Using Landscape Characteristics to Predict Distribution of Temperate-Breeding Canada Geese

Abstract

Land cover classification is a valuable tool for professionals in a diverse range of fields, ranging from environmental and ecosystem management, to land use planning and fire management, these applications play an important role in both the public and private sectors. The most thorough and recent of the datasets used for land cover classification have been the 1992 and 2001 30-meter-posted National Land Cover Database (NLCD) datasets created by the United States Geological Survey (USGS). While these datasets provide a good medium-scale land cover dataset, there are limitations to the NLCD's accuracy and use in finer-scale applications. Under the NEXTMap ® USA program, Intermap Technologies™ is assembling a nationwide dataset of high-resolution 1.25-m orthorectified radar imagery (ORI) and 5 m elevation datasets for the entire conterminous United States. NLCD data and NEXTMap ® Land Cover Data were compared in five different study areas across the United States (California, Colorado, Montana, and two locations in Minnesota), and verified with field measurements. Nine land cover classes (water, barren, grassland, urban, shrub, mixed forest, deciduous forest, evergreen forest, and wetlands) constituted the majority of the study areas. Overall, the result of using NEXTMap radar and elevation data for classification of land cover yielded very favorable results. The majority of the land cover classes were delineated with an overall accuracy ranging between 86.30% -86.91% on the order of 90%, versus 59.16% -63.93% for the NLCD. The NLCD map often confusing deciduous and wetland, underestimating evergreen, and overestimating shrub vegetation classes.

A comparison of field methods to estimate Canada goose abundance

“…Plot‐survey data are easily processed and analyzed by the collecting entity; abundance estimates are available soon after surveys are completed. Finally, the plot survey can potentially provide information on goose–habitat relationships, including regional distributions and relative abundance (McAlister et al ).…”

“…). Eliminating nonhabitat before selecting survey plots eliminated regional effects on goose abundance, and we assumed that goose densities were similar in goose habitat regardless of physiographic region (McAlister et al ). We assigned plots a unique number and used a random number generator to choose 300 plots to survey (Fig.…”

Habitat Suitability modeling of Urial (Ovis orientalis arkal) in the Samelghan plain by using Maximum Entropy Method

“…Generally, due to the irregular distribution of species in habitats, it is difficult and costly to determine the exact distribution of species [9], therefore, such models are naturally static and probable, because statistically, the geographical distribution of species or communities are related to their current habitats [10], so these models can perform well in describing the natural distribution of species (within their current range). It is essential to accurately estimate the distribution of species, to ensure that environmental protection planning efforts are more beneficial in managed areas [9]. Species distribution models or habitat suitability models allow potentially predicting human effects on biodiversity patterns at different spatial scales, the origin more of modeling methods to predict of fauna and flora distribution have in environmental-species relations [11].…”

Habitat Suitability Modeling for Wildlife Management Objectives by Using Maximum Entropy Method

“…Generally, due to the irregular distribution of species in habitats, it is difficult and costly to determine the exact distribution of species [9]. Therefore, such models are naturally static and probable because, statistically, the geographical distribution of species or communities is related to their current habitats [10], so these models can perform well in describing the natural distribution of species (within their current range).…”