118th Congress: Resilient Agriculture, Society & Environment

Over the past several years, instability has been a national and global constant. The COVID-19 pandemic upended supply chains and production systems. Floods, hurricanes, heat waves, droughts, and fires have imposed catastrophic consequences and forced people to reconsider where they can safely live. Russia’s war with Ukraine and other geopolitical conflicts have forced countries around the world to scramble for reliable energy sources.

Congress must act decisively to fortify the United States against these and future destabilizing threats. Priorities include revitalizing U.S. agriculture to ensure a dependable, affordable, and diverse food supply; improving disaster preparation and response; and driving development and oversight of critical environmental technologies.

Revitalizing U.S. Agriculture. Every society needs a robust food supply to survive, thrive, and grow. But skyrocketing food prices and agricultural supply-chain disruptions indicate that our nation’s food supply may be on shaky ground. Congress can take measures to rebuild a world-leading U.S. agricultural sector that is sustainable amid evolving external pressures.

A first step is to invest in agricultural innovation and entrepreneurship. The 2018 Farm Bill created the Agriculture Advanced Research and Development Authority (AgARDA) as a driver of transformative progress in agriculture, but failed to equip the institution with a key tool: prize authority. Prizes have proven to be force multipliers for innovation dollars invested by many institutions, including other Advanced Research Projects Agencies (ARPAs). It would be simple for Congress to extend prize authority to AgARDA as well.

Prize authority at AgARDA would be especially powerful if coupled with additional support for agricultural entrepreneurship. Congress should fund the U.S. Department of Agriculture (USDA), the Small Business Administration (SBA), and the Minority Business Development Administration (MBDA) with $25 million per year for five years to jointly develop a “Ground Up” program to help Americans start small businesses focused on sustainable agriculture.

We must also begin viewing our nation’s soil as a strategic resource. Farmers and ranchers cannot succeed without good places to plant crops and graze livestock. But our nation’s fertile soil is being lost ten times faster than it is being produced. At this rate, many parts of the country will run out of arable land in the next 50 years. Some places—such as the Piedmont region of the eastern United States—already have. States including New Mexico, Illinois, and Nebraska have already introduced or passed legislation to preserve and restore soil health; Congress should follow their example. A comprehensive soil-health bill could, for instance, create bridge-loan projects for farmers transitioning to soil-protective farm practices, expand the USDA’s Environmental Quality Incentives Program (EQIP) program to cover such practices, fund USDA Extension offices to provide related technical assistance, and support regenerative agriculture in general.

Finally, Congress should extend funding for two programs that are delivering clear benefits to U.S. food systems. With major food production concentrated in five states, often far from major population centers, the farm-to-table pathway is extraordinarily susceptible to disruptions. The American Rescue Plan Act created the Food Supply Chain Guaranteed Loan Program to help small- and medium-sized enterprises strengthen this pathway, including through “aggregation, processing, manufacturing, storing, transporting, wholesaling or distribution of food.” This program should be continued and resourced going forward. In addition, the Bioproduct Pilot Program studies how materials derived from agricultural commodities can be used for construction and consumer products. This program increases economic activity in rural areas while also lowering commercialization risks associated with bringing bio-based products to market. Congress should extend funding for this program (currently set to expire after FY 2023) for at least $5 million per year through the end of FY 2028.

Improving Disaster Preparation and Response. Every year, Americans lose billions of dollars to natural hazards including hurricanes, wildfires, floods, heat waves, and droughts. We know these disasters will happen…yet only 15% of federal disaster funding is invested to blunt their effects. In particular, current disaster policy and practice lacks incentives for local governments to proactively reduce risks.

Congress can address this failure by amending aspects of the Stafford Act of 1988. In particular, Congress should redefine the disaster threshold in ways that factor in local capacity and ability to recover. Congress should also consider (i) reducing the federal cost share for disaster response, (ii) implementing other incentive models that may induce better local hazard-reduction decisions and improve long-term resilience, and (iii) strengthening existing incentive programs. For example, the National Flood Insurance Program (NFIP) Community Rating System (CRS) could be improved by requiring local governments to take stronger actions to qualify for reduced insurance rates and increasing transparency about how community ratings are calculated. 

Disaster management response is not the sole purview of FEMA. For example, the Community Development Block Grant Disaster Recovery (CDBG-DR) program positions the Department of Housing and Urban Development (HUD) as a primary disaster-response funder. To ensure efficiency and prevent duplication of effort, Congress must clarify the role of each federal agency involved in disasters.

Congress should also ensure adequate research funding to investigate evidence-based and cost-effective disaster mitigation and response strategies. A useful first step would be doubling the interagency Disaster Resilience Research Grant (DRRG) program, which already supports researchers in groundbreaking modeling, simulations, and solutions development to protect Americans from the most catastrophic consequences.

Driving Development and Oversight of Critical Environmental Technologies. Environmental technologies are critical to ensure energy and resource security. Congress can use market-shaping mechanisms to pull critical environmental technologies, such as carbon capture and storage (CCS), forward. Operation Warp Speed demonstrated breakthrough capacity of federally backed advance market commitments (AMCs) to incentivize rapid development and scaling of transformative technologies. Building on this example, Congress should authorize a $1 billion AMC for scalable carbon-removal approaches—providing the large demand signal needed to attract market entrants, and helping to advance a clean all-of-the-above energy portfolio. This approach could then be extended to other environmentally relevant applications, such as building infrastructure to enable next-generation transportation.

Congress must also ensure responsible deployment and reasonable oversight of new environmental technologies. For instance, DOE recently launched an ambitious “Carbon Negative Shot” to foster breakthroughs in carbon dioxide removal (CDR) technology, and is also leading an interagency CDR task force pursuing the advancement of many CDR approaches. But we lack a national carbon-accounting standard and tool to ensure that CDR initiatives are being implemented consistently, honestly, and successfully. Congress should work with the Department of Energy and the Environmental Protection Agency to address this assessment gap.

Similarly, the IRA appropriates over $405 million across federal agencies for activities including “the development of environmental data or information systems.” This could prove a prescient investment to efficiently guide future federal spending on environmental initiatives—but only if steps are taken to ensure that these dollars are not spent on duplicative efforts (for instance, water data are currently collected by 25 federal entities across 57 data platforms and 462 data types). Congress should therefore authorize and direct the creation of a Digital Service for the Planet “with the expertise and mission to coordinate environmental data and technology across agencies”, thus promoting efficiencies in the data enterprise. This centralized service could be established either as a branch of the existing U.S. Digital Service or as a parallel but distinct body.

Return to introduction

Assessing Agency-Reported Progress on the Justice40 Initiative

Question: What do family game nights and federal government initiatives have in common?

Answer: They’re both much easier to successfully start than to successfully finish.

Coordinating multiple stakeholders—each with their unique interests and perspectives—around a common goal is simply difficult. At FAS, we have yet to figure out how to best tackle family game nights. But we have found that for complex federal initiatives involving many agencies, taking the time to step back and assess progress to date often paves the way for continued future success. We also recognize that unless specifically tasked and resourced, Executive Branch agencies and offices generally lack capacity to do this on their own.

That’s why today, FAS is releasing an independent assessment of agency-reported progress on the Administration’s Justice40 Initiative—a landmark whole-of-government effort to ensure that 40% of the overall benefits of certain federal investments flow to disadvantaged communities that are marginalized, underserved, and overburdened by pollution.

The complete assessment is freely available here. A supplemental spreadsheet to the assessment is available here.

The assessment focuses on the 175 Justice40 recommendations issued by the White House Environmental Justice Advisory Council (WHEJAC) in May 2021. Key takeaways include:

Additional background and insights from the assessment are provided below. 

The WHEJAC and the Justice40 Initiative

President Biden launched the Justice40 Initiative within days of taking office in January 2021. Executive Order (E.O.) 14008, which created the Initiative, also established the first-ever White House Environmental Justice Advisory Council (WHEJAC). The WHEJAC comprises two dozen experts in environmental justice, climate change, disaster preparedness, racial inequity, and related fields. 

The WHEJAC’s mission is to provide advice and recommendations to the Chair of the Council on Environmental Quality (CEQ) and the White House Environmental Justice Interagency Council (IAC) “on how to increase the Federal Government’s efforts to address current and historic environmental injustice.” The WHEJAC’s first suite of recommendations, released in May 2021, included 175 specifically focused on the Justice40 Initiative. In May 2022, CEQ delivered a required report to Congress that included responses from the federal agencies named in each of these.

Assessing agency-reported progress on Justice40

To inform the WHEJAC’s future efforts, and to support ongoing implementation of the Justice40 Initiative, we at FAS conducted an independent assessment of the WHEJAC’s Justice40 recommendations and CEQ’s corresponding report. We emphasize that this assessment was scoped to elucidate key insights and trends from agency-reported progress on Justice40, and did not include independent verification of agency responses. The assessment includes five sections:

Read the full assessment:

Creating a Digital Service for the Planet

Summary

Challenge and Opportunity

The Biden administration—through directives such as Executive Order 14008 on Tackling the Climate Crises at Home and Abroad and President Biden’s Memorandum on Restoring Trust in Government Through Scientific Integrity and Evidence-Based Policymaking, as well as through initiatives such as Justice40 and America the Beautiful (30×30)—has laid the blueprint for a data-driven environmental agenda. 

However, the data to advance this agenda are held and managed by multiple agencies, making them difficult to standardize, share, and use to their full potential. For example, water data are collected by 25 federal entities across 57 data platforms and 462 different data types. Permitting for wetlands, forest fuel treatments, and other important natural-resource management tasks still involves a significant amount of manual data entry, and protocols for handling relevant data vary by region or district. Staff at environmental agencies have privately noted that it can take weeks or months to receive necessary data from colleagues in other agencies, and that they have trouble knowing what data exist at other agencies. Accelerating the success and breadth of environmental initiatives requires digitized, timely, and accessible information for planning and execution of agency strategies.

The state of federal environmental data today echoes the state of public-health data in 2014, when President Obama recognized that the Department of Health and Human Services lacked the technical skill sets and capacity needed to stand up Healthcare.gov. The Obama administration responded by creating the U.S. Digital Service (USDS), which provides federal agencies with on-demand access to the technical expertise they need to design, procure, and deploy technology for the public good. Over the past eight years, USDS has developed a scalable and replicable model of working across government agencies. Projects that USDS has been involved in—like improving federal procurement and hiring processes, deploying healthcare.gov, and modernizing administrative tasks for veterans and immigrants—have saved agencies such as the Department of Veterans Affairs millions of dollars.

But USDS lacks the specialized capacity and skills, experience, and specific directives needed to fully meet the shared digital-infrastructure needs of environmental agencies. The Climate and Economic Justice Screening Tool (CEJST) is an example of how crucial digital-service capacity is for tackling the nation’s environmental priorities, and the need for a DSP. While USDS was instrumental in getting the tool off the ground, several issues with the launch point to a lack of specialized environmental capabilities and expertise within USDS. Many known environmental-justice issues—including wildfire, drought, and flooding—were not reflected in the tool’s first iteration. In addition, the CEJST should have been published in July 2021, but the beta version was not released until February 2022. A DSP familiar with environmental data would have started with a stronger foundation to help anticipate and incorporate such key environmental concerns, and may have been able to deliver the tool on a tighter timeline.

There is hope in this challenge. The fact that many environmental programs across multiple federal agencies have overlapping data and technology needs means that a centralized and dedicated team focused on addressing these needs could significantly and cost-effectively advance the capacities of environmental agencies to:

Plan of Action

To best position federal agencies to meet environmental goals, the Biden administration should establish a “Digital Service for the Planet (DSP).” The DSP would build off the successes of USDS to provide support across three key areas for environmental agencies:

  1. Strategic planning and procurement. Scoping, designing, and procuring technology solutions for programmatic goals. For example, a DSP could help the Fish and Wildlife Service (FWS) accelerate updates to the National Wetlands Inventory, which are currently estimated to take 10 years and cost $20 million dollars.
  2. Technical development. Implementing targeted technical-development activities to achieve mission-related goals in collaboration with agency staff. For example, a DSP could help update the accessibility and utility for many government tools that the public rely heavily on, such as the Army Corps system that tracks mitigation banks (known as the Regulatory In lieu fee and Bank Information Tracking System (RIBITS)).
  3. Cross-agency coordination on digital infrastructure. Facilitating data inventory and sharing, and development of the databases, tools, and technological processes that make cross-agency efforts possible. A DSP could be a helpful partner for facilitating information sharing among agencies that monitor interrelated events, environments, or problems, including droughts, wildfires, and algal blooms. 

The DSP could be established either as a new branch of USDS, or as a new and separate but parallel entity housed within the White House Office of Management and Budget. The former option would enable DSP to leverage the accumulated knowledge and existing structures of USDS. The latter option would enable DSP to be established with a more focused mandate, and would also provide a clear entry point for federal agencies seeking data and technology support specific to environmental issues.

Regardless of the organizational structure selected, DSP should include the essential elements that have helped USDS succeed—per the following recommendations.

Recommendation 1. The DSP should emulate the USDS’s staffing model and position within the Executive Office of the President (EOP).

