Summary1. Stream habitat quality assessment complements biological assessment by providing a mechanism for ruling out habitat degradation as a potential stressor and provides reference targets for the physical aspects of stream restoration projects. This study analysed five approaches for predicting habitat conditions based on discriminant function, linear regressions, ordination and nearest neighbour analyses.2. Quantitative physical and chemical habitat and riparian conditions in minimally‐impacted streams in New Hampshire were estimated using United States Environmental Protection Agency's Environmental Monitoring and Assessment Program protocols. Catchment‐scale descriptors were used to predict segment‐scale stream channel and riparian habitat, and the accuracy and precision of the different modelling approaches were compared.3. A new assessment index comparing and summarizing the degree of correspondence between predicted and observed habitat based on Euclidean distance between the standardized habitat factors is described. Higher index scores (i.e. greater Euclidean distance) would suggest a greater deviation in habitat between observed conditions and expected reference conditions. As in most biotic indices, the range in index scores in reference sites would constitute a situation equivalent to reference conditions. This new index avoids the erroneous prediction of multiple, mutually exclusive habitat conditions that have confounded previous habitat assessment approaches.4. Separate linear regression models for each habitat descriptor yielded the most accurate and precise prediction of reference conditions, with a coefficient of variation (CV) between predictions and observations for all reference sites of 0.269. However, for a unified implementation in regions where a classification‐based approach has already been taken for biological assessment, a discriminant analysis approach, that predicted membership in biotic communities and compared the mean habitat features in the biotic communities with the observed habitat features, was similar in prediction accuracy and precision (CV = 0.293).5. The best model had an error of 27% of the mean index value for the reference sites, indicating substantial room for improvement. Additional catchment characteristics not readily available for this analysis, such as average rainfall or winter snow‐pack, surficial geological characteristics or past land‐use history, may improve the precision of the predicted habitat features in the reference streams. Land‐use history in New Hampshire and regional environmental impacts have greatly impacted stream habitat conditions even in streams considered minimally‐impacted today; thus as regional environmental impacts change and riparian forests mature, reference habitat conditions should be re‐evaluated.

Summary Algal assemblages have been included in biological monitoring in a number of national, large‐scale and long‐term biological monitoring and assessment programmes of streams, lakes and wetlands. However, there has been no substantial investigation of alternative enumeration strategies. We compared the sampling efficiency of the common periphyton subsampling strategies of counting 300 cells by strip counts or random fields and measuring 10 cells per taxon for biovolume, with line‐intercept sampling (LIS), using samples collected from four streams differing in canopy cover and discharge. We also simulated the optimum ratio of cells tallied to cells measured for biovolume for all approaches. LIS was 1·6–5·5 times more efficient at estimating algal taxa biovolume than the 300‐cell count methods. LIS performed best relative to conventional 300‐cell count methods in streams with high biovolume, although efficiency gains were still noticeable in low biovolume streams. In addition, LIS tended to detect more taxa than the conventional 300‐cell target biological assessment methods. In simulations, LIS was more efficient than the conventional methods for nearly all combinations of target cell count and cells measured for biovolume. No single optimal ratio of cells counted to cells measured was found. The target cell count that resulted in the greatest efficiency was higher in assemblages with large variation in cell density. Likewise, assemblages with high variance in taxa biovolume were most efficiently sampled by measuring more cells. Synthesis and applications. An increase in sampling efficiency as a result of greater adoption of LIS sampling may result in more accurate estimation of the relationships between periphyton community composition and environmental factors, and more sensitive detection of impacts using stream periphyton. Additionally, LIS may improve estimates in other ecological research fields employing microscopic counting chambers.

