Human activities have transformed natural landscapes into human-dominated areas at unprecedented rates in the last centuries. Land cover transformation is associated with loss of natural habitat, thus a threat to biodiversity. Because habitat loss will likely continue in the future due to population growth and increase demand for natural resources, an important question in ecological studies is whether land cover features (i.e. amount, variety, shape, configuration) can be used as predictors to estimate species loss from habitat modification. This thesis investigates the predictive ability of landscape features in predicting species distributions at the landscape level and across large regions. It tests several predictions from classic hypotheses such as the species-area relationship and habitat fragmentation, utilizing a macroecological approach. Response variables (e.g. species richness, species’ probability of occurrence) and independent variables (e.g. proportion of natural areas, metrics of fragmentation, temperature, etc.) are analysed in cell sizes of 25-900km2 covering large regions (e.g. southern Ontario, New York State). Bird species were chosen as the main biological model. Most literature assumes that species richness should vary positively as a function of remaining natural area, following the well-known species–area relationship (i.e. classic SAR). Prior studies have shown that avian species richness has a peaked, rather than a monotonic increasing, relationship with the proportion of natural land cover in landscapes of southern Ontario. The first chapter of the thesis showed improvements in the predictive power of classic SARs by proposing the “Lost-habitat SAR”, which demonstrates that richness of open-habitat species can be predicted when we partition human-dominated land cover into an ‘‘available human-dominated’’ component and ‘‘lost’’ habitat (i.e. parts of the landscape that can no longer be utilized by any species). The second chapter addresses a current contention in the literature about the effect of habitat fragmentation beyond habitat amount at the landscape level. Specifically, I tested the effect of fragmentation (e.g. number of patches) on both avian richness and the probability of occurrence (pocc) of individual species, after controlling for habitat amount in 991 landscapes, each 100-km2, in southern Ontario. The analysis showed that overall species richness responds primarily to habitat amount, and that the effect of habitat fragmentation, holding the total amount of habitat constant, is negligible. The probability of occurrence of a few bird species did relate negatively to the size, number and isolation of the patches within the landscape. We argue that the evidence is inconsistent with the hypothesis that reducing habitat fragmentation would be an effective conservation strategy for birds at the landscape level. Chapter 3 tested the speculation in the climate change literature that habitat loss may impede the colonization or movement of species whose ranges are shifting northwards in response to climate. Using the same 100-km2 landscapes of southern Ontario, I examined individual bird species’ probability of occupancy as a function of the amount of remaining natural land cover for three groups of species: i) those whose northern range limit falls within the study area, ii) those whose southern range limit is in the study area, and iii) those whose ranges cover the entire study area. The results showed that the probability of occupancy of southern-edge species is a positive function of the amount of natural land cover (forest) in the landscape, while the probability of occupancy of northern-edge species is a negative function of natural land cover. Hence, I conclude that at southern range limits species faces the dual stresses of climatic warming and habitat conversion. Whereas, at northern (potentially expanding) range edges, partially disturbed landscapes are more readily occupied than undisturbed landscapes. In the final chapter, I challenge widely accepted hypothesis that habitat loss causes biodiversity loss by testing whether conserving natural land cover would conserve species diversity. More specifically, I tested whether broad-extent relationships between avian species richness and natural land cover are independent of: 1) whether species distribution data come from systematic censuses (atlases) versus range maps, and 2) the grain size of the analysis in grid cells covering southern Ontario, CA, and New York State, US. My findings showed that over regional extents, range-map-based richness relates strongly to temperature, irrespective of spatial grain, and that censused species richness relates to temperature less strongly. Moreover, range-map richness is a negative function of the proportion of natural land cover, while realized richness is a peaked function. Therefore, I conclude that conserving natural land cover would not conserve species diversity in southern Ontario or in New York State, since greater natural cover does not imply higher richness. We argue that habitat loss has become a panchreston. It may be misguiding conservation biology strategies by focusing on a threat that is too general to be usefully predictive.