The USDS hires employees on short-term contracts, with each contract term lasting between six months and four years. This contract-based model enables USDS to attract high-level technologists, product designers, and programmers who are interested in public service, but not necessarily committed to careers in government. USDS’s staffing model also ensures that the Service does not take over core agency capacities, but rather is deployed to design and procure tech solutions that agencies will ultimately operate in-house (i.e., without USDS involvement). USDS’s position within the EOP makes USDS an attractive place for top-level talent to work, gives staff access to high-level government officials, and enables the Service to work flexibly across agencies.

Recommendation 2. Staff the DSP with specialists who have prior experience working on environmental projects.

Working on data and technology issues within environmental contexts requires specialized skill sets and experience. For example, geospatial data and analysis are fundamental to environmental protection and conservation, but this has not been a focal point of USDS hiring. In addition, a DSP staff fluent in the vast and specific terminologies used in environmental fields (such as water management) will be better able to communicate with the many subject-matter experts and data stewards working in environmental agencies. 

Recommendation 3. Place interagency collaboration at the core of the DSP mission.

Most USDS projects focus on a single federal agency, but environmental initiatives—and the data and tech needs they present—almost always involve multiple agencies. Major national challenges, including flood-risk management, harmful algal blooms, and environmental justice, all demand an integrated approach to realize cross-agency benefits. For example, EPA-funded green stormwater infrastructure could reduce flood risk for housing units subsidized by the Department of Housing and Urban Development. DSP should be explicitly tasked with devising approaches for tackling complex data and technology issues that cut across agencies. Fulfilling this mandate may require DSP to bring on additional expertise in core competencies such as data sharing and integration.

Recommendation 4. Actively promote the DSP to relevant federal agencies.

Despite USDS’s eight-year existence, many staff members at agencies involved in environmental initiatives know little about the Service and what it can do for them. To avoid underutilization due to lack of awareness, the DSP’s launch should include an outreach campaign targeted at key agencies, including but not limited to the U.S. Army Corps of Engineers (USACE), the Department of Energy (DOE), the Department of the Interior (DOI), the Environmental Protection Agency (EPA), the National Aeronautics and Space Administration (NASA), the National Oceanic and Atmospheric Administration (NOAA), the U.S. Department of Agriculture, and the U.S. Global Change Research Program (USGCRP).

Conclusion

A new Digital Service for the Planet could accelerate progress on environmental and natural-resource challenges through better use of data and technology. USDS has shown that a relatively small and flexible team can have a profound and lasting effect on how agencies operate, save taxpayer money, and encourage new ways of thinking about long standing problems. However, current capacity at USDS is limited and not specifically tailored to the needs of environmental agencies. From issues ranging from water management to environmental justice, ensuring better use of technology and data will yield benefits for generations to come. This is an important step for the federal government to be a better buyer, partner, and consumer of the data technology and innovations that are necessary to support the country’s conservation, water, and stewardship priorities.

Frequently Asked Questions
How would the DSP differ from the U.S. Digital Service?

The DSP would build on the successful USDS model, but would have two distinguishing characteristics. First, the DSP would employ staff experienced in using or managing environmental data and possessing special expertise in geospatial technologies, remote sensing, and other environmentally relevant tech capabilities. Second, DSP would have an explicit mandate to develop processes for tackling data and technology issues that frequently cut across agencies. For example, the Internet of Water found that at least 25 different federal entities collect water data, while the USGCRP has identified at least 217 examples of earth observation efforts spanning many agencies. USDS is not designed to work with so many agencies at once on a single project—but DSP would be.

Would establishing the DSP prohibit agencies from independently improving their data and tech practices? 

Not in most cases. The DSP would focus on meeting data and technology needs shared by multiple agencies. Agencies would still be free—and encouraged!—to pursue agency-specific data- and tech-improvement projects independently.


Indeed, a hope would be that by showcasing the value of digital services for environmental projects on a cross-agency basis, the DSP would inspire individual agencies to establish their own digital services teams. Precedent for this evolution exists: the USDS provided initial resources to solve digital challenges for healthcare.gov and Department of Veteran Affairs. The Department of Veteran Affairs and Department of Defense have since started their internal digital services teams. However, even with agency-based digital service teams, there will always be a need for a team with a cross-agency view, especially given that so many environmental problems and solutions extend well beyond the borders of a single agency. Digital-service teams at multiple levels can be complementary and would focus on different project scopes and groups of users. For example, agency-specific digital-service teams would be much better positioned to help sustain agency-specific components of an effort established by DSP.

How much would this proposal cost?

We propose the DSP start with a mid-sized team of twenty to thirty full-time equivalent employees (FTEs) and a budget around $8 million. These personnel and financial allocations are in line with allocations for USDS. DSP could be scaled up over time if needed, just as USDS grew from approximately 12 FTEs in fiscal year (FY) 2014 to over 200 FTEs in FY 2022. The long-term target size of the DSP team should be informed by the uptake and success of DSP-led work.

Why would agencies want a DSP? Why would they see it as beneficial?

From our conversations with agency staff, we (the authors) have heard time and again that agencies see immense value in a DSP, and find that two scenarios often inhibit improved adoption of environmental data and technology. The first scenario is that environmental-agency staff see the value in pursuing a technology solution to make their program more effective, but do not have the authority or resources to implement the idea, or are not aware of the avenues available to do so. DSP can help agency staff design and implement modern solutions to realize their vision and coordinate with important stakeholders to facilitate the process.


The second scenario is that environmental-agency staff are trained experts in environmental science, but not in evaluating technology solutions. As such, they are poorly equipped to evaluate the integrity of proposed solutions from external vendors. If they end up trialing a solution that is a poor fit, they may become risk-averse to technology at large. In this scenario, there is tremendous value in having a dedicated team of experts within the government available to help agencies source the appropriate technology or technologies for their programmatic goals.

Reduce, Repurpose, Recharge: Establishing a Collaborative Doctrine of Groundwater Management in the Ogallala Aquifer

Summary

Climate change has resulted in extreme and irregular rain events across the United States. Consequently, farmers in the High Plains region have been increasingly dependent on the Ogallala Aquifer for water supplies. With an estimated value of $35 billion, this aquifer supports one-fifth of the nations’ wheat, corn, cotton, and cattle. The Ogallala once held enough water to fill Chicago’s Sears Tower over 2,000 times. Today, the aquifer has lost 30% of its supply — and it is being recharged at half the rate it is being depleted. The consequence of inaction is 70% aquifer depletion by 2060, which will reduce crop output by 30–40%.

Figure 1. Changes in groundwater levels in the Ogallala Aquifer from predevelopment to 2015. Adapted in the Fourth National Climate Assessment from McGuire et al. (2017).

This $14 billion loss to the High Plains agricultural production may be slowed and eventually reversed by (1) reducing Ogallala use, (2) repurposing existing supplies, and (3) recharging the aquifer. The U.S. Department of Agriculture (USDA), in collaboration with the Department of the Interior (DOI) and the Federal Emergency Management Agency (FEMA), should accordingly create the Reduce, Repurpose, Recharge Initiative (RRRI), a voluntary program designed to keep farmers engaged in groundwater conservation. This multi-state program will provide financial incentives to participating farmers in exchange for pledges to limit groundwater withdrawal and participate in training that will equip them with knowledge needed to fulfill those pledges. The RRRI will also make expert advisors available to consult with farmers on policies and funding opportunities related to groundwater conservation. Finally, this program will connect farmers across state lines, allowing them to learn from each other and work together on sustainable management of the Ogallala. The program should be funded through the various water-sustainability budgets of the DOI and USDA, as well as through FEMA’s Building Resilient Infrastructure and Communities grant program.

Challenge and Opportunity

Climate-change-induced droughts have increased the nation’s dependence on groundwater as a source for agriculture, industry, and domestic use. Excessive groundwater pumping has led to land subsidence and deterioration of water quality, increasing water-use cost and jeopardizing crop yield. The problem is especially acute in the Ogallala Aquifer of the High Plains region. The aquifer underlies eight states of the nation’s breadbasket — including Nebraska, Kansas, and Texas — and spans 175,000 square miles. Dependence on the Ogallala has depleted its supply by 30% to date, as shown in Figure 1. 90% of water withdrawn from the Ogallala is used for agricultural irrigation.

Strategic plans for the USDA and DOI make it clear that drought preparedness and water conservation/sustainability are national priorities. Multiple federal efforts exist to advance these priorities. Publicly accessible platforms hosting and providing groundwater data exist at the United States Geological Survey (USGS), the National Institute of Food and Agriculture (NIFA), and the cross-agency National Integrated Drought Information System (NIDIS) partnership. The 2018 Farm Bill strengthened technical- and financial-assistance programs to help individual farms implement water-conservation technology; the bill also created an incentive program for agriculture-to-wetland conversion. From 2011–2018, the USDA’s Natural Resources Conservation Service (NRCS) ran the Ogallala Aquifer Initiative (OAI) to “support targeted, local efforts to conserve the availability of water, both its quantity and quality, in each of the States” covering the Ogallala. The OAI was successful in meeting its water-conservation goals. Recent surveys found that 93% of agricultural producers in the High Plains region believe that water conservation is important.

These past and ongoing initiatives demonstrate that federal will and stakeholder buy-in for aquifer conservation and restoration are there. The key need is for a program that provides farmers the incentives and technical assistance needed to minimize groundwater reliance, ending the tragedy of the commons in the Ogallala once and for all.

Plan of Action

USDA, DOI, and FEMA should launch a joint program designed to embed the three pillars of groundwater conservation — Reduce, Repurpose, and Recharge — into the practices of farmers in Ogallala states. The RRRI will provide a financial incentive to farmers in exchange for farmer commitments to:

  1. Achieve specified water-conservation targets.
  2. Participate in training opportunities and workshops teaching best practices for water conservation and aquifer recharge.

To succeed, the RRRI will require enthusiastic, voluntary participation from farmers across the High Plains region. Participation should be voluntary because studies have shown that voluntary programs are significantly more effective than mandates in achieving water-conservation goals. In a comparative case study about implementing

In a comparative case study about implementing a voluntary versus mandated water restriction, farmers under the voluntary restriction conserved more water relative to the mandatory regulation. A survey of these farmers attributed the group-education component of the voluntary program as the driving force for their restriction. Another survey similarly found that farmers’ altruistic views of water conservation led to longer-lasting participation in water-conservation activities. A comprehensive review of the outcomes of different water policies found that educational programs about water conservation were more effective in water use reduction and improving attitudes towards water conservation relative to mandatory water use restrictions.

To encourage voluntary participation, farmers who enroll in the RRRI would receive a financial incentive. The exact nature of the incentive would need to be determined by the implementing agencies, but could include preferential price setting, preferential market placement, or subsidies based on crop type. In exchange, farmers would agree to an initial water-use assessment performed by field experts (either employees or contractors of USDA or DOI). An appointed advisor (again, either employees or contractors of USDA or DOI) would then work with each farmer to establish long-term (5-year) water-conservation targets based on the assessment results. Each participating farmer would meet quarterly with their advisor to review their water-conservation plan, assess progress towards targets, make mutually agreeable target adjustments, and discuss challenges and solutions. Advisors would also be available in between quarterly meetings for interim questions and concerns.

Farmers who enroll in the RRRI would also commit to attending group trainings and workshops designed to help them identify and implement best water-conservation practices. These learning opportunities would be led by experts sourced from existing agricultural committees (e.g., NRCS Conservation Planners and Technical Service Providers, State Technical Committees, etc.) and water-conservation groups (e.g., Ogallala Water Coordinated Agriculture Project, Groundwater Protection Council, etc.). The group-education curriculum would cover the three tenets of groundwater conservation: reduce, repurpose, and recharge. Table 1 provides a brief description of each tenet, along with examples of aligned activities and potential sources of funding for those activities. The curriculum would teach farmers how each tenet contributes to groundwater conservation, existing and emerging technologies and practices that farmers can implement to achieve each tenet, and financial vehicles available to fund implementation. An added benefit of the group education will be the establishment of a community of farmers across the Ogallala states in which ideas and experiences can be shared.

TenetDefinitionExample activitiesPotential funding source(s)
ReduceMinimizing water needs for existing systemsMore efficient irrigation NRCS’s Agricultural Management Assistance and Conservation Innovation Grants
RepurposeMove away from water-intensive practicesSwitch to less water-intensive cropsNRCS’s Regional Conservation Partnership Program andConservation Stewardship Program
RechargeReplenish groundwater source (aquifer)Capture excess stormwater; convert agricultural land to wetlands FEMA’s Building Resilient Infrastructure and Communities Grant; NRCS’s Agricultural Conservation Easement Program
Table 1. Definition, example activities, and potential funding sources for each groundwater-conservation tenet.

The RRRI should be established as a multi-agency collaboration. Each involved agency (USDA, DOI, and FEMA) can provide unique expertise. USDA can leverage its research arm, NIFA, to produce up-to-date technology recommendations and scientific assessments. USDA’s NRCS can provide the underlying technical and financial support for realizing the RRRI tenets. DOI can rely on USGS’s existing groundwater database and the NIDIS’s affiliated expert community of data scientists to support the granular, up-to-date groundwater measurements needed to assess water-conservation progress. DOI’s Bureau of Land Management (BLM) can ensure the RRRI tenets are enacted (in parallel with implementation on privately owned farmland) across public lands in the High Plains region. Finally, FEMA can collaborate with NIDIS and with USDA’s Risk Management Agency (RMA) to formally assess risks of drought and Ogallala depletion — assessments that can be used to make the case for the RRRI to farmers, funders, and policymakers. 