Effects of the non-indigenous shrub Rhamnus frangula L. (glossy buckthorn) on tree recruitment, herb cover, forest floor plant species richness, and R. frangula recruitment were tested in two southeastern New Hampshire Pinus forests using a randomized complete-block field experiment. The treatment, applied in January of 2000, was the presence of well-established R. frangula populations with three levels: R. frangula absent prior to experiment ("uninvaded"), > 90% R. frangula cover ("Rhamnus present"), and removal of > 90% R. frangula cover ("Rhamnus removed"). After 2 years of measurements, Rhamnus present had significantly lower first-year native tree seedling density than Rhamnus removed and uninvaded plots (0.11, 0.40, and 0.40 seedlings/m(2) respectively). First-year native tree seedling density in the Rhamnus removed and uninvaded treatments were similar. Neither percent herb cover nor plant species richness were significantly affected by the removal of R. frangula in the two years following treatment. We believe these results indicate that the presence of dense R. frangula inhibits the establishment of tree seedlings. Rhamnus removed plots sampled one year after removal had five-fold greater first-year R. frangula seedling density than the other treatments. However, after two years first-year R. frangula seedling density was similarly low in all treatments (< 0.5 R. frangula seedlings/m(2)). Control efforts for R. frangula may need to focus on conspecific seedling emergence for at least two years following initial control.

We tested the ability of the simple reaction–diffusion spread model to quantify the rate of spread of Rhamnus frangula L., a non-indigenous shrub, at the scale of a single forest stand in southeastern New Hampshire. Cut stems of R. frangula were collected in a Cartesian grid pattern within the invaded stand and aged. A contour map of the invasion pattern was produced using kriging. The empirical invasion rate was calculated as the linear distance per year since initial invasion. The intrinsic rate of increase and diffusion coefficient were estimated in the field and the model prediction calculated. Observed mean rate of linear expansion was measured to be 6.7m per year with a standard deviation of 1.24m per year, while the predicted linear asymptotic rate of spread was calculated be 6.3m per year. Thus, the simple reaction–diffusion model yielded a very accurate prediction of actual spread rate. There may have been a lag-phase in the invasion at this site as the spread rate was found to be slower in the early stages of invasion compared to later stages. At the scale of single stands, the simple reaction–diffusion model may be accurate enough to predict the spread of new invaders without recourse to complex life-history spread rate models. Additionally, if lag-phases in spread rate occur with regular frequency, then attempts to retrospectively predict which species may be invasive will be hampered by incorrectly classifying a future invader as currently non-invasive.

Disturbance and land transformation seem to enhance the ability of exotic species to invade, possibly due to increased resource levels and greater immigration potential from surrounding disturbed areas. To investigate the role disturbance may play in enhancing exotic species invasion, we used retrospective analysis to discriminate the characters associated with naturalized and non-naturalized exotic woody plants in New Hampshire utilizing the biological information contained in the US Department of Agriculture (USDA) PLANTS database and selected other sources. Exotic plants were partitioned into two groups: those known to have self-sustaining populations and those that did not display self-sustaining populations. To predict membership into these groups, we used stepwise logistic regression. Variables were screened for correlation coefficients higher than 0.5. Model development was constrained using Akaike’s information criteria with the correction factor for sample size. The 11th model step containing 11 characters was selected and it differed significantly from the constant-only model (χ2=66.383, P<0.001). Exotic species were correctly classified into the known non-naturalized or naturalized categories in 90% of the cases. The resulting model could be used to reliably screen novel woody plant introductions. Additionally, we provide a framework for preliminarily assessing the importance of different factors in encouraging woody plant invasion in the northeastern US.

This paper investigated the potential for the exotic shrub Rhamnus frangula L. (glossy buckthorn) to alter native plant community composition in southeastern New Hampshire. Stratified random sampling was performed with 2 m x 2 m plots randomly located in 5 m intervals along three 50 m transects in four even-aged Pinus-mixed hardwood forests, three of which were managed stands. The associations between R. frangula and the measured species abundances and environmental variables were investigated using linear, least-squares multiple regression and Non-metric Multidimensional Scaling Ordination. Plot basal area of R. frangula was inversely related to woody seedling density (p < 0.001), herb cover (p < 0.05), and species richness (p < 0.01). The relative contribution of R. frangula to explaining variance in seedling density was greater than canopy openness, soil pH, soil clay, or soil sand. Abundance of R. frangula was a statistically significant predictor (p < 0.05) of individual herb species abundances for all study sites. This evidence supports the hypothesis that R. frangula causes a decline in seedling density and alters native ground level plant species abundances. Furthermore, the patterns agree with the suppression of ground level plant species abundances by R. frangula found in removal experiments.