Accounting for hidden wild inputs in conservation substitutes

Artificial propagation, wildlife farming, ranching, cultivation, and aquaculture support commercial markets and livelihoods, and their products may be promoted as substitutes for wild harvest. However, production labels for these products cannot be used to determine conservation benefit because output may depend on wild inputs. Thus, we developed hidden wild-input (HWI) accounting as a bounded assessment framework. HWIs are wild-derived biological inputs that are important for producing substitutes but also potentially for conservation, and they are underrepresented in labels, statistics, and assessments. Through conceptual analysis and comparison, we identified establishment, maintenance, and scaling production inputs and conducted a counterfactual evaluation to determine the avoided harvest of wild targets, source-population burdens, and other ecological effects, such as wild-input intensity, wild-origin proportion, supplementation frequency, and elasticity. We examined four examples of wild-input use to illustrate the framework: Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) source code Y exposed the gap between production-origin categories and upstream accounting; crocodile ranching showed recurring wild inputs can coexist with conservation gains under regulated management; capture-based European eel (Anguilla anguilla) aquaculture demonstrated production-extraction coupling; and the artificial production of Chinese caterpillar fungus (Ophiocordyceps sinensis) revealed potential commercial dependence when recurring inputs are not reported. Together, the examples showed that source categories cannot be used to quantify upstream burdens, recurring inputs do not indicate conservation failure, and input-output coefficients demonstrate coupling but not net effects. Conservation benefit should be assessed against a credible no-expansion counterfactual by asking whether additional output reduces target harvest, added wild-input burdens remain sustainable, and other ecological effects reverse the gain. Complete biological independence is neither necessary nor sufficient. When input, demographic, and market-substitution data are unavailable, uncertainty should be reported rather than replaced by assumptions based on production labels.