Early actions needed to launch the RRRI include:

Conclusion

Climate-change-induced droughts have increased farmer dependence on groundwater, resulting in a 30% depletion of the Ogallala Aquifer to date. Under current management practices, depletion of the Ogallala will reach 70% by 2060. We can solve the problem. The technology, technical expertise, programmatic and data infrastructure, and financial support for groundwater conservation exist. The key need is to directly connect farmers with — and motivate them to use — these resources. A joint USDA/DOI/FEMA program founded in the “Reduce, Repurpose, Recharge” tenets of water conservation can do just that for farmers across the High Plains region. By coupling financial incentives with tailored water-conservation targets, technical expertise, and group educational opportunities, the RRRI will meaningfully advance the long-term security of the critically important Ogallala—and the farmers whose livelihoods depend on it.

Frequently Asked Questions
What is the estimated cost of this program?

Based on the budget for the Ogallala Aquifer Initiative, the RRRI would require $25 million per year for 10-20 years to support the program’s staff and cover travel costs. This funding can be drawn from water-sustainability discretionary funds already allocated at USDA and DOI as well as FEMA’s Building Resilient Infrastructure and Communities grant program.

What existing technologies can promote sustainable groundwater management?

Publications from the Ogallala Water Coordinated Agriculture Project cite numerous examples of existing technologies that can promote sustainable groundwater management, including irrigating with recycled water (i.e., direct non-potable reuse) and shifting to dryland irrigation.

How does the hydrology of the Ogallala region lend itself to aquifer recharge?

The sandy soils of the High Plains are ideal for managed aquifer recharge as they allow for fast infiltration.

Why focus on the Ogallala Aquifer when groundwater depletion is an issue across the US?

With no existing federal regulation on groundwater use, the country needs a pilot program to demonstrate the effectiveness of an interstate groundwater use policy to create precedent for future policymaking and begin to optimize water use policies at such a large scale. The Ogallala Aquifer is the largest and most productive aquifer in the world and conserving the agriculture it supports is required for a sustainable future.

Why won’t the federal government just put a limit on groundwater pumping?

While the federal government has regulations in place dictating water quality through the Environmental Protection Agency’s Clean Water Act and Safe Drinking Water Act, water-allocation policy is left up to the states. Between the eight states above the Ogallala Aquifer, there are four distinct doctrines that define groundwater law, some in direct conflict with one another. State authority over water resources makes it difficult for the federal government to implement mandatory groundwater conservation measures. Voluntary programs like RRRI are an effective mechanism to reach groundwater conservation goals without infringing on states’ water rights.

Establishing the AYA Research Institute: Increasing Data Capacity and Community Engagement for Environmental-Justice Tools

Summary

Environmental justice (EJ) is a priority issue for the Biden Administration, yet the federal government lacks capacity to collect and maintain data needed to adequately identify and respond to environmental-justice (EJ) issues. EJ tools meant to resolve EJ issues — especially the Environmental Protection Agency (EPA)’s EJSCREEN tool — are gaining national recognition. But knowledge gaps and a dearth of EJ-trained scientists are preventing EJSCREEN from reaching its full potential. To address these issues, the Administration should allocate a portion of the EPA’s Justice40 funding to create the “AYA Research Institute”, a think tank under EPA’s jurisdiction. Derived from the Adinkra symbol, AYA means “resourcefulness and defiance against oppression.” The AYA Research Institute will functionally address EJSCREEN’s limitations as well as increase federal capacity to identify and effectively resolve existing and future EJ issues.

Challenge and Opportunity

Approximately 200,000 people in the United States die every year of pollution-related causes. These deaths are concentrated in underresourced, vulnerable, and/or minority communities. The EPA created the Office of Environmental Justice (OEJ) in 1992 to address systematic disparities in environmental outcomes among different communities. The primary tool that OEJ relies on to consider and address EJ concerns is EJSCREEN. EJSCREEN integrates a variety of environmental and demographic data into a layered map that identifies communities disproportionately impacted by environmental harms. This tool is available for public use and is the primary screening mechanism for many initiatives at state and local levels. Unfortunately, EJSCREEN has three major limitations:

  1. Missing indicators. EJSCREEN omits crucial environmental indicators such as drinking-water quality and indoor air quality. OEJ states that these crucial indicators are not included due to a lack of resources available to collect underlying data at the appropriate quality, spatial range, and resolution. 
  2. Small areas are less accurate. There is considerable uncertainty in EJSCREEN environmental and demographic estimates at the census block group (CBG) level. This is because (i) EJSCREEN’s assessments of environmental indicators can rely on data collected at scales less granular than CBG, and (ii) some of EJSCREEN’s demographic estimates are derived from surveys (as opposed to census data) and are therefore less consistent.
  3. Deficiencies in a single dataset can propagate across EJSCREEN analyses. Environmental indicators and health outcomes are inherently interconnected. This means that subpar data on certain indicators — such as emissions levels, ambient pollutant levels in air, individual exposure, and pollutant toxicity — can compromise the reliability of EJSCREEN results on multiple fronts. 

These limitations must be addressed to unlock the full potential of EJSCREEN as a tool for informing research and policy. More robust, accurate, and comprehensive environmental and demographic data are needed to power EJSCREEN. Community-driven initiatives are a powerful but underutilized way to source such data. Yet limited time, funding, rapport, and knowledge tend to discourage scientists from engaging in community-based research collaborations. In addition, effectively operationalizing data-based EJ initiatives at a national scale requires the involvement of specialists trained at the intersection of EJ and science, technology, engineering, and math (STEM). Unfortunately, relatively poor compensation discourages scientists from pursuing EJ work — and scientists who work on other topics but have interest in EJ can rarely commit the time needed to sustain long-term collaborations with EJ organizations. It is time to augment the federal government’s past and existing EJ work with redoubled investment in community-based data and training.

Plan of Action

EPA should dedicate $20 million of its Justice40 funding to establish the AYA Research Institute: an in-house think tank designed to functionally address EJSCREEN’s limitations as well as increase federal capacity to identify and effectively resolve existing and future EJ issues. The word AYA is the formal name for the Adinkra symbol meaning “resourcefulness and defiance against oppression” — concepts that define the fight for environmental justice.

The Research Institute will comprise three arms. The first arm will increase federal EJ data capacity through an expert advisory group tasked with providing and updating recommendations to inform federal collection and use of EJ data. The advisory group will focus specifically on (i) reviewing and recommending updates to environmental and demographic indicators included in EJSCREEN, and (ii) identifying opportunities for community-based initiatives that could help close key gaps in the data upon which EJSCREEN relies.

The second arm will help grow the pipeline of EJ-focused scientists through a three-year fellowship program supporting doctoral students in applied research projects that exclusively address EJ issues in U.S. municipalities and counties identified as frontline communities. The program will be three years long so that participants are able to conduct much-needed longitudinal studies that are rare in the EJ space. To be eligible, doctoral students will need to (i) demonstrate how their projects will help strengthen EJSCREEN and/or leverage EJSCREEN insights, and (ii) present a clear plan for interacting with and considering recommendations from local EJ grassroots organization(s). Selected students will be matched with grassroots EJ organizations distributed across five U.S. geographic regions (Northeast, Southeast, Midwest, Southwest, and West) for mentorship and implementation support. The fellowship will support participants in achieving their academic goals while also providing them with experience working with community-based data, building community-engagement and science-communication skills, and learning how to scale science policymaking from local to federal systems. As such, the fellowship will help grow the pipeline of STEM talent knowledgeable about and committed to working on EJ issues in the United States.

The third arm will embed EJ expertise into federal decision making by sponsoring a permanent suite of very dominant resident staff, supported by “visitors” (i.e., the doctoral fellows), to produce policy recommendations, studies, surveys, qualitative analyses, and quantitative analyses centered around EJ. This model will rely on the resident staff to maintain strong relationships with federal government and extragovernmental partners and to ensure continuity across projects, while the fellows provide ancillary support as appropriate based on their skills/interest and Institute needs. The fellowship will act as a screening tool for hiring future members of the resident staff.

Taken together, these arms of the AYA Research Institute will help advance Justice40’s goal of improving training and workforce development, as well as the Biden Administration’s goal of better preparing the United States to adapt and respond to the impacts of climate change. The AYA Research Institute can be launched with $10 million: $4 million to establish the fellowship program with an initial cohort of 10 doctoral students (receiving stipends commensurate with typical doctoral stipends at U.S. universities), and $6 million to cover administrative expenses and staff expert salaries. Additional funding will be needed to maintain the Institute if it proves successful after launch. Funding for the Institute could come from Justice40 funds allocated to EPA. Alternatively, EPA’s fiscal year (FY) 2022 budget for science and technology clearly states a goal of prioritizing EJ — funds from this budget could hence be allocated towards the Institute using existing authority. Finally, EPA’s FY 2022 budget for environmental programs and management dedicates approximately $6 million to EJSCREEN — a portion of these funds could be reallocated to the Institute as well.

Conclusion

The Biden-Harris Administration is making unprecedented investments in environmental justice. The AYA Research Institute is designed to be a force multiplier for those investments. Federally sponsored EJ efforts involve multiple programs and management tools that directly rely on the usability and accuracy of EJSCREEN. The AYA Research Institute will increase federal data capacity and help resolve the largest gaps in the data upon which EJSCREEN depends in order to increase the tool’s effectiveness. The Institute will also advance data-driven environmental-justice efforts more broadly by (i) growing the pipeline of EJ-focused researchers experienced in working with data, and (ii) embedding EJ expertise into federal decision making. In sum, the AYA Research Institute will strengthen the federal government’s capacity to strategically and meaningfully advance EJ nationwide. 

Frequently Asked Questions
How does this proposal align with grassroots EJ efforts?

Many grassroots EJ efforts are focused on working with scientists to better collect and use data to understand the scope of environmental injustices. The AYA Research Institute would allocate in-kind support to advance such efforts and would help ensure that data collected through community-based initiatives is used as appropriate to strengthen federal decision-making tools like EJSCREEN.

How does this proposal align with the Climate and Economic Justice Screening Tool (CEJST) recently announced by the Biden administration?

EJSCREEN and CEJST are meant to be used in tandem. As the White House explains, “EJSCREEN and CEJST complement each other — the former provides a tool to screen for potential disproportionate environmental burdens and harms at the community level, while the latter defines and maps disadvantaged communities for the purpose of informing how Federal agencies guide the benefits of certain programs, including through the Justice40 Initiative.” As such, improvements to EJSCREEN will inevitably strengthen deployment of CEJST.

Has a think tank ever been embedded in a federal government agency before?

Yes. Examples include the U.S. Army War College Strategic Studies Institute and the Asian-Pacific Center for Security Studies. Both entities have been successful and serve as primary research facilities.

What criteria would the AYA Research Institute use to evaluate doctoral students who apply to its fellowship program?

To be eligible for the fellowship program, applicants must have completed one year of their doctoral program and be current students in a STEM department. Fellows must propose a research project that would help strengthen EJSCREEN and/or leverage EJSCREEN insights to address a particular EJ issue. Fellows must also clearly demonstrate how they would work with community-based organizations on their proposed projects. Priority would be given to candidates proposing the types of longitudinal studies that are rare but badly needed in the EJ space. To ensure that fellows are well equipped to perform deep community engagement, additional selection criteria for the AYA Research Institute fellowship program could draw from the criteria presented in the rubric for the Harvard Climate Advocacy Fellowship.

What can be done to avoid politicizing the AYA Research Institute, and to ensure the Institute’s longevity across administrations?

A key step will be grounding the Institute in the expertise of salaried, career staff. This will offset potential politicization of research outputs.

What is the existing data the EJSCREEN is using?

EJSCREEN 2.0 is largely using data from the 2020 U.S. Census Bureau’s American Community Survey, as well as many other sources (e.g., the Department of Transportation (DOT) National Transportation Atlas Database, the Community Multiscale Air Quality (CMAQ) modeling system, etc.) The EJSCREEN Technical Document explicates the existing data sources that EJSCREEN relies on.

7. What are the demographic and environmental indicators of interest included in EJSCREEN?

The demographic indicators are: people of color, low income, unemployment rate, linguistic isolation, less than high school education, under age 5 and over age 64. The environmental indicators are: particulate matter 2.5, ozone, diesel particulate matter, air toxics cancer risk, air toxics respiratory hazard index, traffic proximity and volume, lead paint, Superfund proximity, risk management plan facility proximity, hazardous waste proximity, underground storage tanks and leaking UST, and wastewater discharge.

The Next Ten Years of Climate Policy, According to Our Experts

Two weeks ago, the IPCC released their most dire warning yet – that we have just three years to prevent the most catastrophic storms, natural disasters, and droughts human civilization might ever see. We are getting closer and closer to the temperature that scientists have warned us for decades would do irreversible damage to our societies and ecosystems.

But the role of scientists is not just one of town criers, warning us of what will come. Scientists are also activists taking fate into their own hands. Last week, scientists across the world staged sit-ins, held demonstrations, and handcuffed themselves to some of the worst climate offenders to send a bold message: the time for action is now.

As a science policy organization, we seek to bridge the gap between experts and policymakers. We have published dozens of proposals and policy memos outlining bold, perhaps even radical, climate policy ideas that would not only save the world, but will invigorate the U.S. and global economy with it.

Below, our experts, researchers, and staff share some of their thoughts on what the next ten years of climate policy will look like, how scientists can get involved in policymaking, and what they hope to see.

