Publications by authors named "Alexander O Headman"
Article Synopsis
- Understanding the extent of subsurface drainage (tile-drain) is crucial for analyzing landscape responses to rainfall and soil management impacts on stream health and water quality.
- A UNet machine-learning model was developed to detect tile-drain networks in satellite images without needing detailed data on soil or terrain, achieving high accuracy similar to expert manual tracing.
- The model performed best in spring conditions, identifying tile drains with 93%-96% accuracy, thus helping to manage nutrient and sediment flow which is essential for addressing issues like harmful algal blooms.
