Northern
Bobwhite
Landscape Level
Evaluation of Northern Bobwhite
Habitats in
Eastern Virginia Using Landsat TM Imagery
Northern
Bobwhite
Northern Bobwhite (Colinus
virginianus) are a small and exciting game bird associated with agricultural
lands, edges between habitat types, young pine plantations, and mature
pine stands that have been extensively thinned to allow light penetration
to the forest floor. In the past 30+ years there has been an almost
universal decline in bobwhite population numbers despite a long history
of management. The Virginia Bobwhite Quail Management Plan was implemented
in 1996 to slow and stop the current population declines in Virginia.
Remote Sensing
A large scale land cover map along with a detailed understanding of the spatial arrangements of bobwhite habitats will not only aid Virginia’s quail management plan, but also allow focused efforts by our wildlife managers. I explored the possibilities of using remote sensing to map various habitats important to bobwhite. I compared several classification algorithms applied to Landsat TM imagery prior to selecting the classification method that best delineated early successional habitats. The final method was a hybrid between the traditional supervised and unsupervised algorithms. After I selected this method, I created a classified land cover map for the coastal plain and piedmont of Virginia.
Models based on Classified Habitat Maps
A detailed understanding of the spatial arrangement bobwhite habitats would allow more focused efforts by wildlife managers. I used a 4-year average of northern bobwhite call count data in conjunction with the remotely-sensed habitat maps to study landscape-level habitat associations. Landscape metrics were calculated for the landscape surrounding each stop and were used in two modeling exercises to differentiate between high and low northern bobwhite populations. Both pattern recognition (PATREC) and logistic regression models predicted levels of northern bobwhite abundance well for the modeled (73.5% and 73.9% respectively) and independent (74.6% and 76.6% respectively) data sets. The revised models were applied to the remotely-sensed habitat maps of the eastern 2/3 of Virginia to develop maps expressing the quality of a landscape for supporting a high population of bobwhite based on existing land cover. Both models predicted similar percentages in each of the quality classes.
| The modeled results for the eastern portion of Virginia showing final PATREC probabilities. Areas in yellow and red have the highest probability of supporting a high bobwhite population while areas in the blues and green have the lowest probability of supporting the high quail population. | The modeled results for the eastern portion of Virginia showing final Logistic Regression probabilities. Again, areas in yellow and red have the highest probability of supporting a high bobwhite population while areas in the blues and green have the lowest probability of supporting the high quail population. |
 |
 |
Wildlife Models Directly From Remotely-Sensed Imagery
Finally, I explored the possibility
of eliminating the time consuming and very costly step of classifying a
remotely-sensed image prior to examining its quality for a particular species.
Using raw Landsat TM imagery and bobwhite call count data, I developed
predictive logistic regression models expressing the quality of a landscape
surrounding a pixel. The first model predicted the probability of
the landscape supporting a high bobwhite population. Due to a number
of stops with an average of zero, I was also able to generate a model that
expressed the probability of the landscape supporting any number of bobwhite.
This method also satisfactorily predicted high/low population and presence/absence
for the modeled data (65.7% and 83.1%, respectively) and independent data
(65.3% and 83.7%, respectively). The method described will allow
for rapid assessment of our wildlife resources without having to classify
remotely-sensed images into habitat classes prior to analyses.
 |
 |
| Final probability map for the high/low population model developed directly from the raw, remotely sensed imagery. Areas in yellow and red have the highest probability of supporting a high bobwhite population while areas in the blues and green have the lowest probability of supporting the high quail population. | Final probability map for the presence/absence population model developed directly from the raw, remotely sensed imagery. Areas in yellow and red have the highest probability of supporting any number of bobwhite while areas in the blues and green have the lowest probability of supporting any number of bobwhite. |
More Detailed Information
• View a slide show with more information
• Download thesis for ultimate detail
• Quail
Information and Management Options
For more information contact Garrett Schairer or Jeff Waldon