Disturbances alter ecosystem carbon dynamics, often by reducing carbon uptake and stocks. We compared the impact of two types of disturbances that represent the most likely future conditions of currently dense ponderosa pine forests of the southwestern United States: (1) high-intensity fire and (2) thinning, designed to reduce fire intensity. High-severity fire had a larger impact on ecosystem carbon uptake and storage than thinning. Total ecosystem carbon was 42% lower at the intensely burned site, 10 years after burning, than at the undisturbed site. Eddy covariance measurements over two years showed that the burned site was a net annual source of carbon to the atmosphere whereas the undisturbed site was a sink. Net primary production (NPP), evapotranspiration (ET), and water use efficiency were lower at the burned site than at the undisturbed site. In contrast, thinning decreased total ecosystem carbon by 18%, and changed the site from a carbon sink to a source in the first posttreatment year. Thinning also decreased ET, reduced the limitation of drought on carbon uptake during summer, and did not change water use efficiency. Both disturbances reduced ecosystem carbon uptake by decreasing gross primary production (55% by burning, 30% by thinning) more than total ecosystem respiration (TER; 33-47% by burning, 18% by thinning), and increased the contribution of soil carbon dioxide efflux to TER. The relationship between TER and temperature was not affected by either disturbance. Efforts to accurately estimate regional carbon budgets should consider impacts on carbon dynamics of both large disturbances, such as high-intensity fire, and the partial disturbance of thinning that is often used to prevent intense burning. Our results show that thinned forests of ponderosa pine in the southwestern United States are a desirable alternative to intensively burned forests to maintain carbon stocks and primary production.

ABSTRACT1. Migratory marine species (MMS) include many of the world's most charismatic organisms such as marine mammals, seabirds, turtles, sharks, and tuna. Many are now among the most threatened due to the diverse range of pressures they encounter during their extensive movements. This paper shows that 21% of MMS are classified as threatened (i.e. categorized as Critically Endangered, Endangered or Vulnerable). Sea turtles are the most threatened group (85%), followed by seabirds (27%), cartilaginous fish (26%), marine mammals (15%) and bony fish (11%). Taken together 48% of MMS are threatened, Near Threatened or Data Deficient.2. As well as being threatened they share in common being wide-ranging animals, travelling through the waters of multiple nations as well as in Areas Beyond National Jurisdiction (ABNJ) during different times of the year. This makes their conservation a challenge, requiring coordinated action by many nations, international organizations, Multilateral Environmental Agreements (MEAs) and other stakeholders if their populations are to recover to healthy levels and be safeguarded into the future.3. Even though they are wide-ranging, long-term studies reveal considerable site fidelity and well-defined habitats for many species and areas. These sites are prime candidates for enhanced management such as via Marine Protect Area (MPA) designations. However, existing management frameworks do not yet contribute sufficiently to MMS conservation, MPA networks need to be expanded to capture key areas, in many cases through the application of new dynamic management techniques such as time area closures.4. Data on the distribution, abundance, behaviours and threats faced by many MMS are now available. These data should be used to inform the design of effective management regimes, such as MPAs, both within and beyond national jurisdictions. MEAs should ensure a full complement of MMS are included within species listings, and encourage further action to safeguard their populations.

Post-fire changes in desert vegetation patterns are known, but the mechanisms are poorly understood. Theory suggests that pulse dynamics of resource availability confer advantages to invasive annual species, and that pulse timing can influence survival and competition among species. Precipitation patterns in the American Southwest are predicted to shift toward a drier climate, potentially altering post-fire resource availability and consequent vegetation dynamics. We quantified post-fire inorganic N dynamics and determined how annual plants respond to soil inorganic nitrogen variability following experimental fires in a Mojave Desert shrub community. Soil inorganic N, soil net N mineralization, and production of annual plants were measured beneath shrubs and in interspaces during 6 months following fire. Soil inorganic N pools in burned plots were up to 1 g m(-2) greater than unburned plots for several weeks and increased under shrubs (0.5-1.0 g m(-2)) more than interspaces (0.1-0.2 g m(-2)). Soil NO(3) (-)-N (nitrate-N) increased more and persisted longer than soil NH(4) (+)-N (ammonium-N). Laboratory incubations simulating low soil moisture conditions, and consistent with field moisture during the study, suggest that soil net ammonification and net nitrification were low and mostly unaffected by shrub canopy or burning. After late season rains, and where soil inorganic N pools were elevated after fire, productivity of the predominant invasive Schismus spp. increased and native annuals declined. Results suggest that increased N availability following wildfire can favor invasive annuals over natives. Whether the short-term success of invasive species following fire will direct long-term species composition changes remains to be seen, yet predicted changes in precipitation variability will likely interact with N cycling to affect invasive annual plant dominance following wildfire.

Aim Why are some species widespread and abundant while others are restricted and rare? Darwin (1859) and others have claimed that some species have become widespread because they are competitively dominant over related, geographically restricted species. An alternative hypothesis is that wide-ranging species are ecological generalists that have been able to opportunistically colonize many kinds of new, disturbed and/or marginal habitats, whereas related narrow-ranging species are ecological specialists that competitively dominate speci®c kinds of relatively stable habitats/resources. We tested these opposing hypotheses using small mammals, for which considerable data on competitive interactions and geographical range sizes are available.Location North America, Europe, Asia, Africa and Australia.Methods We analysed data for forty-three competitive interactions between congeneric species in twenty genera, seven families, and four orders of small mammals. Competitive dominance was identi®ed from both ®eld and laboratory experiments, which were lumped because they showed similar results. ResultsWe found that (1) when all studies were analysed as individual data points, species with smaller geographical ranges tended to be dominant over congeneric species with larger geographical ranges, (2) lumping of interaction pairs sharing one or more of the same species yielded a similar result, (3) lumping all interactions involving the same genus also showed the same result, although it was not statistically signi®cant, probably because of the small sample size resulting from this attempt to remove phylogenetic effects. Examination of the taxonomic relationships of the studied genera revealed no obvious phylogenetic effects on the relationship of competitive dominance with geographical range size. Furthermore, although body-size differences appeared to have played a role in the results observed, they cannot completely explain them.Main conclusions We tentatively conclude that Darwin's hypothesis is falsi®ed for small mammals, while the alternative hypothesis that geographically restricted species are competitively dominant over related widespread species receives some support.