Matt Korda, Senior Research Associate and Project Manager, Nuclear Information Project

Climate change and nuclear weapons have a symbiotic relationship: each threat exacerbates the other. Climate change is setting the stage for conflict between nuclear-armed states, and a recent study suggests that even a regional nuclear war could cause mass global starvation for over a decade. Not to mention the fact that even during peacetime, decades of uranium mining, nuclear testing, and nuclear waste dumping have contaminated some of our planet’s ecosystems beyond repair, displacing entire communities—often communities of color—in the process. 

Given the interconnectedness of both the climate crisis and nuclear weapons, we can’t afford for these two existential issues to be tackled in silos. Progressive climate change policies should include demilitarization and disarmament provisions, and progressive nuclear policies should address the climate and humanitarian impacts of nuclear weapons. With that in mind, nuclear disarmament activists and climate change activists are natural allies in the fight to mitigate global catastrophic risks. 

Ishan Sharma, Fellow and Policy Analyst, Technology & Innovation

The IPCC report is clear that a range of solutions is needed to reduce and remove carbon from the atmosphere, “an essential element of scenarios that limit warming to to 1.5℃ or likely below 2℃ by 2100”. 

Across buildings, transport, energy transformation, infrastructure, and industry sectors, climate solutions technologies abound. The question is how to move the needle on their technology readiness levels. Or, in other words, what will spur their adoption at scale?

Let’s start with creating the right liftoff conditions — climate solutions startups have long described the trouble of gaining access to funding given the large upfront costs of hardtech (when compared to software). As a result, many exciting ideas fail to cross the “Valley of Death” and achieve scaled adoption. 

But what if we employed new ways of de-risking investments into climate solutions startups that incentivized more capital flows? One example: committing to buy a certain technology in advance, or advanced market commitments (AMCs). AMCs were recently used in Operation Warp Speed to de-risk and galvanize Pfizer, Moderna, and other pharma companies’ investments to produce COVID-19 vaccines. Last week, an alliance of big tech companies under the organization Frontier poured $925 million towards carbon removal, including large portions towards AMCs because by  “committing to buy a product early, you can help bring it to market faster”.  

Fortunately, the White House Council on Environmental Quality is challenging each agency to leverage its purchasing power in each  sustainability plan, turning the federal government into a massive source of clean demand. 

There’s also the related problem of scaling this type of innovation across the United States. How do you encourage legacy energy communities to support these transitions into the green economy? One way is to create the right set of economic conditions that incentivizes — and provides for — well paying jobs in new growth sectors. 

Currently, there are only a handful of cities with the “industries and a solid base of human capital [to] keep attracting good employers and offering high wages … ecosystems form in these hot cities, complete with innovation companies, funding sources, highly educated workers and a strong service economy.” Spreading innovation to underleveraged regions is difficult for a number of reasons, including training a willing workforce and securing a viable investment ecosystem for startup liftoff. 

But if done right, it could be another answer to rising inflation, as it would bring new demand to “stone cold” markets and stabilize prices from the bottom up and middle out. 

Fortunately, nascent regional investment efforts like the Economic Development Administration’s Build Back Better Regional Challenge and the Department of Energy’s Energy Program for Innovation Clusters seeks a holistic approach to green development, bringing breakthrough research happening in the labs, training workforces in the locality, empowering startup liftoff, and facilitating spillover economic benefits from climate solutions’ commercialization. Federal assets — from national labs to federally funded R&D at universities to manufacturing innovation institutes — should be brought to bear in supporting this innovation-cluster based approach. 

But part of this clean growth climate solution will require coupling our investments in cutting-edge R&D with our investments in manufacturing, in order to produce and commercialize these technologies at scale in the United States. Decades of underinvestment in manufacturing has sacrificed the art of “learning-by-building” — the substantial, value-add interactions that happen when manufacturers are seated at the table with designers. 

This is the reason why China-based companies own 80% of the solar panel market share, despite photovoltaic cells’ invention at Bell Labs in 1954. After heavily subsidizing the manufacturing of solar panels, China capitalized on the benefits of learning-by-building, perhaps the most important factor explaining the nearly 100% drop in PV cells’ module costs over the last 30 years. 

If America is going to lead on solving climate change with breakthrough R&D, we need to ensure we can produce the technologies at scale — lest we risk losing our competitive and economic advantage. But we also need to think seriously about the additional regulatory structures promoting oil and gas above clean growth expansion. 

One example is geothermal energy, which could have already supplied unlimited clean electricity for a cost of around 3¢/kWh, but progress in the field was stunted by entrepreneurial risks of dealing with permitting requirements at up to 2 years of approval timelines. This is despite the Energy Policy Act of 2005 granting oil and gas companies carveouts to perform the same type of drilling geothermal would need. What if we did the same, today, for geothermal? 

Another example is the well-known fossil fuels subsidies, which have been rather hard to kill. These subsidies are responsible for as much as 68-78% of the average rate of return for fossil fuel endeavors. But what might a targeted decoupling look like for fossil fuels subsidies? Several researchers have investigated this very question, identifying which are the most high-leverage subsidies to kill. At first place is the intangible drilling costs subsidy (IDC), which “increases [the] U.S.-wide average IRR by 11 and 8 percentage points for oil and gas fields, respectively.” I’m not the first to advocate for the removal of this subsidy, it’s been decades in the making, but those before me underscore the crucial need for attention focused on high-leverage subsidies — chipping away bit-by-bit at the unnecessary market advantage oil and gas has received for nearly a century. 

Erica Goldman, Director of Science Policy

My back porch looks out into an urban forest in Silver Spring, MD, actually just a few miles from where Rachel Carson lived and wrote Silent Spring. Over the past few weeks, like all over the DC region, the bare trees behind my house bloomed brilliant pinks and whites and purples, and then went quickly green. Now, the forest gets noisier each day – birds in the morning, frogs and cicadas at night, occasional owls and foxes in intermittent conversation.

Most years, I don’t pay attention to how fast all of this happens. But this year I noticed. I’ve been thinking a lot about thresholds for change, sudden inflection points that make the incremental, exponential. With climate change, these tipping points often signal catastrophe – melting glaciers, intensifying storms, and rising seas. But what if tipping points are the key to positive change as well, and targeted interventions can bring larger than expected returns?

Last week, the California Air Resources Board proposed to raise the sale of new cars that are electric, hydrogen-powered or plug-in hybrids to 100% by 2035. In late 2021, the Federal Sustainability Plan issued the ambitious target for the whole-of-government fleet to be fully electric by the same year. Meanwhile, the Bipartisan Infrastructure Law ​​allocates $21 billion over the next five years for the Office of Clean Energy Demonstration, which could dramatically accelerate the pace of deployment of key low-carbon technologies like clean hydrogen, advanced nuclear energy, and carbon capture and storage (CCS) from industrial facilities and power plants. With these kinds of targeted interventions, strategic subsidies, and technologies ready for lift off, might these positive climate tipping points be within reach?

Scientists think maybe. An article in Nature last week showed that the climate pledges made by nations at the COP26 meeting could in fact ensure that global warming does not exceed 2 ºC before 2100 — but only if backed up by short-term policies. Meanwhile, other researchers are studying how to intentionally  trigger positive tipping points through social, technological and ecological innovations, policy interventions, public investment, private investment, broadcasting public information, and behavioral nudges. 

Taken together, my sense is that positive tipping points give us agency and help “unlock paralysis by complexity,” transcending incrementalism and offering “plausible grounds for hope.” We know what nature is capable of, if we can get smart about how to jumpstart recovery and then get out of the way. 

Regulating Probiotic Use and Improving Veterinary Care to Bolster Honeybee Health

This memo is part of the Day One Project Early Career Science Policy Accelerator, a joint initiative between the Federation of American Scientists & the National Science Policy Network.

Summary

One-third of the food Americans eat comes from honeybee-pollinated crops. Honeybees used for commercial pollination operations are routinely treated with antibiotics as a preventative measure against bacterial infections. Pre- and probiotics are marketed to beekeepers to help restore honeybee gut health and improve overall immune function. However, there is little to no federal oversight of these supplements. Apiculture supplements currently on the market are expensive but often ineffective. This leaves unaware farmers wasting money on “snake oil” products while honeybee colonies remain weakened — threatening not just the U.S. agricultural economy, but also the livelihoods of beekeepers and farmers. At the same time, widespread use of antibiotics in apiculture puts honeybees at high risk of spreading antibiotic resistance.

To address these issues, the Food and Drug Administration (FDA)’s Office of Human and Animal Food Operations and the U.S. Department of Agriculture (USDA)’s National Institute for Food and Agriculture (NIFA) should work together to (1) create an FDA review and approval process for pre- and probiotic apiculture products, (2) design educational programs designed to educate veterinarians on best practices for beekeeping health and husbandry, and (3) offer grants to help farmers and apiculturists access high-quality veterinary care for honeybee colonies.

Challenge and Opportunity

Honeybee pollination services are pivotal to the U.S. agricultural economy. It is estimated that about one-third of the food Americans eat comes from crops pollinated by honeybees. Throughout the past decade, beekeepers have suffered colony losses that make commercial apiculture challenging. These colony losses are caused by complex and interconnected issues including the rise of honeybee diseases such as bacterial infections like American Foulbrood or viral infections linked to pests like the Varroa mite, a general increase in hive pests, habitat fragmentation and nutrition loss, and increased use of pesticides and/or pesticide exposure. 

The substantial threats posed by bacterial and viral diseases to honeybee colonies have driven commercial beekeeping operations to routinely treat their hives with antibiotics (mainly oxytetracycline). Unfortunately, antibiotic treatment can also (i) compromise honeybee health by wiping out beneficial bacteria in the honeybee microbiome, and (ii) promote antibiotic resistance. Routine use of antibiotics in apiculture hence compounds the challenges mentioned above and further compromises the livelihoods of U.S. farmers and the security of U.S. food systems.

In 2017, the FDA responded to antibiotic overuse in apiculture by amending the Veterinary Feed Directive (VFD) section of the Animal Drug Availability Act of 1996 (ADAA). The 2017 amendment required beekeepers to obtain veterinary approval to treat their colonies with antibiotics against certain diseases. While attractive on paper, the implementation of this policy has encountered challenges in practice. Finding a vet who understands the highly complex dynamics of apiculture has been a substantial challenge for commercial beekeepers, especially in rural areas. Improvements to the implementation of the VFD are needed to contain the spread of antibiotic resistance in apiculture.

Relatedly, researchers, beekeepers, and companies alike have all been on the hunt for a solution to restore honeybee health after antibiotic treatment. Pre- and probiotic therapy has recently been proposed as a promising and cost-effective strategy to enhance human and animal health, particularly to restore beneficial gut bacteria following antibiotic treatment. Several companies have developed pre- and probiotic supplements targeted at commercial apiculturists. Two popular supplements are HiveAliveTM and SuperDFM®-HoneyBeeTM. HiveAliveTM is marketed as a prebiotic and is composed of seaweed, thymol, and lemongrass extracts. Although there is some evidence that HiveAliveTM decreases infectious fungal-spore counts and reduces winter honeybee mortality, the value of this supplement as a honeybee prebiotic (i.e., to boost growth or activity of beneficial gut bacteria prior to antibiotic treatment) has not been tested. SuperDMF®-HoneyBeeTM is marketed as a probiotic that can restore the honeybee gut microbiome. But SuperDMF®-HoneyBeeTM is exclusively composed of microbes — isolated from mammals or the environment — that have never been found in honeybees and therefore are probably incapable of colonizing the bee gut. To date, neither HiveAliveTM nor SuperDFM®-HoneyBeeTM has been scientifically shown to protect or restore the honeybee gut microbiome from adverse effects of antibiotic treatment.

A big part of the reason why pre- and probiotic supplements for honeybees (as well as for other agricultural uses) have not been externally validated is that such products are not subject to FDA or USDA regulation. This lack of federal oversight means that beekeepers interested in using such products have only the manufacturer’s word that the products will work as promised. Federal intervention is needed to protect commercial farmers and beekeepers from predatory companies selling expensive “snake oil” products.

Plan of Action

To ensure the long-term sustainability of U.S. apiculture and agriculture, the FDA and USDA should work together on the following three-part strategy to improve the administration of antibiotics in apiculture and to strengthen the regulation of pre- and probiotic supplements marketed to commercial beekeepers. 

Part 1. Educate veterinarians in beekeeping to limit misuse and overuse of antibiotics.

For instance, the USDA’s Office of Pest Management Policy (OPMP) and National Institute for Food and Agriculture (NIFA) could collaborate with the U.S. Honeybee Veterinary Consortium on an annual training program, hosted at the USDA’s Bee Research Laboratory, to educate vets working in agricultural areas on the basics of honeybee disease, prevention, treatment, and post-treatment options. The training could also discuss the latest evidence on the efficacy of pre- and probiotic supplements, ensuring that vets can help beekeepers navigate this emerging marketplace of products. Additionally, for veterinarians who are unable to travel to in-person training, these resources could be made available in an online portal. 

Part 2. Strengthen regulation of pre- and probiotics marketed to beekeepers. 

