Numerous global initiatives and policy frameworks, including blue carbon strategies, biodiversity targets, and ecosystem restoration commitments, highlight the need to conserve and restore seagrasses, yet losses continue to accelerate across many regions. Building on three decades of “Scientists' Warning” papers, we deliver a focused warning rather than another exhaustive review: we argue that the future of seagrasses depends less on further documenting declines and more on (i) prioritizing the protection and conservation of existing meadows, which remain threatened primarily by local stressors such as declining water quality, anchoring, dredging, aquaculture, fisheries interactions, and altered hydrodynamics, (ii) explicitly treating cumulative and interacting stressors as the analytical unit of management, and (iii) closing four operational gaps: standardized global mapping, transparent literature and evidence synthesis, scalable restoration grounded in biotechnology, genetic and microbial science, and governance frameworks capable of translating blue carbon and biodiversity commitments into effective protection and restoration actions. We synthesize evidence across bioregions to identify common patterns of degradation, propose a practical framework that organizes threats according to their physical, chemical, biological, and climatic origins while explicitly accounting for cumulative effects, and outline a prioritized research and policy agenda. Our central message is operational: halting seagrass loss over the next decade is achievable, but only if conservation and political efforts focus first on safeguarding existing meadows and addressing the dominant local drivers of decline rather than relying primarily on climate mitigation potential.
Although direct vertebrate mortality is one of the major impacts of extreme wildfires, it remains poorly documented. Here, we quantified vertebrate mortality through post-fire carcass surveys following severe wildfire in the Iberá Wetlands (Argentina)—a key restoration landscape—and compared outcomes with the Pantanal (Brazil) to evaluate whether extreme conditions produce consistent mortality patterns across Neotropical wetlands. We conducted post-fire line transect surveys (164.7 km) within the 2425.5 km2 burned area, recording 194 carcasses from 42 taxa. Accounting for carcass body size, we estimated at least 641,290 carcasses in the Iberá Nature Reserve. Carcass density was highest in small reptiles (170 carcasses/km2), heavily driven by semi-aquatic specialists such as Helicops leopardinus, and large mammals (32.8 carcasses/km2). Among large mammals, capybara (Hydrochoerus hydrochaeris) carcasses were equivalent to ~63% of the minimum pre-fire population within burned areas and to ~30% of the population of the Iberá National Park. Comparatively, Iberá showed higher carcass densities of medium- and large-bodied vertebrates than Pantanal (50.08 vs. 24.18 carcasses/km2, respectively), whereas Pantanal fires have been reported to impact small-bodied fauna more strongly. In Iberá, carcass density did not differ significantly among vegetation types or between protection levels, although these patterns should be interpreted in terms of absolute carcass density rather than proportional mortality, given the lack of species-specific pre-fire densities. Massive losses of dominant species (capybara) could trigger cascading ecological effects and hinder restoration efforts, underscoring the need to move beyond fire suppression toward integrated fuel management and adaptive strategies under increasingly extreme fire regimes.
Conservation solutions often lack transferability between contexts, a weakness that stems from a lack of connection between general theories and context-specific research. To address this problem, we propose a methodology for developing middle-range theory in conservation science through the formalised application of an operational conceptual framework. Our approach involves (1) defining a focal conservation problem and system; (2) developing a social-ecological system description using expert knowledge and a standardised grammar and ontology, based on the Coupled Infrastructure Systems framework; (3) identifying key feedbacks and expressing these as a causal loop diagram; (4) translating identified social-ecological system dynamics into a formal model which precisely expresses existing theory by specifying quantitative relationships between variables; and then (5) treating the model as an experimental system that can be explored in silico to advance conceptual understanding and mid-range generalisation. We demonstrate our approach in the context of protected area management, using the example of managing human-wildlife interactions to show how expressing conservation challenges as generalized models can connect scientific enquiry to both more specific theories of practice and more universal theory. Key innovations that facilitated progress included adopting a standard, explicitly social-ecological framework to both describe and connect multiple case studies; focusing on feedbacks; and integrating previous research on social-ecological syndromes and dilemmas. Our approach lays a foundation for the deliberate identification of shared principles and context dependence across locally different instances of similar conservation problems and offers a starting point for the explicit development of mid-range theory and transferable solutions in conservation science.
