Wiebe Nijland

Fine-spatial scale predictions of understory species using climate- and LiDAR-derived terrain and canopy metrics

Abstract Food and habitat resources are critical components of wildlife management and conservation efforts. The grizzly bear (Ursus arctos) has diverse diets and habitat requirements particularly for understory plant species, which are impacted by human developments and forest management activities. We use light detection and ranging (LiDAR) data to predict the occurrence of 14 understory plant species relevant to bear forage and compare our predictions with more conventional climate- and land cover-based models. We use boosted regression trees to model each of the 14 understory species across 4435     km 2 using occurrence (presence–absence) data from 1941 field plots. Three sets of models were fitted: climate only, climate and basic land and forest covers from Landsat 30-m imagery, and a climate- and LiDAR-derived model describing both the terrain and forest canopy. Resulting model accuracies varied widely among species. Overall, 8 of 14 species models were improved by including the LiDAR-derived variables. For climate-only models, mean annual precipitation and frost-free periods were the most important variables. With inclusion of LiDAR-derived attributes, depth-to-water table, terrain-intercepted annual radiation, and elevation were most often selected. This suggests that fine-scale terrain conditions affect the distribution of the studied species more than canopy conditions.

Intertidal resource use over millennia enhances forest productivity

Human occupation is usually associated with degraded landscapes but 13,000 years of repeated occupation by British Columbias coastal First Nations has had the opposite effect, enhancing temperate rainforest productivity. This is particularly the case over the last 6,000 years when intensified intertidal shellfish usage resulted in the accumulation of substantial shell middens. We show that soils at habitation sites are higher in calcium and phosphorous. Both of these are limiting factors in coastal temperate rainforests. Western redcedar (Thuja plicata) trees growing on the middens were found to be taller, have higher wood calcium, greater radial growth and exhibit less top die-back. Coastal British Columbia is the first known example of long-term intertidal resource use enhancing forest productivity and we expect this pattern to occur at archaeological sites along coastlines globally.

Assessing the impact of field of view on monitoring understory and overstory phenology using digital repeat photography.

Abstract Phenological patterns of the components within forest ecosystems, such as understory vegetation, are important indicators of climate variability, productivity, and additional ecosystem services such as food and habitat availability for wildlife. Proximal sensing systems can provide detailed phenological records at a relatively low cost. As interest in these datasets increases, we need additional information regarding the effect of different approaches on the scale of observations and camera field of view. In this research, we examine the impact of field of view on the capacity of cameras to detect changes in phenology of individual species in an image time series. We examine two co-located series of oblique images acquired using a fine and broad field of view and compare a number of phenological indicators, including the start and end of season derived for individual plant species. Our results indicate both fine and broad field of view camera systems are highly effective at detecting key markers of plant phenology with no significant differences between the two. This result supports environmental monitoring using cost-effective broad field of view cameras, or even—subject to some constraints—readily available camera stations installed for tourism or traffic monitoring. Résumé Les patrons phénologiques des composantes au sein des écosystèmes forestiers, comme la végétation de sous-bois, sont des indicateurs importants de la variabilité du climat, de la productivité et des services écosystémiques supplémentaires tels que la disponibilité de nourriture et d’habitats pour la faune. Les systèmes de détection par caméra dans l’environnement peuvent fournir des observations phénologiques détaillées à un coût relativement faible. Comme l’intérêt pour ces ensembles de données s’accroît, nous avons besoin d’informations supplémentaires concernant l’effet des différentes approches sur l’échelle des observations et sur le champ de vision de la caméra. Dans cette recherche, nous examinons l’impact du champ de vision sur la capacité des caméras à détecter des changements dans la phénologie des espèces individuelles dans une série temporelle d’images. Nous examinons deux séries colocalisées d’images obliques acquises en utilisant des champs de vision étroit et large et nous comparons quelques indicateurs phénologiques, y compris le début et la fin de la saison, dérivés pour des espèces de plante individuelle. Nos résultats indiquent que les systèmes de caméras avec des champs de vision large et étroit sont très efficaces pour la détection de marqueurs clés de la phénologie des plantes sans différence significative entre les deux. Ce résultat supporte le concept de la surveillance de l’environnement en utilisant des caméras économiques avec de larges champs de vision, ou même—avec certaines contraintes—des stations de caméras facilement accessibles installées pour le tourisme ou la surveillance du trafic.

Heading for the hills? Evaluating spatial distribution of woodland caribou in response to a growing anthropogenic disturbance footprint

