A spectral three-dimensional color space model of tree crown health

Abstract

Protecting the future of forests in the United States and other countries depends in part on our ability to monitor and map forest health conditions in a timely fashion to facilitate management of emerging threats and disturbances over a multitude of spatial scales. Remote sensing data and technologies have contributed to our ability to meet these needs, but existing methods relying on supervised classification are often limited to specific areas by the availability of imagery or training data, as well as model transferability. Scaling up and operationalizing these methods for general broadscale monitoring and mapping may be promoted by using simple models that are easily trained and projected across space and time with widely available imagery. Here, we describe a new model that classifies high resolution (~1 m2) 3-band red, green, blue (RGB) imagery from a single point in time into one of four color classes corresponding to tree crown condition or health: green healthy crowns, red damaged or dying crowns, gray damaged or dead crowns, and shadowed crowns where the condition status is unknown. These Tree Crown Health (TCH) models trained on data from the United States (US) Department of Agriculture, National Agriculture Imagery Program (NAIP), for all 48 States in the contiguous US and spanning years 2012 to 2019, exhibited high measures of model performance and transferability when evaluated using randomly withheld testing data (n = 122 NAIP state x year combinations; median overall accuracy 0.89–0.90; median Kappa 0.85–0.86). We present examples of how TCH models can detect and map individual tree mortality resulting from a variety of nationally significant native and invasive forest insects and diseases in the US. We conclude with discussion of opportunities and challenges for extending and implementing TCH models in support of broadscale monitoring and mapping of forest health.

Scattered tree death contributes to substantial forest loss in California

“…Brown-stage trees appear brownish or reddish on RGB images (Fig. 3e ) and imply recent tree death with dried or decoloured foliage remaining on stems 48 . Grey-stage trees have lost their foliage, appear greyish or whitish (Fig.…”

“…Grey-stage trees have lost their foliage, appear greyish or whitish (Fig. 3f ), and imply long-standing dead woods 48 . Differentiating between brown- and grey-stage allows for crude approximations of recent mortality from single-year NAIP images, along with species and damage agent information.…”

An approach to assessing subsea pipeline-associated mercury release into the North Sea and its potential environmental and human health impact

“…Such preferences were not considered in the present work, but future research would benefit from a more detailed assessment of society’s seafood consumption habits. This is especially relevant when considering contaminants such as methylmercury, with the potential to bioaccumulate and biomagnify, as all preferred fish species fall within trophic levels ≥4 and thus pose an increased risk of exceeding the FS for mercury, as well as the EWI MeHg [ 115 ]. It should also be noted that most of the salmon consumed in the UK is farmed, thus being fed with fishmeal containing marine caught species of a lower commercial value such as shrimp or sand eel.…”

A multi-scale approach to detecting standing dead trees in UAV RGB images based on improved faster R-CNN

“…The digital cameras carried by UAVs account for the largest proportion of all sensors, but only a few researchers proposed using RGB images obtained by digital cameras to identify SDTs. Existing identification methods include visual interpretation, HSV thresholding, the method of combining spectral with texture analyses, and a super-pixel segmentation algorithm, also called linear spectral clustering for SVM [27][28][29]. Though traditional object detection methods can effectively identify SDTs in different types of images, given the different characteristics of SDTs of different species, such methods need to first calculate extraction factors based on artificially designed features, which leads to poor robustness.…”