The Acadian Forest of eastern Canada has shown a pervasive signal of forest degradation since 1985 (Fig. 1). Since 1985, >3 million ha have been clear-cut (Fig. 1d), with most of this area now occupied by either tree plantations and thinnings (Fig. 1c–e), which are dominated by single tree species20, or a mix of early successional tree species (Fig. 1a,d,e). Despite some ingrowth due to succession, old forest has declined by 39% during the period observed (Extended Data Fig. 1a,b; Supplementary Methods). The pattern of extensive harvest of old forest, followed by rapid regeneration of young forest appears to be common across many forest regions of North America (for example, central Canada, southeastern United States, western United States; Fig. 1b) (ref. 10) and can be considered ‘forest degradation’ in that these practices simplify forest structure, reduce tree species diversity and truncate old-forest age classes6. During the same 35-year time period, forest cover remained relatively stable, increasing by a net 6.5% (Fig. 3a, red line)21.
a, Habitat trends (1985–2020) for the seven bird species exhibiting the greatest population declines according to SDMs; all of these species are old forest associated. During the same time interval, total forest cover did not decline (red line, right axis), indicating that habitat loss is a function of forest degradation rather than loss. b,c, Predicted habitat loss (pink) and gain (blue) between 1985 and 2020 for two example species: Blackburnian warbler (33% habitat loss; b) and golden-crowned kinglet (38% habitat loss; c). Habitat loss was quantified using SDMs with Landsat data as independent variables strongly predicted population trends for forest bird species.
Overall, SDMs using Landsat reflectance bands as predictors performed well for most forest bird species when tested on 50% spatially discrete hold-out data (Extended Data Fig. 2; (bar x) area under the curve (AUC) = 0.73 [range: 0.60–0.90]). SDMs therefore provided reliable estimates of habitat suitability and distribution for most of the 54 species. Species with lower model-prediction success tended to be associated with fine-scale forest structure (for example, individual tall trees, standing and fallen dead wood) which are poorly captured by satellite imagery.
We back cast SDMs to quantify habitat change for all 54 forest bird species from 1985 to 2020. Habitat declines occurred for 66% of species during 1985–2020; 93% of species exhibited habitat reductions over the past decade (Fig. 3 and Extended Data Fig. 3). Species showing the greatest decreases in habitat were golden-crowned kinglet (Regulus satrapa; −38%) and Blackburnian warbler (Setophaga fusca; −33%; Supplementary Video 1) with seven species showing habitat declines >25% (Fig. 3). Most species with strongly declining habitat are associated with old forests22 (Fig. 4a,b), which is consistent with forest degradation due to harvesting of old forest. Indeed, clear-cut harvest alone was strongly associated with habitat declines for all old forest-associated species (Fig. 4c and Extended Data Figs. 4 and 5). Forest succession into old age classes was apparently insufficient to compensate for this rate of loss. Fifteen species exhibited habitat increases, but most (14 out of 15) of these tend to be associated with young or immature forests (Fig. 4a,b).
a, The relationship between habitat change, estimated from SDMs and independently derived population change estimates from the BBS for the Acadian forest. Bird species of mature (old) forests (M; dark green dots) exhibit the greatest habitat loss; this is generally reflected in strongly negative population trends. Bird species associated with regenerating forest (R; red dots) tend to have stable or increasing habitat but still show BBS population declines. b, The relationship between quantitatively derived estimates of mature-forest association and habitat change from 1985 to 2020. Mature forest-associated species tend to be losing the most habitat in relation to immature- (I; light-green dots) and regeneration-associated species. Successional stage categorizations (R, I, M) are from Birds of the World (BOW). The regression line was fit using a hierarchical Bayesian model (Supplementary Methods) and grey shading in b shows 95% credible intervals. Only a subset of species is shown in b (those with quantitative data for mature-forest associations; Supplementary Methods). c, The relationship between area clear-cut occurring from 1985 to 2020 in each species’ habitat within a 200 m-diameter buffer surrounding BBS routes (N = 90) and habitat loss (1985–2020) at the same scale for six mature forest-associated species. Black lines are regression lines and grey bands are 95% confidence intervals (regression estimates in Supplementary Table 3). As expected, clear-cutting is strongly associated with habitat loss, which indicates that ingrowth of new habitat is rarely compensated for by habitat loss (a signature of forest degradation via old age–class truncation).