Currently, the market for pre- and probiotics targeted at beekeepers is a veritable “wild west”: one that allows the marketing and sale of essentially any product as long as the ingredients included are deemed safe per the Official Publication of the Association of American Feed Control Officials and are either (i) approved for addition to animal feed (per part 573 in Title 21 of the Code of Federal Regulations (21 CFR 573)), and/or (ii) generally recognized as safe (GRAS) for the intended use (including those listed in 21 CFR 582 and 584). Notably, the efficacy of marketed pre- and probiotics does not have to be demonstrated. Therefore, in alignment with an FDA guidance document recommending stronger oversight of pre- and probiotics targeted at beekeepers, FDA’s Office of New Animal Drug Evaluation (ONADE) should extend its normal animal drug approval process to include pre- and probiotic supplements marketed to beekeepers.

Part 3: Provide apiculturists with better access to high-quality veterinary care.

USDA could create a new Honeybee Veterinary Services Grant Program (HVSGP) that offers rural farmers and beekeepers funding to obtain vet care for their colonies. This program would be modeled after the American Veterinary Medical Association (AVMA)’s Veterinary Services Grant Program, which provides funding to help rural farmers access high-quality vet care for farm animals. The USDA could also consider launching a parallel Honeybee Veterinary Medicine Loan Repayment Program (HVMLRP; again modeled on an AVMA program), which would help place vets trained in beekeeping husbandry “in high-need rural areas by providing strategic loan repayment help in exchange for service”. Vets participating in this program would agree to provide the following services:

Conclusion

Widespread use of antibiotics in commercial beekeeping is a problem for bees, beekeepers, and the larger ecosystem due to the spread of antibiotic resistance and the negative effects of antibiotic treatment on honeybee health. The federal government can mitigate these adverse effects by improving the knowledge and reach of vets trained in best practices for antibiotic treatment in apiculture, as well as by improving regulation of pre- and probiotic supplements purported to restore honeybee gut microbiomes following antibiotic treatment. These actions will collectively secure the health of honeybees — and the livelihoods of farmers who depend on them — for the long term.

Frequently Asked Questions
Why should pre- and probiotics be regulated in honeybees and not humans?

Pre- and probiotics should be regulated in both humans and animals. Pre- and probiotic supplements marketed for human use, like those marketed for apicultural use, are poorly regulated and rife with misleading, untested, or simply false claims. While this memo focuses on the apicultural sector, there is certainly a broader need for increased federal intervention with respect to the safety and efficacy of pre- and probiotics.

What would incentivize veterinarians to participate in educational programs like those proposed in this memo?

The FDA’s 2017 amendments to the VFD mean that if a beekeeper needs to administer antibiotics to their honeybees, they must obtain a prescription or feed directive from a licensed veterinarian. Therefore, vets have a new professional incentive to better understand the dynamics of beekeeping husbandry.

Why treat the “symptoms” of antibiotic treatment and not the root “cause” of widespread antibiotic use in the first place?

In an ideal world, commercial beekeeping would rely on antibiotics only as a last resort. But the reality is that commercial beehives today — due to factors such as a history of intensive antibiotic use in apiculture and the practice of transporting colonies en masse from place to place — are so susceptible to deadly bacteria that imposing major restrictions on antibiotic use in apiculture would seriously compromise U.S. agricultural productivity and the livelihoods of American farmers. Farmers, researchers, and policymakers should continue to collaborate on strategies for phasing out apicultural antibiotic use in the long term. But in the near term, actions should still be taken to promote best practices for apicultural antibiotic treatment and to better regulate supplements that could help minimize adverse impacts of antibiotic treatment on honeybee health.

How retroactive would new FDA regulations on honeybee pre- and probiotic supplements be? I.e., would these regulations apply to products already on the market?

Yes, these regulations should apply to existing products as well as products developed in the future.

Where will the money come from to support the HBVSGP and new research ventures in pre- and probiotic development for honeybees?

The AVMA’s Veterinary Services Grant Program (VSGP) receives funding annually through Congressional appropriations. This funding was $3.5 million for Fiscal Year 2022 (FY22). The HBVSGP could be launched with a similar amount. HBVSGP funding could come from new Congressional appropriations, and/or from existing USDA programs. For instance, the 2008e Farm Bill designated pollinators and their habitat a priority for the USDA and authorized money for projects that promoted pollinator habitat and health under the Environmental Quality Incentives Program (EQIP). Money could also be earmarked from the USDA National Institute of Food and Agriculture (NIFA), Agriculture and Food Research Initiative – Education and Workforce Development grant program to encourage the research and development of better pre- and probiotic supplements and continuing education programs in honeybee veterinary care.

Eliminating Childhood Lead Poisoning Worldwide

An estimated 815 million children (one in three) around the globe have dangerous levels of lead in their bloodstream, levels high enough to cause irreversible brain damage and impose severe health, economic, and societal consequences. 96% of these children live in low- and middle-income countries (LMICs), where collectively only about $6–10 million from non-governmental organizations is available each year to address the problem. To help eliminate childhood lead poisoning worldwide, the U.S. Federal Government should (1) add blood lead level (BLL) testing to the USAID-led Demographic and Health Survey Program, (2) create a Grand Challenge for Development to end childhood lead poisoning, and (3) push forward a global treaty on lead control.

Challenge and Opportunity

Lead is a potent toxin that causes irreversible harm to children’s brains and vital organs. Elevated body lead levels result in reduced intelligence, lower educational attainment, behavioral disorders, violent crime, reduced lifetime earnings, anemia, kidney disease, and cardiovascular disease. Impacts of lead on cognitive development are estimated to cause nearly $1 trillion of income loss in LMICs annually. Adverse health effects related to lead poisoning account for 1% of the global disease burden, causing 1 million deaths annually and substantial disability.

This enormous burden of lead poisoning in LMICs is preventable. It results from a combination of sources of exposure, some of the most important being:

These sources of lead exposure have been effectively regulated in the United States and other high-income countries, which have seen average blood lead levels in their populations decline dramatically over the last 40 years. To achieve the same success, LMICs will need to prioritize policies such as:

LMICs generally face three major barriers to implementing such policies:

  1. Lack of data on blood lead levels and on the scale and severity of lead poisoning. Most LMICs have no studies measuring blood lead levels. Policymakers are therefore unaware of the extent of the problem and hence unlikely to act in response. 
  2. Lack of data on which sources of lead exposure are the biggest local contributors. Causes of lead poisoning vary spatially, but the vast majority of LMICs have not conducted source-apportionment studies. This makes it difficult to prioritize the most impactful policies. 
  3. Limited access to equipment needed to detect lead in paint, spices, water, other sources, or the environment. Without needed detection capabilities, regulators cannot investigate the lead content of potential sources, nor can they monitor and enforce regulation of known sources. 

These barriers are relatively simple to overcome, and when they are overcome do indeed result in action. As an example, at least 20 LMICs introduced legally binding lead paint regulation after the Global Alliance to Eliminate Lead Paint and its partners helped those countries confirm that lead paint was an important source of lead poisoning. Moreover, addressing childhood lead poisoning is in line with the priorities of the Biden Administration and the U.S. Agency for International Development (USAID). The Administration has already proposed an ambitious $15 billion plan to address childhood lead poisoning in the United States by eliminating lead pipes and service lines. By contributing to global elimination efforts (for only a fraction of what it will cost to solve the problem domestically), the Administration can multiply its impact on reducing childhood lead poisoning. Doing so would also advance USAID’s mission of “advanc[ing] a free, peaceful, and prosperous world”, since a reduction in childhood lead poisoning worldwide would improve health, strengthen economies, and prevent crime and conflict.

Plan of Action

Lead poisoning, from a variety of sources, affects one in three children worldwide. This is an unacceptable situation that demands action. The United States should adopt a three-part roadmap to help LMICs implement and enforce policies needed to achieve global elimination of childhood lead poisoning. 

Recommendation 1. Add blood lead level (BLL) testing to the USAID-led Demographic and Health Survey. 

USAID, through its Demographic and Health Survey (DHS), is in an ideal position to address the first barrier that LMICs face to implementing anti-lead poisoning policies: lack of data and awareness. The DHS collects, analyzes, and disseminates accurate and representative data on health in over 90 countries. Including BLL testing in the DHS would:

As such, USAID should add BLL testing of children into the DHS Biomarker Questionnaire for all host countries. This could be done in DHS revision for Phase 9, beginning in 2023. Including BLL testing in the DHS is also the first step to addressing the second barrier that LMICs face: lack of data on sources of lead exposure. BLL data collected through the DHS would reveal which countries and populations have the greatest lead burdens. These data can be leveraged by researchers, governments, and NGOs to investigate key sources of lead exposure.

BLL testing of children is feasible to carry out in the context of the DHS. It was successfully piloted in 1998 and 2002 via the DHS presence in India and Uzbekistan, but not rolled out further. Testing can be carried out using finger-stick capillary sampling and portable analyzers, so venipuncture and laboratory analysis are not required. Further, such testing can be carried out by health technicians who are already trained in capillary blood testing of children for anemia as part of the DHS. The testing can be conducted while questionnaires are administered, and results and any follow-up actions can be shared with the parent/guardian immediately. Alternatively, laboratory lead tests can be added onto sample analysis if blood draws are already being taken. Costs are low in both cases, estimated at around $10 per test. 

Recommendation 2. Create a Grand Challenge for Development to end childhood lead poisoning.

Childhood lead poisoning in LMICs is dramatically neglected relative to the scale of the problem. Though childhood lead poisoning costs LMICs nearly $1 trillion annually and accounts for 1% of the global disease burden, only about $6–10 million per year is dedicated to addressing the problem. A USAID-led Grand Challenge for Development to end childhood lead poisoning would mobilize governments, companies, and foundations around generating and implementing solutions. In particular, the Challenge should encourage solutions to the second and third barriers presented above: lack of data on sources of lead exposure and limited detection capacity. 

Recommendation 3. Push forward a global treaty on lead control.

A global push is needed to put childhood lead poisoning on the radar of decision-makers across the world and spur implementation and enforcement of policies to address the issue. The Biden Administration should lead an international conference to develop a global treaty on lead control. Such a treaty would set safe standards for lead in a variety of products (building on the Global Alliance to Eliminate Lead Paint’s toolkit for establishing lead-paint laws) and recommend regulatory measures to control sources of lead exposure. The success of the UN’s Partnership for Clean Fuels and Vehicles in bringing about global elimination of leaded gasoline illustrates that international political will to act can indeed be generated around lead pollution. 

Conclusion

By implementing this three-part roadmap the Biden administration and USAID can make a historic and catalytic contribution towards global elimination of lead poisoning. There is true urgency; the problem becomes harder to solve each year as more lead enters the environment where it will remain a source of exposure for decades to come.  Acting now will improve the health, wellbeing and potential of hundreds of millions of children. 

Frequently Asked Questions
How do we know that childhood lead poisoning is a global problem if there is very little data on it?

Though relatively little investigation has been done on childhood lead poisoning in LMICs, the studies that do exist have consistently shown very high levels of lead poisoning. A recent systematic reviewidentified studies of background levels of childhood lead exposure in 34 LMICs. According to the review, “[o]f the 1.3 billion children (aged 0–14 years) living in the 34 LMICs with acceptable data on background blood lead levels in children, approximately 632 million…were estimated to have a level exceeding the CDC [Centers for Disease Control and Prevention] reference value of 5 μg/dL, and 413 million…were estimated to exceed the previous reference value of 10 μg/dL.” Data collected by the Institute of Health Metrics and Evaluation and analyzed in a joint UNICEF/Pure Earth report published in 2020 similarly concluded that dangerously elevated BLLs affect over 800 million children worldwide.

Why are levels of lead poisoning so high in LMICs and what are the sources?

Major sources of lead poisoning in LMICs include paint, spices, cookware, pottery, pipes, cosmetics, toys, unsafe lead-acid battery recycling, unsafe e-waste recycling, and poorly controlled mining and smelting operations. High-income countries like the United States have relatively low levels of lead poisoning due to strong regulations around these sources of lead poisoning. Most high-income countries have, for instance, banned lead in gasoline and paint, set enforceable standards around the lead content of water, and imposed strong regulations around food adulteration. As a result, median BLLs in high-income countries have declined dramatically (in the United States, from 15ug/dL in the 1970s to <1µg/dL today). LMICs generally lack many of these effective controls around lead exposure and therefore have very high levels of childhood lead poisoning.

What policies and interventions need to be in place in LMICs to eliminate childhood lead poisoning?

The most important thing that can be done to tackle the scourge of childhood lead poisoning is to impose source controls that prevent lead from entering the environment or consumer products. Though the relative contributions of different sources to childhood lead poisoning differ by country, effective policies and interventions tend to include:



  • Regulations limiting the lead content of paint available on the market.



  • Regulation of lead-acid battery and e-waste recycling practices.



  • Inclusion of lead parameters in national drinking-water-quality standards.



  • Regulation of the use of lead compounds in other locally important sources, such as spices, ceramics, cookware, toys, and cosmetics. 


To enforce these policies, LMICs need testing capacity sufficient to monitor lead levels in potential exposure sources and in the environment. LMICs also need BLL monitoring to track the impact of policies and interventions. Fully eliminating childhood lead poisoning will ultimately involve abatement: i.e., removing lead already in the environment, such as by taking off lead paint already on walls and by replacing lead pipes. However, these interventions are extremely costly, with much lower impact per dollar than preventing lead from entering the environment in the first place.

Why is global childhood lead poisoning such a neglected issue despite its enormous scale?

