Fine Resolution Imagery and LIDAR-Derived Canopy Heights Accurately Classify Land Cover with a Focus on Shrub/Sapling Cover in a Mountainous Landscape

Abstract

Publicly available land cover maps do not accurately represent shrubs and saplings, an uncommon but ecologically relevant cover type represented by woody vegetation <4 m tall. This omission likely occurs because (1) the resolution is too coarse, (2) poor training data are available, and/or (3) shrub/saplings are difficult to discriminate from spectrally similar classes. We present a framework for classifying land cover, including shrub/saplings, by combining open-source fine-resolution (1 m) spectral and structural data across a large (>6000 km2) mountainous region. We hypothesized that the combination of spectral (imagery) and structural (LIDAR) data would allow for discrimination of shrub/sapling cover from other cover types. Specifically, we created training data using segmented four-band imagery from the National Agricultural Imagery Program (NAIP). In addition to spectral information from imagery, we used topographic information (elevation, slope, and aspect) and a LIDAR-derived canopy height model to classify land cover within a pixel-based random forests framework. To assess model accuracy, we used image interpretation and an independent sample of validation points. Due to the fine resolution of predictor rasters across such a large geographic region, we classified five subregions (counties) separately. We also compared the landscape metrics calculated for our custom classification at fine (1 m) and coarse resolution (resampled to 30 m) to metrics calculated with National Land Cover Data (NLCD). We achieved an overall accuracy of 89% and >80% accuracy for each land cover class. The LIDAR-derived canopy height model was consistently ranked as the most important predictor of vegetative land cover classes. Compared with our custom classification, NLCD underrepresented pasture/grassland by up to 10% and overrepresented forest up to 30%. There was no correlation between percent shrub/sapling cover in our custom classification and NLCD, suggesting that NLCD is not reliable for applications concerned with this ecologically relevant cover type.

Breeding season space use and habitat selection by Blue-winged Warblers in managed shrublands

“…Other sources of remotely sensed data, like the National Land Cover Database (Dewitz 2021) are recognized to poorly capture cover types that shrubland birds are dependent on (i.e., old fields, young forest, etc. ; Bulluck et al 2022). Recent work by McNeil et al (2023) assessed the value of LiDAR data in Pennsylvania for predicting the occupancy of Golden-winged Warblers in the Pocono Mountains region.…”

LiDAR Data Fusion to Improve Forest Attribute Estimates: A Review

“…6) Other: 9.1% of the papers [148][149][150][151][152][153][154][155][156][157][158][159][160] include a variety of applications, such as mapping the pigment distribution and quantifying taxonomic, functional, and phylogenetic diversity, tree age estimation etc. 7) Fuel load: 6.3% of the papers [161][162][163][164][165][166][167][168][169] include applications that deal with fuel load and forest fire modeling.…”