Several lines of evidence support forest management as the primary driver of forest degradation rather than alternative mechanisms (for example, climate-mediated forest decline, natural disturbance, permanent deforestation). First, our SDMs did not include climate data so the reflectance changes from satellite imagery used in our SDMs were predominantly due to forest compositional changes. Although climate (for example, inter-annual differences in precipitation) can cause subtle differences in reflectance (leaf colour) over time, most changes in the magnitude of reflectance are due to changes in forest composition or cover rather than effects of climate23 (Supplementary Figs. 1 and 2). Indeed, if the observed habitat declines were due to climate effects or natural disturbance, we would expect to see parallel habitat declines in protected areas, which we did not (Extended Data Figs. 6 and 7). Second, species exhibiting the greatest declines in habitat are those most strongly associated with old forest (Fig. 4a,b), which is the primary target of timber harvest. Indeed, the amount of area clear-cut was strongly associated with habitat loss for old forest-associated bird species (Fig. 4c and Extended Data Figs. 4 and 5). Third, deforestation (defined as permanent conversion to another land-cover type)24 was not a primary driver of habitat loss in our region; deforestation contributed <2% of total habitat loss for all 54 species (Supplementary Information and Supplementary Table 2). We acknowledge that due to the complex nature of changes in forest structure and composition through forest management, our evidence for forest-management effects on bird habitat is necessarily indirect. However, given the apparent minimal effects of climate and deforestation on habitat change, forestry-driven degradation is the most parsimonious remaining explanation for substantial habitat declines.
Next, we tested the hypothesis that habitat loss was positively correlated with bird-population declines using BBS data for the Maritime provinces (Methods). We used SDMs to quantify habitat change (1985–2019) in landscapes surrounding BBS routes (N = 90; Supplementary Methods). We then used Bayesian hierarchical models19,25 in a space-for-time approach to test whether the SDM-predicted habitat amount in each year of the time series was associated with population size for each species along each route. Importantly, BBS data are entirely independent of our SDMs, so this test also represents a strong validation of our SDM-derived habitat models. Second, using an additional model parameter, we tested whether annual change in SDM-modelled habitat changes (increases or decreases) along routes in each year could predict annual bird-abundance changes.
Bayesian models revealed a strong effect of habitat amount on BBS bird abundance for all but three species (Fig. 5). Abundance of all but three species tracked annual habitat amount with 95% posterior distributions that did not include zero (vertical line in Fig. 5a; posterior probability, Fig. 5b). The effect of habitat was substantial and probably biologically meaningful for most species, with abundance decreasing a median of 7.99 times from landscapes with the highest to lowest habitat amounts (Supplementary Table 4).
a, Posterior distributions for the effects of SDM-derived habitat amount across routes (x axis) on bird abundance, using BBS data. The vertical black line at zero reflects no positive or negative population trend. Abundance of most species was positively influenced by habitat, which supports the hypothesis that bird populations are strongly linked to breeding habitat amount. b, The posterior probability that habitat had an effect on population size for 54 forest bird species. The vertical black line indicates 95% posterior probability of an effect.
We also found that annual changes in habitat along BBS routes were associated with bird-abundance changes (Extended Data Fig. 8); in other words, habitat loss in one year resulted in abundance declines along routes in the same year. For thirteen species, the Bayesian estimate for the effect of habitat loss on population decline had posterior probabilities >0.95, and 20 species had posterior probabilities >0.8. Importantly, most of the species showing an effect of habitat loss along routes on changes in population decline have lost substantial habitat over the time period and are associated with old forest (for example, Blackburnian warbler, northern parula [Setophaga americana], red-breasted nuthatch [Sitta canadensis], boreal chickadee [Poecile hudsonicus], dark-eyed junco [Junco hyemalis]; Extended Data Fig. 8), which would be expected with the harvest of old forest—a component of forest degradation. It is important to note that this test is highly challenging because many factors can drive annual fluctuations in bird abundance (for example, weather, phenology, conditions during migration or on the wintering grounds). Also, in any given year, habitat change along BBS routes can be quite small for some species; this low inter-annual variation in a predictor variable can preclude high statistical power to detect effects.