An extreme lack of awareness, lack of data, and lack of advocacy around childhood lead poisoning in LMICs has created a vicious cycle of inattention. A large part of the problem is that lead poisoning is invisible. Unlike a disease like malaria, which causes characteristic cyclical fevers that indicate their cause, the effects of lead poisoning are more difficult to trace back.

Deploy a National Network of Air-Pollution and CO2 Sensors in 300 American Cities by 2030

Summary

The Biden-Harris Administration should deploy a national network of low-cost, co-located, real-time greenhouse gas (GHG) and air-pollution emission sensors in 300 American cities by 2030 to help communities address environmental inequities, combat global warming, and improve public health. Urban areas contribute more than 70% of total GHG emissions. Aerosols and other byproducts of fossil-fuel combustion — the major drivers of poor air quality — are emitted in huge quantities alongside those GHGs. A “300 by ‘30” initiative establishing a national network of local, ground-level sensors will provide precise and customized information to drive critical climate and air-quality decisions and benefit neighborhoods, schools, and businesses in communities across the nation. Ground-level dense sensor networks located in community neighborhoods also provide a resource that educators can leverage to engage students on co-created “real-time and actionable science”, helping the next generation see how science and technology can contribute to solving our country’s most challenging issues.

Challenge and Opportunity

U.S. cities contribute 70% of our nation’s GHG emissions and have more concentrated air pollutants that harm neighborhoods and communities unequally. Climate change profoundly impacts human health and wellbeing through drought, wildfire, and extreme-weather events, among numerous other impacts. Microscopic air pollutants, which penetrate the body’s respiratory and circulatory systems, play a significant role in heart disease, stroke, lung cancer, and asthma. These diseases collectively cost Americans $800 billion annually in medical bills and result in more than 100,000 Americans dying prematurely each year. Also, health impacts are experienced more acutely for certain communities. Some racial groups and poorer households, especially those located near highways and industry, face higher exposure to harmful air pollutants than others, deepening health inequities across American society. 

GHG emissions and ground-level air pollution are both negative products of fossil-fuel combustion and are inextricably linked. But our nation lacks a comprehensive approach to measure, monitor, and mitigate these drivers of climate change and air pollution. Furthermore, key indicators of air quality — such as ground-level pollutant measurements — are not typically considered alongside GHG measurements in governmental attempts to regulate emissions. A coordinated and data-driven approach across government is needed to drive policies that are ambitious enough to simultaneously and equitably tackle both the climate crisis and worsening air-quality inequities in the United States.

Technologies that are coming down in cost enable ground-level, real-time, and neighborhood-scale observations of GHG and air-pollutant levels. These data support cost-effective mapping of carbon dioxide (CO2) and air-quality related emissions (such as PM2.5, ozone, CO, and nitrogen oxides) to aid in forecasting local air quality, conducting GHG inventories, detecting pollution hotspots, and assessing the effectiveness of policies designed to reduce air pollution and GHG emissions. The result can be more successful, targeted strategies to reduce climate impacts, improve human health, and ensure environmental equity.

Pilot projects are proving the value of hyper-local GHG and air-quality sensor networks. Multiple universities, philanthropies, and nongovernmental organizations (NGOs) have launched pilot projectsdeploying local, real-time GHG and air-pollutant sensors in cities including Los Angeles, New York City, Houston, TX, Providence, RI, and cities in the San Francisco Bay Area. In the San Francisco Bay Area, for instance, a dense network of 70 sensors enabled researchers to closely investigate how movement patterns changed as a result of the COVID-19 pandemic. Observations from local air-quality sensors could be used to evaluate policies aimed at increasing electric-vehicle deployment, to demonstrate how CO and NOx emissions from vehicles change day to day, and to prove that emissions from heavy-duty trucks disproportionately impact lower-income neighborhoods and neighborhoods of color. The federal government can and should incorporate lessons learned from these pilot projects in designing a national network of air-quality sensors in cities across the country. 

Components of a national air-quality sensor network are in place. On-the-ground sensor measurements provide essential ground-level, high-spatial-density measurements that can be combined with data from satellites and other observing systems to create more accurate climate and air-quality maps and models for regions, states, and the country. Through sophisticated computational models, for instance, weather data from the National Oceanic and Atmospheric Administration (NOAA) are already being combined with existing satellite data and data from ground-level dense sensor networks to help locate sources of GHG emissions and air-pollution in cities throughout the day and across seasons. The Environmental Protection Agency (EPA) is working on improving these measurements and models by encouraging development of standards for low-cost sensor data. Finally, data from pilot projects referenced above is being used on an ad hoc basis to inform policy. Data showing that CO2 emissions from the vehicle fleet are decreasing faster than expected in cities with granular emissions monitoring are that policies designed to reduce GHG emissions are working as or better than intended. Federal leadership is needed to bring the impacts of such insights to scale on larger and even more impactful levels.

A national network of hyper-local GHG and air-quality sensors will contribute to K–12 science curricula. The University of California, Berkeley partnered with the National Aeronautics and Space Administration (NASA) on the GLOBE educational program. The program provides ideas and materials for K–12 activities related to climate education and data literacy that leverage data from dense local air-quality sensor networks. Data from a national air-quality sensor network would expand opportunities for this type of place-based learning, motivating students with projects that incorporate observations occurring on the roof of their schools or nearby in their neighborhoods to investigate the atmosphere, climate, and use of data in scientific analyses.Scaling a national network of local GHG and air-quality sensors to include hundreds of cities will yield major economies of scale. A national air-quality sensor network that includes 300 American cities — essentially, all U.S. cities with populations greater than 100,000 — will drive down sensor costs and drive up sensor quality by growing the relevant market. Scaling up the network will also lower operational costs of merging large datasets, interpreting those data, and communicating insights to the public. This city-federal collaboration would provide validated data needed to prove which national and local policies to improve air quality and reduce emissions work, and to weed out those that don’t. 

Plan of Action

The National Oceanic and Atmospheric Administration (NOAA), in partnership with the Bureau of Economic Analysis, the Centers for Disease Control and Prevention (CDC), the Environmental Protection Agency (EPA), the National Aeronautics and Space Administration (NASA), the National Institute of Standards and Technology (NIST), and the National Science Foundation (NSF) should lead a $100 million “300 by ’30: The American City Urban Air Challenge” to deploy low-cost, real-time, ground-based sensors by the year 2030 in all 300 U.S. cities with populations greater than 100,000 residents.

The initiative could be organized and managed by region through an expanded NOAA Regional Collaboration Network, under the auspices of NOAA’s Office of Oceanic and Atmospheric Research. NOAA is responsible for weather and air-quality forecasting and already manages a large suite of global CO2 and global air-quality-related observations along with local weather observations. In a complementary manner, the “300 by ‘30” sensor network would measure CO2, CO (carbon monoxide), NO (nitric oxide), NO2 (nitrogen dioxide), O3 (ozone), and PM2.5 (particulate matter down to 2.5 microns in size) at the neighborhood scale. “300 by ‘30” network operators would coordinate data integration and management within and across localities and report findings to the public through a uniform portal maintained by the federal government. Overall, NOAA would coordinate sensor deployment, network integration and data management and manage the transition from research to operations. NOAA would also work with NIST and EPA to provide uniform formats for collecting and sharing data.

Though NOAA is the natural agency to lead the “300 by ‘30” initiative, other federal agencies can and should play key supporting roles. NSF can support new approaches to instrument design and major innovations in data and computational science methods for analysis of observations that would transition rapidly to practical deployment. NIST can provide technical expertise and leadership in much-needed standards-setting for GHG measurements. NASA can advance the STEM-education portion of this initiative (see below), showing educators and students how to observe GHGs and air quality in their neighborhoods and how to link ground-level observations to observations made from space. NASA can also work with NOAA to merge high-density ground-level and wide-area space-based datasets. BEA can develop local models to provide the nonpartisan, nonpolitical economic information cities will need to inform urban air-policy decisions triggered by insights from the sensor network. Similarly, the EPA can help guide cities in using climate and air-quality information from the sensor network. The CDC can use network data to better characterize public-health threats related to climate change and air pollution, as well as to coordinate responses with state and local health officials. 

The “300 by ‘30” challenge should be deployed in a phased approach that (i) leverages lessons learned from pilot projects referenced above, and (ii) optimizes cost savings and efficiencies from increasing the number of networked cities. Leveraging its Regional Collaboration Network, NOAA would launch the Challenge in 2023 with an initial cohort of nine cities (one in each of NOAA’s nine regions). The Challenge would expand to 25 cities by 2024, 100 cities by 2027, and all 300 cities by 2030. The Challenge would also be open to participation by states and territories whose largest cities have populations less than 100,000.

The challenge should also build on NASA’s GLOBE program to develop and share K–12 curricula, activities, and learning materials that use data from the sensor network to advance climate education and data literacy and to inspire students to pursue higher education and careers in STEM. NOAA and NSF could provide additional support in promoting observation-based science education in classrooms and museums, illustrating how basic scientific observations of the atmosphere vary by neighborhood and collectively contribute to weather, air-quality, and climate models.

Frequently Asked Questions
Has something like the “300 by ‘30” initiative been tried before?

Recent improvements in sensor technologies are only now enabling the use of dense mesh networks of sensors to precisely pinpoint levels and sources of GHGs and air pollutants in real time and at the neighborhood scale. Pilot projects in the San Francisco Bay Area, Los Angeles, Houston, Providence, and New York City have proven the value of localized networks of air-quality sensors, and have demonstrated how data from these sensors can inform emissions-reductions policies. While individual localities, states, and the EPA are continuing to support pilot projects, there has never been a national effort to deploy networked GHG and air-quality sensors in all of the nation’s largest cities, nor has there been a concerted effort to link data collected from such sensors at scale.

If the proposed sensor networks will be inherently local, then why does the federal government need to get involved?

Although urban areas are responsible for over 70% of national GHG emissions and over 70% of air pollution in urban environments, even cities with existing policy approaches to GHGs and air quality lack the information to rapidly evaluate whether their emissions-reduction policies are effective. Further, COVID-19 has impacted local revenue, strained municipal budgets, and has understandably detracted attention from environmental issues in many localities. Federal involvement is needed to (i) give cities the equipment, data, and support they need to make meaningful progress on emissions of GHGs and air pollutants, (ii) coordinate efforts and facilitate exchange of information and lessons learned across cities, and (iii) provide common standards for data collection and sharing.

Is there precedent for this initiative in other countries?

A pilot project including a 20-device sensor network was led by U.S. scientists and developed for the City of Glasgow, Scotland as a demonstration for the COP26 climate conference. The City of Glasgow is an active partner in efforts to expand sensor networks, and is one model for how scientists and municipalities can work together to develop needed information presented in a useful format.

Where will the sensors come from?

Sensors appropriate for this initiative can be manufactured in the United States. A design for a localized network air-quality sensors the size of a shoe box has been described in freely available literature by researchers at the University of California, Berkeley. Domestic manufacture, installation, and maintenance of sensors needed for a national monitoring network will create stable, well-paying jobs in cities nationwide.

Which organizations are already working in this space?

Leading scientific societies Optica (formerly OSA) and the American Geophysical Union (AGU) are spearheading the effort to provide “actionable science” to local and regional policymakers as part of their Global Environmental Measurement & Monitoring (GEMM) Initiative. Optica and AGU are also exploring opportunities with the United Nations Human Settlements Program (UN-Habitat) and the World Meteorological Organization (WMO) to expand these efforts. GHG- and air-quality-measurement pilot projects referenced above are based on the BEACO2N Network of sensors developed by University of California, Berkeley Professor Ronald Cohen.

What about methane

Procurement as an Instrument of Policy: Novel Approaches to Climate Challenges

Our roundtable of climate and procurement experts discussed the underutilized role of the federal government as a strategic buyer to overcome market failures in scaling innovative solutions to climate change. Speakers included:

Initial Perspectives

The session began with remarks from Thomas Kalil (Schmidt Futures), who provided an introduction into procurement levers, or “demand-pull” mechanisms.

The session then shifted to a discussion from Andrew Mayock (CEQ) about the Bidenadministration’s key priorities in combatting climate change, along with nascent conversations on procurement.

Q&A Panel Discussion

Day One Project Policy Entrepreneur in Residence Jetta Wong (ITIF) moderated the panelist Q&A session, asking the following questions with summarized responses:

If we all agree that demand-pull mechanisms could have an outsized impact on driving climate solutions, what are some of the barriers preventing the widespread adoption of these tools?

Thomas Kalil:

One problem is that we have a shortage of people that know how to design effective demand-pull mechanisms. Then there’s the lack of cultural awareness within agencies over useful examples like Falcon 9 for accomplishing strategic objectives. This has resulted in an underutilization of Other Transactions Agreements — perhaps the broadest laws passed by Congress as it is defined by what it is not (not a grant, not a contract) rather than what it is. Under the America COMPETES Reauthorization Act of 2010, federal agencies can support incentive prizes of up to $50 million, but few agencies are using these authorities for ambitious prizes. Finally, appropriations processes pose another obstacle, as Congress traditionally requires that agencies spend their annual appropriations in two years. But for architecting successful demand-pull mechanisms, agencies may need to disperse these funds over a longer time period. It’s important that Congress consider utilizing “No-year” funding foragency R&D and market-shaping efforts. The private sector has an equally important role. Consider the automotive industry: research by the Boston Consulting Group indicates that if an automobile was made with carbon neutral materials, it would only increase the sticker price by 2%. The federal government has not realized the potential to shift markets in this direction.

