algorithms or models and symbolic representations of operations on data bits within a computer memory. An algorithm is here, and generally, conceived to be steps leading to a desired result. The steps are those requiring physical transformations or manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It has proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like.

[00216] It should be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise as apparent from the following discussion, it is appreciated that throughout the description, discussions utilizing terms such as “processing” or “computing” or “calculating” or “determining” or “displaying” or the like, refer to the action and processes of a computer system, or similar electronic computing device, that manipulates and transforms data represented as physical (electronic) quantities within the computer system’s registers and memories into other data similarly represented as physical quantities within the computer system memories or registers or other such information storage, transmission or display devices. [00217] Some of the operations described herein are performed by a computer physically mounted within a machine. This computer may be specially constructed for the required purposes, or it may comprise a general-purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored in a computer readable storage medium, such as, but is not limited to, any type of disk including floppy disks, optical disks, CD-ROMs, and magnetic-optical disks, read-only memories (ROMs), random access memories (RAMs), EPROMs, EEPROMs, magnetic or optical cards, or any type of non-transitory computer readable storage medium suitable for storing electronic instructions.

[00218] The figures and the description above relate to various embodiments by way of illustration only. It should be noted that from the following discussion, alternative embodiments of the structures and methods disclosed herein will be readily recognized as viable alternatives that may be employed without departing from the principles of what is claimed.

[00219] One or more embodiments have been described above, examples of which are illustrated in the accompanying figures. It is noted that wherever practicable similar or like reference numbers may be used in the figures and may indicate similar or like functionality. The figures depict embodiments of the disclosed system (or method) for purposes of illustration only. One skilled in the art will readily recognize from the following description that alternative embodiments of the structures and methods illustrated herein may be employed without departing from the principles described herein.

[00220] Some embodiments may be described using the expression “coupled” and “connected” along with their derivatives. It should be understood that these terms are not intended as synonyms for each other. For example, some embodiments may be described using the term “connected” to indicate that two or more elements are in direct physical or electrical contact with each other. In another example, some embodiments may be described using the term “coupled” to indicate that two or more elements are in direct physical or electrical contact. The term “coupled,” however, may also mean that two or more elements are not in direct physical or electrical contact with each other, but yet still co-operate or interact with each other. The embodiments are not limited in this context.

[00221] As used herein, the terms “comprises,” “comprising,” “includes,” “including,” “has,” “having” or any other variation thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, article or apparatus that comprises a list of elements is not necessarily limited to only those elements but may include other elements not expressly listed or inherent to such process, method, article or apparatus. Further, unless expressly stated to the contrary, “or” refers to an inclusive or and not to an exclusive or. For example, a condition A or B is satisfied by any one of the following: A is true (or present), and B is false (or not present), A is false (or not present), and B is true (or present), and both A and B is true (or present).

[00222] In addition, use of the “a” or “an” are employed to describe elements and components of the embodiments herein. This is done merely for convenience and to give a general sense of the system. This description should be read to include one or at least one and the singular also includes the plural unless it is obvious that it is meant otherwise.

[00223] Upon reading this disclosure, those of skill in the art will appreciate still additional alternative structural and functional designs for a system and a process for performing the various functionalities described herein. Thus, while particular embodiments and applications have been illustrated and described, it is to be understood that the disclosed embodiments are not limited to the precise construction and components disclosed herein. Various modifications, changes and variations, which will be apparent to those, skilled in the art, may be made in the arrangement, operation and details of the method and apparatus disclosed herein without departing from the spirit and scope defined in the appended claims.