Abstract Anthropogenic landscape change (i.e., disturbance) is recognized as an important factor in the decline and extirpation of wildlife populations. Understanding and monitoring the relationship between wildlife distribution and disturbance is necessary for effective conservation planning. Many studies consider disturbance as a covariate explaining wildlife behavior. However, we propose that there are several advantages to considering the spatial relationship between disturbance and wildlife directly using utilization distributions (UDs), including objective assessment of the spatially explicit overlap between wildlife and disturbance, and the ability to track trends in this relationship over time. Here, we examined how central mountain woodland caribou (Rangifer tarandus caribou) distribution changed over time in relation to (i) anthropogenic disturbance, baseline range (defined using telemetry data from 1998 to 2005), and alpine habitat; and (ii) interannual climate variation (North Pacific Index; NPI). We developed seasonal UDs for caribou in west‐central Alberta and east‐central British Columbia, Canada, monitored with GPS collars between 1998 and 2013. We mapped the cumulative annual density of disturbance features within caribou range and used indices of overlap to determine the spatial relationship and trend between caribou UDs, anthropogenic disturbance, baseline range, alpine habitat, and the NPI. Anthropogenic disturbance increased over time, but the overlap between caribou UDs and disturbance did not. Caribou use of alpine habitat during spring, fall, and late winter increased over time, concurrent with a decrease in use of baseline range. Overlap between caribou UDs and disturbance increased during spring and fall following relatively cold, snowy winters (high NPI), but overall, climate did not explain changes in caribou distribution over time. We provide evidence supporting the hypothesis that caribou populations adjust their spatial distribution in relation to anthropogenic landscape change. Our findings could have implications for population persistence if distributional shifts result in greater use of alpine habitat during winter. Monitoring long‐term changes in the distribution of populations is a valuable component of conservation planning for species at risk in disturbed landscapes.

Characterizing a Decade of Disturbance Events Using Landsat and MODIS Satellite Imagery in Western Alberta, Canada for Grizzly Bear Management

Abstract Mapping and quantifying the area and type of disturbance within forests is critical for sustainable forest management. Grizzly bear (Ursus arctos) have large home ranges and diverse habitat needs and as a result, information on the extent, type, and timing of disturbances is important. In this research we apply a remote-sensing-based disturbance mapping technique to the southeastern extent of a grizzly bear range. We apply a data fusion approach with MODIS 250 m and Landsat 30 m spatial resolution imagery to map disturbances biweekly from 2001–2011. A regression tree classifier was applied to classify the disturbance events based on spatial and temporal characteristics. Fire was attributed as a disturbance based on a national fire database. Results indicate that across the 130,727 km2 study area, 4,603 km2 of forest were disturbed over the past decade (2001–2011), impacting 0.35% of the study area annually. Overall, 68.7% of the disturbance events were attributed to forest harvest, followed by well sites 13.4%, fires 9.3% and road development, 8.6%. Primary source habitat contained 3.8% of disturbed land, and primary sink areas had 5.9% disturbed land. Our findings quantify habitat change, which can aid managers by identifying significant areas for grizzly bear conservation.

Deriving Rich Coastal Morphology and Shore Zone Classification from LIDAR Terrain Models

ABSTRACT Nijland, W.; Reshitnyk, L.Y.; Starzomski, B.M.; Reynolds, J.D.; Darimont, C.T., and Nelson, T.A., 2017. Deriving rich coastal morphology and shore zone classification from LIDAR terrain models. Comprehensive mapping of shore-zone morphology supports evaluation of shore habitat, monitoring of environmental hazards, and characterization of the transfer of nutrients between marine and terrestrial environments. This article shows how rich shore-zone morphological metrics can be derived from LIDAR terrain models and evaluates the application of LIDAR to classify shore-zone substrates. The utility of LIDAR methods was tested in comparison with the current best-practice method of photo interpretation (i.e. the BC ShoreZone system) on Calvert Island, British Columbia, Canada. Wider applications are considered. Indicators of shore-zone morphology (i.e. slope, width, roughness, backshore elevation) were calculated from LIDAR terrain models for regularly spaced transects perpendicular to the coastline. A combination of boosted regression-tree modeling and direct-rule application was used to classify the shore-zone morphology according to the British Columbia (BC) ShoreZone system. Classification accuracy was assessed against existing ShoreZone classification data. Shore-zone substrate was classified from LIDAR-derived morphometric indicators with 90% accuracy (five classes). A full classification, which combined substrate with shore width and slope, results in lower correspondence (40%; 25 classes) when compared with ShoreZone classes. Differences can likely be attributed, in part, to variation in spatial resolution of elevation-based methods and photo interpretation. It is concluded that LIDAR data can be used to support characterization of shore-zone morphology. Differences in processing and interpretation cause a low direct correspondence with the current image-based classification system, but LIDAR has the advantage of higher resolution, rich terrain information, speed, and an objective and repeatable method for monitoring future change in coastal environments.

Regression tree modeling of spatial pattern and process interactions

In forestry, many fundamental spatial processes cannot be measured directly and data on spatial patterns are used as a surrogate for studying processes. To characterize the outcomes of a dynamic process in terms of a spatial pattern, we often consider the probability of certain outcomes over a large area rather than on the scale of the particular process. In this chapter we demonstrate data mining approaches that leverage the growing availability of forestry-related spatial data sets for understanding spatial processes. We present classification and regression trees (CART) and associated methods, including boosted regression trees (BRT) and random forests (RT). We demonstrate how data mining or machine learning approaches are useful for relating spatial patterns and processes. Methods are applied to a wildfire data and covariate data are used to contextualize the quantified patterns. Results indicate that fire patterns are mostly related to processes influenced by people. Given the growing number of multi-temporal and large area datasets on forests and ecology machine learning and data mining approaches should be leveraged to quantify dynamic space-time relationships.