Jetta Wong:

During the Obama administration, only one person in the entire agency knew about prizes and challenges; it was her job to teach others. Now the Department of Energy has many people working on these issues, but a lot more opportunities to execute on. We need more people willing to leverage these procurement innovations.

Andrew Mayock:

People, policy, and resources matter most. The CEQ and Administration appreciates that many good civil servants with this know-how have left. We have a people deficit; we need more individuals who possess the right skills to effect the transformation we’re hoping for. In terms of policy, the federal sustainability plan we hope to release soon comments on these issues, and sends a clear signal to markets with goals of carbon-free electricity and emission-free federal vehicles. Finally, on resources, now is an opportune time in Washington with the Build Back Better agenda. It has the ability to turbocharge the federal procurement space into truly solving climate challenges.

Lara Pierpoint:

We also have a structural problem with how the federal government is set up. The Department of Energy is focused on sources of energy, not climate, which has caused an organizational and resource challenge in meeting their climate objectives. The climate challenge is a different problem and requires new experiences. In the previous administration, the Department of Energy actually took a step forward. They had robust conversations with industry players and took a systemic view to thinking about how to unlock new technology pathways. It’s important to acknowledge where industry is coming from, what barriers are in their way, and how we might unlock more pathways for scaling innovations.

Despite these barriers, we know a few agencies have started deploying these tools. How should agencies think about getting this right?

Thomas Kalil:

When it comes to the Other Transaction Authorities, agencies have a large amount of flexibility in terms of how to use it. The authority is there, but agencies either haven’t used it or aren’t using it creatively enough. The goal of these partnerships, in the climate context, needs to be on accelerating the scale of deployment and eliminating the “green premium.” Climate solutions need to be scaled and profitable. The government should be partnering with the consumers of new technologies, not just the producers.

Andrew Mayock:

The “Buy Clean” policy is a critical tool towards making progress on these fronts and agencies should recognize the high-level strategic support for these procurement solutions. We should think more about how to scale up and work with the private sector.

One of the solutions mentioned was for Congress to utilize no-year funding, is there a concern with transparency over this and other demand- pull mechanisms?

Thomas Kalil:

If anything, demand-pull mechanisms require more transparency as the government must get specific about what it wants and when. The sponsor of the prize sets a very clear goal, and the results usually follow from this. Performance-based goals versus “we know it when we see it” is a useful mode of operation. The no-year funding element is important for addressing the appropriations obstacle, so agencies have available pools of funding to disperse at later stages.

Lara Pierpoint:

Milestone-based approaches are a great way to achieve progress with transparency, especially with complex technologies like nuclear power. If you want to get the innovation you need, it’s important to have clarity on regulatory milestones in addition to procurement and other targets. Milestone-based payments offer this support.

If you could challenge the audience to do one thing related to tackling the climate crisis with procurement, what would it be?

Thomas Kalil:

Find technologies that have potential for impact and describe the challenges and opportunities to applying demand-pull mechanisms in those sectors. Get specific. You can read about the different levers on the Day One Project website.

Lara Pierpoint:

For those in the agencies, and outside of government, embrace a broader problem- solving mindset. It’s too easy to fall into climate tech silos. Let’s think about what the end-point is and what tools can help us get there. This will involve an interdisciplinary collaboration across offices and fields, so let’s be open to it.

Andrew Mayock:

We need resources, and if we get them, we’ll figure out the people and policy. Stay tuned for the federal sustainability plan and send us your thoughts!

Follow-up Opportunities

The Day One Project is committed to sourcing diverse ideas from a wide community. We encourage you to stay engaged with us as we cultivate more science and technology policy ideas to set the agenda on industrial policy and beyond.

Resource Allocation Questions To Be Answered

The Day One Project recently conducted a white-boarding session with 20 PPBE experts. The product of this seminar is the following list of broad questions about the financial barriers to the Department of Defense’s efforts to modernize the US military. These questions, and the research necessary to answer them, can serve as a roadmap for the Commission on Planning, Programming, Budgeting, and Execution (PPBE) Reform’s work.

  1. Do PPBE and related resource allocation processes, including the appropriations process, limit the ability of emerging technologies to cross the “valley of death” into operations and contribute to DoD’s inability to compete in time with agile competitors?
  2. Is the DoD’s current planning process able to translate future concepts of operations into the programming guidance necessary to develop future warfighting capabilities, or is it overly constrained by the construct of a weapons system program?
  3. Does the current emphasis on a predictive requirements system hinder the Department’s ability to rapidly adopt emerging technologies and undermine its use of recent procurement reforms?
  4. Is the Department’s reliance on manual data calls, PowerPoint presentations, and PDF spreadsheets hosted on different enterprise systems a hindrance to effective budgetary oversight and digital transparency?
  5. Are year-of-execution reprogramming authorities big enough or flexible enough to allow the Department to take advantage of the dynamics of the emerging technology market?
  6. Are DoD’s programmatic measures of effectiveness and performance structured to value adherence to original predictions over the potential of unforeseen outcomes? Is the DoD measuring the right things?

Pathways to Net-zero Soil Loss by 2050

The current administration should announce its intention to achieve net-zero soil loss by 2050. This target aligns with President Biden’s plan to “mount a historic, whole-of-Government-approach to combating climate change,” would help fulfill the administration’s commitment to achieving a net-zero-emissions economy by 2050, and is key to protecting our nation’s agricultural productivity.

Healthy soil is essential to food production. Less well recognized is the vital role that soil plays in climate modulation. Soil is the largest terrestrial carbon repository on the planet, containing three times the amount of carbon in Earth’s atmosphere. Soil represents a potential sink for 133 billion tons of carbon (equal to 25 years of U.S. fossil-fuel emissions). Using soil to offset emissions generates significant co-benefits. Carbon sequestration in soil nourishes soil ecosystems by improving soil architecture and increasing water-holding capacity. Deeper and more fertile soil also supports biodiversity and enriches natural habitats adjacent to agricultural land.

Over two-thirds of the United States is grassland, forestland, and cropland. Land practices that increase the amount of carbon stored underground present a relatively low-cost means for President Biden’s administration to pursue its goal of net-zero carbon emissions by 2050. But lost soil can no longer serve as a carbon repository. And once lost, soil takes centuries to rebuild. Increasingly extreme climate events and soil-degrading industrial farming practices are combining to rapidly deplete our nation’s strategic soil resources. The United States is losing 10.8 tons of fertile soil per hectare per year: a rate that is at least ten times greater than the rate of soil production. At this rate, many parts of the United States will run out of soil in the next 50 years; some regions already have. For example, in the Piedmont region of the eastern United States, farming practices that were inappropriate for the topography caused topsoil erosion and led to the abandonment of agriculture. The northwestern Palouse region has lost 40–50% of its topsoil, and one-third of the Midwest corn belt has lost all of its topsoil.

Soil loss reduces crop yields, destroys species’ habitats that are critical to food production, and causes high financial losses. Once roughly half of the soil is lost from a field, crop yields and nutrient density suffer. Maintaining a desired level of agricultural output then requires synthetic fertilizers that further compromise soil health, unleashing a feedback loop with widespread impacts on air, land, and water quality — impacts that are often disproportionately concentrated in underserved populations.

Climate change and soil erosion create a dual-threat to food production. As climate change progresses, more extreme weather events like intense flooding in the northeastern United States and prolonged drought in the Southwest make farmland less hospitable to production. Concurrently, soil erosion and degradation release soil carbon as greenhouse gases and make crops more vulnerable to extreme weather by weakening the capabilities of plants to fix carbon and deposit it in the soil. Halting soil erosion could reduce emissions, and building stable stores of soil carbon will reduce atmospheric carbon.

Prioritizing soil health and carbon sequestration as a domestic response to the climate and food-security crises is backed by centuries of pre-industrial agricultural practices. Before European occupation of tribal lands and the introduction of “modern agricultural practices,” Indigenous peoples across North America used soil protective practices to produce food while enhancing the health of larger ecosystems. Some U.S. farmers adhere to principles that guide all good soil stewardship — prevent soil movement and improve soil structure. Practices like no-till farming, cover cropping, application of organic soil amendments, and intercropping with deep-rooted prairie plants are proven to anchor soil and can increase its carbon content. In livestock production, regenerative grazing involves moving animals frequently to naturally fertilize the soil while allowing plants to recover and regrow. If all farms implemented these practices, most soil erosion would halt. The challenge is to equip farmers with the knowledge, financial incentives, and flexibility to use soil-protective techniques.

This document recommends a set of actions that the federal government — working with state and local governments, corporations, research institutions, landowners, and farmers — can take towards achieving net-zero soil loss by 2050. These recommendations are supported by policy priorities outlined in President Biden’s Discretionary Budget Request for Fiscal Year 2022 and the bipartisan infrastructure deal currently under negotiation in Congress. Throughout, we emphasize the importance of (1) prioritizing storage of stable carbon (i.e., carbon that remains in soils for the long term) and (2) addressing environmental injustices associated with soil erosion by engaging a broad group of stakeholders.

Firm commitments to restore degraded land will establish the United States as an international leader in soil health, help avoid the worst impacts of climate change, strengthen food security, advance environmental justice, and inspire other countries to set similar net-zero targets. The health of our planet and its people depend on soil preservation. Our nation can, and should, lead the way.

Plan of Action

Action 1. Become a signatory of “4 per mille,” the international initiative encouraging countries to collectively increase global soil carbon by 0.4 percent per year.

The United States should officially join the international effort, “4 per mille” (4p1000), and commit to increasing stable soil carbon by at least 0.4 percent per year. By signing onto this effort, President Biden would send a powerful message of appreciation for U.S. conservation farmers and signal to the rest of the world that soil and forest management are important strategies for mitigating and adapting to climate change.

Detractors of 4p1000 have raised concerns about its feasibility, measurement, and accountability. These arguments obscure the target’s intent: to motivate a global effort to sequester carbon in soil and avert the worst of anthropogenic climate change. The target gives countries a tangible and common goal to work towards as they identify and implement the soil-carbon sequestration strategies that will work best in their respective domestic environments.

Before COP26, the White House Office of Science and Technology Policy, in partnership with the Secretary of Agriculture and the Biden administration’s climate change leaders (John Kerry and Gina McCarthy), should develop a strategy to accompany the United States’ endorsement of 4p1000 and garner endorsements of the agreement from other nations. A central pillar of this strategy should focus on developing and deploying inexpensive methods to estimate soil carbon. These new tools would help farmers track their net carbon increases and ensure that carbon emissions from soil are not negating their efforts.

This action could be supported by funds allocated to the Department of State for multilateral climate initiatives, Department of Interior funding for ecosystem resilience among all land-management agencies, and USDA’s renewed investment to engage landowners to combat climate change and increase participation in voluntary conservation.

Action 2. Invest in a data repository for agriculture and soil carbon. 

Advances in soil health of agricultural systems, like advances in human health, will depend on the sector’s capacity to aggregate and refine big data. This capacity is needed to develop comprehensive decision-support tools underpinned by hyperlocal data in a publicly accessible and well-maintained database

USDA’s Agricultural Research Service currently supports a data repository through its National Agricultural Library (NAL). The NAL repository houses datasets generated by USDA researchers and other USDA-funded research. Unfortunately, the NAL repository is poorly equipped to handle data originating from additional sources. Nor does the NAL repository support the industry-wide annotation system needed to make data searchable and comparable.

A new repository is needed. The National Library of Medicine (NLM) offers an excellent model in GenBank. By helping researchers compare genes, this open-access bioinformatics tool deepens our understanding of health and accelerates development of medical treatments. GenBank connects to similar databases worldwide, and researchers contribute to and search the databases with associated software. The National Weather Service (NWS) similarly compiles a massive set of weather data that supports research and generates income from business services. Both GenBank and the National Weather Service’s databases have supported an explosion of resources, products, and services, from diagnostic medical tests, precision medicine, and genetic testing to weather apps for phones. These databases also feature budgets an order of magnitude larger than the budget for USDA’s NAL.

A right-sized investment in a broad agricultural research database at the NAL, including data generated with proprietary smart-farm technologies and other public-private collaborations, is the future of modern agriculture and agriculture research. Nationally available, high-quality, and curated agricultural data would seed a wealth of new services and companies in the sector. The database would also support the implementation of reliable, locally tailored, and situationally relevant soil-management practices and decision tools that provide precision health practices for soil.

Specifically, we recommend that USDA take the following steps to establish a broad agricultural data repository:

These steps could be carried out using discretionary funding at USDA earmarked for investments in research and development capacity of farmers. These steps collectively align with the administration’s goal to “support a multi-agency initiative aimed at integrating science-based tools into conservation planning and verifying stable carbon sequestration, greenhouse-gas reduction, wildlife stewardship, and other environmental services at the farm level and on federal lands.”

Action 3. Invest in targeted research to reduce soil erosion and increase carbon sequestration.

General factors contributing to soil loss and mitigation principles are universal. Still, the most effective combination of specific practices for reducing soil erosion and increasing carbon sequestration depends on local soil type, slope, soil condition, land use, and weather. In many farming settings, regenerative practices can increase soil carbon and eliminate soil erosion in as little as one or two growing seasons. But matching best practices to a given location can be complex.

