Equations based on empirical relationships of peat physical properties were used to estimate the average long-term apparent rate of carbon accumulation (LORCA) in Finnish mire vegetation regions. The results were generalized to the boreal and subarctic regions. Analyses of 1302 dated peat cores were used to infer carbon accumulation for each mire vegetation region of Finland. The area-weighted LORCA for Finnish undrained mire areas was 18.5 g m 2 yr 1 and the total carbon sink 0.79 Tg yr 1 (1 Tg = 1012g). The total carbon pool of Finnish undrained mires was estimated as 2257 Tg. The aapa-mire region included 80% of the total net accumulation rate of carbon and 85% of the total carbon reservoirs of Finnish undrained mires. LORCA was signi” cantly higher in the raised-bog region, 26.1 g m 2 yr 1, compared with the aapa-mire region, 17.3 g m 2 yr 1, and bogs generally had a higher LORCA 20.8 g m 2 yr 1, than fens 16.9 g m 2 yr 1. The total C sink for boreal and subarctic mires was estimated at 66 Tg yr 1 which is about 31% lower than the previous estimates. The total C pool of all boreal and subarctic mires was estimated at 270–370 Pg (1 Pg = 1015g).

(1) Models of peat accumulation are developed that include constant, linear and quadratic decay of dry mass remaining. Profiles of dry bulk density of 795 peatlands distributed over Finland are used to infer cumulative carbon for each site. These values and basal ages are themselves used to infer rates of growth and decay of the peat. (2) A method, 'function parameter fitting' (FPF), is devised to estimate parameter values in non-linear functions when there are uncertainties in both variables, as there are in cumulative carbon and age. Where the data are highly variable then results with FPF are substantially different from those used hitherto that assume uncertainty in only the dependent variable. (3) For five regions in Finland and in Boreal Canada the inferred rate of addition, p* [M L-2 T-'I, is related to degree-days above zero, and decay, a* [T-'1 is related logarithmically to mean annual temperature. The present day rate of accumulation of carbon in northern peatlands is about 5.6 Tmol yr-' or, as dry mass, 0.07 Gt yr-I. (4) There are difficulties in the interpretation of LARCA (=LORCA = long term average rate of carbon accumulation). Understanding of peatland dynamics may result from the use of intrinsic models allowing decay: it is unlikely to emerge from the exotic models in common use.

No abstract

Stratigraphical analyses based on 1028 dated Holocene peat cores were used to estimate the rate of carbon accumulation in Finnish peatlands. Results were compared with data from other Boreal areas. The basal age of peat columns was found to be the best predictor of carbon accumulation and a significant correlation between depth and age of peat was evident. When normalized for the mean depth of peatlands in Finland ( c. 1.5 m), the average long-term carbon accumulation was 26.1 g C m-2 yr-1. In individual cores the values ranged from 2.8 to 88.6 g (average 22.5 ± 11.6 g) being much higher in bogs than in fens and almost double in southern mires as compared with those in the northern Boreal zone. The modelled rate of actual carbon accumulation is about 2/3 of that above. Boreal mires are actual sinks of carbon for thousands of years ahead in the present climate. The predicted greenhouse warming may shift the present Sphagnum mires northward and the net effect may be to restrain the radiative forcing. Poorly known 3D mass accumulation in mires and unpredictable functioning of northern mire ecosystems to climatic changes (for example via increased mire fires) complicate this interpretation.

The community structure of littoral macroinvertebrates was explored by multivariate analyses in three basins of the large Lake Saimaa system (eastern Fin land). The basins differed in trophic status and degree of human influence. It was hypothesized that the structure of littoral invertebrate communities is influenced by lake trophic status, as is the case in profundal communities. Three littoral habitat types with different substrate (stony, sandy and vegetated shores) were sampled from the shoreline to a depth of 1.5-3 meters. The habitat type was found to be largely deter mined by the slope of the shore and the wind exposure. Each habitat type supported fairly characteristic fauna, and detrended correspondence analysis grouped the inverte brate assemblages by habitat type rather than by basin. Within each habitat type, canonical correspondence analysis indicated that species composition changed along the trophic gradient. In the vegetated littoral zone, the greatest change in community structure occurred within the macrophyte beds, varying from the outer edge of macro phytes to the shoreline. Two alternative or complementary explanations are given for this horizontal gradient. First, a horizontal gradient of abiotic characteristics results in a change of community composition. Second, the macrophyte beds may form a hori zontal transition zone in predation, from invertebrate predation inshore to fi sh preda tors offshore. On the stony and sandy shores, the magnitude of wave action was also important in structuring the communities. As each habitat type harbors characteristic

Community ecology has broadened considerably with the recognition that it is not only at the species‐level data where biological patterns and their determinants should be studied. Rather, also functional and phylogenetic data should be examined, as they may provide important information for both basic ecology and applied fields such as conservation and bioassessment. We thus explored the distance decay of taxonomic, functional, and phylogenetic community compositions along spatial and environmental gradients within a boreal lake metacommunity. We used distance‐based methods (i.e., Mantel test, Mantel correlograms and db‐RDA) to examine different facets (i.e., taxonomic, functional, and phylogenetic) and components (i.e., total, turnover and nestedness‐resultant) in relation to spatial and environmental variables. We found that the species compositions of lake benthic invertebrate communities varied mostly along environmental gradients, but were also weakly related to spatial distances between lakes. We also showed that functional and phylogenetic compositions were solely related to environmental variation across the lakes, but these relationships were generally weak. Our exploration of different facets and components of beta diversity added to the knowledge of lake invertebrate communities by adding functional and phylogenetic views, which has rarely been done in studies of aquatic metacommunities. Such information is also important in valuing lakes for conservation and bioassessment because it is not only at the species‐level data where ecological patterns and underlying mechanisms should be explored.

The Anthropocene presents formidable threats to freshwater ecosystems. Lakes are especially vulnerable and important at the same time. They cover only a small area worldwide but harbour high levels of biodiversity and contribute disproportionately to ecosystem services. Lakes differ with respect to their general type (e.g. land‐locked, drainage, floodplain and large lakes) and position in the landscape (e.g. highland versus lowland lakes), which contribute to the dynamics of these systems. Lakes should be generally viewed as ‘meta‐systems’, whereby biodiversity is strongly affected by species dispersal, and ecosystem dynamics are contributed by the flow of matter and substances among locations in a broader waterscape context. Lake connectivity in the waterscape and position in the landscape determine the degree to which a lake is prone to invasion by non‐native species and accumulation of harmful substances. Highly connected lakes low in the landscape accumulate nutrients and pollutants originating from ecosystems higher in the landscape. The monitoring and restoration of lake biodiversity and ecosystem services should consider the fact that a high degree of dynamism is present at local, regional and global scales. However, local and regional monitoring may be plagued by the unpredictability of ecological phenomena, hindering adaptive management of lakes. Although monitoring data are increasingly becoming available to study responses of lakes to global change, we still lack suitable integration of models for entire waterscapes. Research across disciplinary boundaries is needed to address the challenges that lakes face in the Anthropocene because they may play an increasingly important role in harbouring unique aquatic biota as well as providing ecosystem goods and services in the future.

No abstract