A bottom-up approach to supply chain legibility

The voluntary carbon markets are undergoing a difficult correction. Credits that once appeared defensible in project documentation are now struggling to withstand independent examination. Modelled baselines often rely on counterfactual scenarios that cannot be validated. Permanence assumptions depend on ecological stability that field data does not support. Biomass, soil fluxes and disturbance patterns fluctuate more than the protocols anticipated. As scrutiny increases, the measurement architecture behind many credits cannot account for the divergence between expected and observed behaviour. The credibility crisis is not driven by sentiment. It reflects a mismatch between the assumptions used to quantify impact and the physical conditions that govern it.

Supply chain sustainability reporting is developing the same structural weaknesses. Corporate impact tables present precise numbers, yet most inputs originate from broad emission factors, generic life cycle coefficients and unverified supplier data. The World Benchmarking Alliance documented widespread retreat from detailed disclosures in 2024 when companies recognised that many figures could not withstand examination. The 2025 KPMG assurance survey found that only a small proportion of supply chain environmental data is supported by physical verification. A technical note from the Science Based Targets initiative highlighted that many Scope 3 values express a confidence the underlying assumptions cannot justify. Research from MIT and the Journal of Industrial Ecology shows that variation among suppliers within a single region often exceeds the variation between entire product categories. When generalised coefficients replace observation, the resulting numbers lose stability.

Waste fate makes the problem clear. Standard accounting treats waste as a terminal category with a representative disposal factor. In practice, environmental outcomes are dictated by the actual physical and operational conditions the material encounters once it leaves the facility.

For organic waste, methane generation depends on moisture, temperature, oxygen availability, carbon to nitrogen ratios and microbial activity. Anaerobic decomposition in unmanaged piles can produce methane fluxes that differ by roughly a factor of ten under otherwise similar conditions. A wet, oxygen limited pile in warm conditions can generate significant methane, while the same material, dried or aerated, produces negligible flux. Landfilled organics follow different degradation trajectories. Cover materials, compaction, waste density and gas collection systems influence the decomposition environment and therefore the emissions profile. Two regions that appear identical in reported numbers can have completely different atmospheric impacts because the decomposition conditions differ in ways the accounting framework does not capture.

Plastic and composite waste shows similar divergence. Mechanical recycling reduces impact only when contamination is low and local infrastructure can maintain adequate yield. When contamination exceeds roughly ten to fifteen percent, operators commonly divert material to incineration or disposal. Emissions from incineration vary with feedstock composition, combustion efficiency, furnace temperature and the share of energy recovered. Stack measurements show that nitrogen oxides and particulates can differ by factors of two to three across facilities performing the same process. A single disposal coefficient cannot represent this variability.

Operational differences compound the problem. Within a single city, identical waste streams can follow different paths depending on subcontractor practices, sorting capacity and the economics of informal recovery. Mixed plastics logged as recycled may be diverted to energy recovery if sorting lines are overloaded. Textile waste reported as downcycled may be burned in open air when secondary processors face bottlenecks. These divergences often dominate the environmental footprint of a waste stream, yet accounting systems compress them into a single category.

The ongoing correction in the carbon markets shows how fragile environmental information becomes when it depends on assumptions that cannot be reconciled with observation. A 2025 investigation by Carbon Market Watch found that a majority of auditors in one registry were associated with problematic crediting outcomes, reflecting the distance between documentary review and physical evidence. A Royal United Services Institute analysis identified structural vulnerabilities that allowed unverifiable claims to pass through verification channels. Sasaki and colleagues documented cases where offset quality deteriorated when permanence and additionality assumptions were tested against field data. Academic work on forest and soil carbon dynamics shows that ecological processes vary at spatial and temporal scales that exceed the assumptions embedded in many protocols. These findings point to a common mechanism. Generalised averages cannot substitute for measurements when local conditions determine environmental outcome.

Supply chain sustainability reporting faces the same constraints. Local variation in moisture, temperature, energy mix, operator behaviour, equipment performance and waste handling often determines the final footprint of a product. When this variation is collapsed into broad coefficients, the resulting data cannot withstand scrutiny.

A bottom up approach strengthens the system by grounding it in observable mechanisms. This does not require comprehensive monitoring of every process. Most steps contribute little to the final outcome. The essential task is to identify the governing variables at the points where environmental divergence occurs and measure those variables directly. In the waste example, these variables include moisture content, oxygen exposure, temperature, contamination levels and the local handling pathway. These parameters can be measured with simple tools and they explain most of the variation that generic factors obscure. Once these measurements are incorporated, modelled elements can be kept within empirically defensible limits and uncertainty can be presented transparently.

The carbon markets are demonstrating, in real time, the vulnerability of environmental information systems built on assumptions that drift away from physical conditions. Supply chain sustainability reporting is showing the same fracture lines. Systems remain credible only when the information reflects the behaviour of the materials and processes that create the impact.