We estimated the net number of breeding individuals that have probably disappeared due to habitat loss from 1985 to 2020 using published accounts of territory sizes for each species22 (Supplementary Table 5). This calculation assumes that available habitat is consistently occupied, which is supported by strong associations between habitat amount along BBS routes and bird abundance over the long term. Across all species, back-cast SDMs indicate that a net 28,215,247 ha (282,153 km2) of habitat has been lost, equating to a loss of between 16,779,704 and 52,243,938 breeding pairs (33,559,408–104,487,876 individuals; Supplementary Methods and Supplementary Table 5). One might expect that forest degradation, rather than resulting in broad-scale declines across species, is simply causing species turnover from old forest-associated bird species to young-forest associates. However, it is important to note that we quantified net bird decline from an unbiased list of the 54 most common forest bird species in eastern Canada. This list included both early and late successional species. Such net bird declines could be due to the fact that (1) even some early seral species are losing habitat (probably due to conversion from diverse early successional forest to species-poor plantations and thinnings)26 and (2) in this region, more species occupy older forests than regenerating forests27.
We also quantified overall population trends for 54 species of forest birds using data from the BBS (Fig. 6). These estimates give the total magnitude of population changes which include, but are not limited to, habitat loss or gain effects. Thirty-nine of the 54 species examined (72%) are in population decline (defined as having 95% credible intervals that do not bound zero). The magnitude of the declines for 15 forest bird species is severe (>5% per year). It is notable that most species exhibiting both habitat loss and population declines are old-forest associates (Fig. 4a; bottom left quadrant, dark green dots), with old-forest species exhibiting the greatest habitat losses (Fig. 4b and Supplementary Methods; hierarchical regression, (hat beta) = −16.66 [6.32 SE]).
a, Population trend parameter estimates and posterior distributions for 54 species of forest birds derived from Bayesian models. Seventy-two percent of species that are sufficiently common to model experienced population declines from 1985 to 2019. Colour key is provided in Fig. 5. The vertical green line indicates a population trend of zero. Dashed vertical lines coincide with trends of −15% (−0.15), −10% (−0.10) and −5% (−0.05) annual population trends. b, Predicted linear population trends for 1985–2019 (regression lines are mean trends derived from Bayesian Poisson models, Supplementary Methods) including annual variation estimated from BBS data. Shaded purple areas reflect 95% credible intervals and reflect the magnitude of species population declines shown in a. Populations of these eight old forest-associated species have declined 60–90% over the period observed.
BBS declines are not restricted to old-forest species; several species in rapid population decline are early seral species (for example, Lincoln’s sparrow [Melospiza lincolnii], mourning warbler [Geothlypis philadelphia]; Fig. 4a, bottom right quadrant). Despite the fact that these species have gained habitat over 35 years, their populations continue to decline. Only three species (black-capped chickadee [Poecile atricapillus], hairy woodpecker [Leuconotopicus villosus] and ruby-throated hummingbird [Archilochus colubris]) are increasing in abundance. Populations of these species increased despite evidence of habitat decline (Fig. 4a, top left quadrant)—perhaps because each benefit from anthropogenic habitats and supplemental food. Importantly, habitat changes from 1985 to 2019 along BBS routes were representative of changes at the scale of the entire region for most species (Extended Data Fig. 9), so BBS population trends are highly likely to reflect population trends at the regional scale. This contrasts to the 1965–1985 period when mature-forest loss along routes was slower than in the broader region28.
We also modelled BBS population trends over the past ten years, as this is the period of importance for informing listing decisions under the Committee on the Status of Endangered Wildlife in Canada (COSEWIC). Nine species have exhibited population declines >30% over ten years (Supplementary Fig. 3), which meets the criterion for consideration as ‘threatened’ under COSEWIC Criterion A (ref. 29).
Source: Ecology - nature.com