For example, intensive tillage is the most soil-erosive practice in agriculture. Reducing the use of this practice has been an important goal for soil-preservation efforts over the last four decades. Organic farms frequently use intensive tillage because organic certification prohibits the use of genetically engineered plants or herbicides—even though herbicide treatment provides excellent weed control and genetic engineering has made it possible to suppress weeds using herbicides without damage to the engineered crop plant. Reducing soil erosion on organic farms hence requires research into new methods of weed control.

The USDA National Institute of Food and Agriculture (NIFA) and the National Science Foundation (NSF) should jointly fund competitive grants for research into practices that reduce soil erosion, increase the nutrient density of food, and sequester carbon stably. Priority projects of these grants might include:

As with Action 2, these steps could be carried out using discretionary funding at USDA earmarked for investments in farmers’ research and development capacity. These steps collectively align with the administration’s goal to support a multi-agency initiative to integrate science-based tools into conservation planning and verify stable carbon sequestration, greenhouse-gas reduction, wildlife stewardship, and other environmental services at the farm level on federal lands.

Action 4. Develop financial and educational programs that help farmers transition to soil-protective practices.

Soil-protective practices have agronomic and economic benefits. Farmers using continuous no-till methods save several thousand dollars each year due to reduced fuel and labor investments. But economic returns on soil-saving practices can take several years to accrue. Growers are rightly concerned about their financial solvency in the short term should they implement such practices, as well as about yield reductions associated with no-till agriculture in some cases. USDA should (i) provide financial assistance to help producers transition to soil-saving practices and (ii) offer training to help producers realize maximal benefits of soil-protective practices at each phase of the transition.

For instance, USDA’s Farm Service Agency (FSA) could offer loans based on cost-saving projections from reduced need for synthetic inputs and increased potential yield once the transition to soil-protective practices is complete. For example, loans could cover the cost of the first five years of projected lost income per acre. At the end of this term, USDA’s Risk Management Agency (RMA) could offer discounted crop insurance rates because the now-healthier soil would engender a more resilient system less likely to experience catastrophic losses during floods and droughts. Farmers could use savings on insurance costs to repay loans and keep premiums constant once repayment begins.

Participation in the loan program could be contingent on farmers’ capacity to maintain soil-protective practices for at least ten years. During the initial five-year loan period, soil-health specialists affiliated with USDA could provide farmers with training on measuring progress, collecting data, and uploading that data to a centralized database. Outcomes across participating farms could be tracked and iteratively inform best practices during the transition period. After the initial five-year period, farmers could qualify for a five-year loan-forbearance period if they demonstrate continued participation in the program.

USDA could also offer direct payments to farmers participating in soil revitalization. Another Day One Project policy proposal recommends that the USDA offer incentive payments for climate-smart practices that produce ecosystem services if the producer cannot find a buyer through an ecosystem-services market. 

Specifically, we recommend that USDA take the following steps to develop financial and educational programs that help farmers transition to soil-protective practices:

These steps could be supported by discretionary funding at the Department of Treasury earmarked for investments in American communities and small businesses and USDA funds dedicated to growing rural economies. These steps align with President Biden’s commitment to expanding the role of Community Development Financial Institutions (CDFIs), which offer loans to start-ups and small businesses in rural communities and create new markets for reaching a net-zero carbon economy by 2050.

Action 5. Develop circular economy practices for young entrepreneurs supporting soil conservation.

Small businesses have a significant role in post-pandemic recovery by providing jobs and combating the climate crisis through innovation. The path to a net-zero carbon economy by 2050 must include circular economy principles that design waste out of economic cycles, keep products and materials in use, and regenerate natural systems. Additionally, closing education gaps and creating new paths to secure jobs for young people who did not complete high school has transformational effects on economic opportunities, health, and life expectancy.

USDA, the Small Business Administration (SBA), and the Minority Business Development Administration (MBDA) should jointly develop a “Ground Up” program that (i) engages the agriculture industry in identifying circular-economy business opportunities and (ii) engages young people without a high-school education in starting small businesses that conserve, restore, and protect soil and other natural resources. Ground Up would fill gaps created by the uneven and insufficient USDA Extension workforce in underserved and under-resourced communities. Ground Up would also provide more extensive business and entrepreneurship training than is typically possible through Extension programs. By leveraging relationships with industry partners, program participants could be connected to byproducts—or “wasted resources”—they need to start a circular business and access to mentoring and markets required to sell their products and services profitably. For example, a Ground Up enterprise might incorporate grounds from commercial or residential coffee-making operations or municipal waste into commercial compost production. The Participants who complete the Ground Up program would be eligible for nointerest federal business loans, with repayment required once the business was profitable. The federal government could partner with Community Development Financial Institutions (CDFIs) to share the cost of loans and build connections among young entrepreneurs, Extension professionals, and potential partner businesses.

Specifically, we recommend that USDA and the White House take the following steps to develop circular economy practices for young entrepreneurs supporting soil conservation:

These steps could be implemented using discretionary funds within USDA, SBA, and MBDA earmarked to support innovative multi-agency business opportunities for rural and minority entrepreneurs. These steps align with the SBA’s commitments help small businesses combat climate change and invest in underserved entrepreneurs; the USDA’s mandate to grow rural economies and foster innovation in the agricultural sector, as well as USDA’s dedication to increasing and protecting biodiversity through good farm stewardship; and the MBDA’s economic-development grants aimed at addressing long-standing racial inequity for minority-owned firms.

Action 6. Support diversity in the agricultural workforce pipeline.

People of color, including Black, LatinX, and Indigenous people, are underrepresented in agriculture and agricultural sciences. To begin addressing this underrepresentation, the Biden administration should ensure diversity in its proposed Civilian Climate Corps (CCC). The CCC is envisioned as a modern-day equivalent of the Depression-era Civilian Conservation Corps work-relief program. The new iteration focuses on enhancing conservation and climate-smart practices across the country. The new CCC represents a terrific way for the Department of the Interior (DOI) to train a diverse workforce in climate- and soil-smart land-management practices with clear pathways to careers in technical assistance, agribusiness, and academic agricultural research, among others.

The administration can boost diversity in agricultural research by directing the USDA’s Office of Civil Rights and the National Institute of Food and Agriculture (NIFA) to conduct an in-depth assessment of challenges faced by researchers of color in agricultural science and develop discipline-wide plans to address them. The administration can also increase research funding and funding for research infrastructure targeted at underrepresented populations. Students from disadvantaged backgrounds are more likely to choose fields with reliable funding. The relative lack of funding for agricultural sciences, as evidenced by outdated educational infrastructure and shrinking training programs, puts agriculture departments at a stark disadvantage compared to the modern facilities (and reliable post-graduate incomes) of other scientific departments (e.g., biomedicine). The National Science Foundation (NSF) should support research and facilities at Historically Black Colleges and Universities (HBCUs) to demonstrate and communicate programmatic stability and cutting-edge innovation in agriculture.

Specifically, we recommend that the Biden administration take the following steps to support diversity in the agricultural-workforce pipeline: 

These steps could be supported by funding allocated at USDA, NSF, and DOI to increase racial equity, specifically the participation of historically underrepresented people in the Civilian Climate Corps and farming, science, and engineering more broadly.

Action 7. Fund existing and proposed advanced research projects agencies (ARPAs) to invest in soil-saving research.

USDA’s research agencies tend to fund low-risk research that delivers incremental changes in agricultural practices. This essential research provides many strategies for stemming soil loss, but remarkably few farms employ these strategies. The nation needs paradigm-shifting advances that farmers will use. The Advanced Research Projects Agency (ARPA) model can help realize such advances by investing deeply in bold ideas outside of mainstream thinking. Several existing and proposed ARPA programs are well-positioned to invest in soil-saving research.

ARPA-Energy (ARPA-E) in the Department of Energy (DOE) is already funding high-impact agricultural research that protects soil. ARPA-E has invested in one soil-centered project, ROOTS, to develop “root-focused” plant cultivars that could dramatically reduce atmospheric carbon. The agency is also gearing up for a new project on carbon farming. These projects match ARPA-E’s energy-focused mission, which includes reducing greenhouse gases in the atmosphere. However, ARPA-E does not have the mandate to invest in specific agricultural projects that build and protect soil. Two additional ARPA-style entities have been proposed that could do so instead: ARPA-I (infrastructure), included as part of the bipartisan Infrastructure Investment and Jobs Act, and AgARDA, a USDA-based ARPA-style agency authorized by the 2018 Agriculture Improvement Act (Farm Bill). If funded, ARPA-I, AgARDA, or both could invest in groundbreaking research to drive soil protection.

To leverage the ARPA model for transformative advances in soil-saving research, we recommend that the Biden administration:

These steps could be supported by discretionary funds allocated to the DOE and USDA. Cumulatively, the President’s most recent budget request directs $1.1 billion to DOE to support breakthroughs in climate and clean-energy research and solutions. Specifically, mitigating and adapting to the climate crisis involves more than inventing cleaner energy; new technologies that help farmers protect soil and fix carbon into the land will also be essential for correcting extreme imbalances in the global carbon budget.

Action 8. Develop criteria and funding for “Earth Cities.”

People feel helpless and fatigued about climate change at the local level partly because they lack the agency to make positive steps to remove greenhouse gases from the atmosphere. The White House should deepen its relationships with mayors and nonprofit coalition groups of cities—such as C40, U.S. Conference of Mayors, and the National League of Cities— to engage urban communities in combating hazards related to climate change.

Like the Arbor Day Foundation’s “Tree Cities” program that encourages communities to steward their tree resources, a national “Earth Cities” program would recognize cities leading the way on urban soil stewardship and management. Criteria for receiving the “Earth City” designation could include implementation of a centralized municipal composting program, large-scale replanting of public parks and rights-ofway with native grasses and perennials that have soil-health benefits, creative management of excavated soil and rock generated by urban construction, becoming a signatory to the 4p1000 initiative, and observance of World Soil Day on December 5. Taking steps to become an “Earth City” and prioritizing soil management at the municipal level offers communities a way to make a positive difference and experience benefits locally while addressing global climate challenges.

Recent research demonstrates that temperatures can vary as much as 20 degrees across different neighborhoods within the same city. Urban heat islands often overlap with communities of color and low-income households in areas with few trees and large amounts of heat-trapping pavement. In these historically redlined communities, rates of heat-related illness and deaths are also higher than wealthier, whiter, and cooler parts of town. Additionally, meeting green building codes and keeping federally supported housing projects affordable has become increasingly difficult in urban centers. Tending to soil health by reusing excavated soil, planting trees and tall grasses on site, and creating more green spaces can inexpensively mitigate the urban heat-island effect while increasing access to nature in historically under-resourced communities. A partnership between soil experts at USDA, pollution and environmental-hazard experts at EPA, and affordable housing programs at the Department of Housing and Urban Development (HUD) would support cities with funding and implementation and further strengthen program viability by tying federal support to local soil stewardship practices.

Specifically, we recommend that the Biden administration take the following steps to recognize and support cities striving to preserve soil and enhance soil-carbon sequestration:

These steps could be supported through earmarked funds at EPA for the Accelerating Environmental and Economic Justice Initiative, HUD funds to modernize and rehabilitate public housing, infrastructure, and facilities in historically underfunded and marginalized communities; and USDA funds that encourage conservation and increased biodiversity on private land.

Action 9. Plant deep-rooted perennials on median strips to foster carbon-rich soils for multi-benefit surface transportation.

As a part of President Biden’s plan to invest in multi-benefit transportation infrastructures, a policy to populate median strips with deep-rooted prairie perennials presents a means to restore soil carbon and simultaneously sustain essential pollinators in agricultural and other ecosystems. Highway medians are supposed to be at least 50 feet wide for safety, creating a minimum of 6 acres of median per mile of highway. The 47,000 miles of U.S. Interstate and 160,000 miles of other highways amount to nearly 300,000 and 1 million acres, respectively, of median strips in the United States. Each acre could sequester 1.7 tons of carbon per year until the soil’s carrying capacity is reached.

Deep carbon stores of soil in the Midwest resulted from centuries of growth of perennial plants that store most of their carbon in their roots. The crops that replaced the prairies shunt most of their carbon to the harvested aboveground tissues, leaving little in the soil. Corn roots, for example, represent only 1% of the plant biomass by the end of the growing season, whereas the roots of perennials—which can grow to as deep as 15 feet underground—can account for as much as 70% of the plant’s biomass. Between 2009 and 2015, 53 million acres of U.S. land was converted from native vegetation to cropland, leading to a loss of 2% of the soil carbon stored in that land per year. This loss translates to 3.2 gigatons of carbon dioxide released into the atmosphere—equivalent to almost one-half of annual U.S. fossil-fuel emissions.

One way to mitigate soil loss is by planting highway median strips with the native, deep-rooted perennials that simultaneously nourish pollinators, enrich soil, and sequester copious amounts of carbon. The Department of Transportation (DOT) could coordinate a large-scale highway-replanting initiative through the effort proposed in the bipartisan infrastructure bill to rebuild the interstate system. In parallel, federal and local “Adopt-a-Highway” programs could enlist citizens, businesses, and municipalities in seeding median strips with native plants.

Specifically, we recommend that:

The administration could pursue these steps using discretionary funds allocated to the Department of Transportation to support competitive-grant programs for infrastructure. The administration could also leverage part of the $110 billion allocated in the bipartisan Infrastructure Investment and Jobs Act towards infrastructure upgrades, including upgrades focused on climate-change mitigation, resilience, and equity.