Skip to main content
ARS Home » Research » Publications at this Location » Publication #244580
Title: Inexpensive Tools To Quantify And Map Vegetative Cover For Large-Scale Research Or Management Decisions.

Author
item WILES, LORI
item WALTERMIRE, R - U.S. GEOLOGICAL SURVEY (USGS)
item BAUSCH, WALTER

Submitted to: Range Management Meeting Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 2/7/2010
Publication Date: 2/7/2010
Citation: Wiles, L., Waltermire, R., Bausch, W.C. 2010. Inexpensive Tools To Quantify And Map Vegetative Cover For Large-Scale Research Or Management Decisions.Range Management Meeting Proceedings.

Interpretive Summary:

Technical Abstract: Vegetative cover can be quantified quickly and consistently and often at lower cost with image analysis of color digital images than with visual assessments. Image-based mapping of vegetative cover for large-scale research and management decisions can now be considered with the accuracy of these methods, the availability of inexpensive digital cameras and GPS units, advances in image handling in program languages and the availability of open-source programs for creating and displaying maps. Ideally, systems for making these maps will be designed for users with limited access to expensive GIS software and expertise and decision makers who will want to examine the images to consider not only the amount of vegetation, but also characteristics such as plant species and size. We designed a simple system and developed software to map weed cover over large areas in fallow fields that quantifies vegetative cover based on color bands of individual pixels and does not require sophisticated and expensive GIS and image analysis software. Images can be collected from a moving vehicle without supplemental lighting. Classification is automated and users with minimal GIS expertise can create a map of vegetative cover and view the original and classified images as well as image and GPS data by selecting a location on the map. We implemented the system for $2500 and have used it to map weed cover in 15 fallow fields and to quantify cover of wheat, sunflowers, dry beans and corn in response to water stress. We could create a map from 1000 images and 5000 GPS coordinates in 30 minutes. Weeds of many different species and as small as two cotyledons were detected and crop cover was accurately quantified throughout the season without changing or calibrating the classification algorithm. The system can be adapted for other uses as the classification algorithm can be easily modified, consumer or professional digital cameras and GPS units can be used, and the file for mapping is in standard GIS format.