The overall health of forests can be assessed using the micro- and macronutrient content of tree leaves to help guide forest management decisions in the face of climate change, species loss, and other variables. Traditional methods for assessing nutrient levels in forests are expensive and labor-intensive.

Researchers recently analyzed the reflected spectra of tree leaves to precisely determine the nutrient content of the leaves, providing a faster and larger-scale method for assessing forest health.

Field methods that involve collecting leaf samples and then measuring the nutrient content of the foliage in the laboratory are time-consuming. As climate change alters growing conditions, new, faster methods are needed to assess the health of the forest.

To solve this problem, a team of researchers from the University of Massachusetts and the University of Virginia analyzed the light reflected from forest leaves (spectra) over a wide range of wavelengths to accurately determine the concentrations of calcium (Ca), magnesium (Mg), potassium (K), phosphorus (P), manganese (Mn), and zinc (Zn) in leaves.

The team published the study on June 27 in Journal of Remote Sensing.

Specifically, the researchers measured reflected light at wavelengths ranging from 400 to 2,450 nanometers (nm, 1.0 x 10^-9 m) to determine nutrient concentrations. The team was able to determine the best wavelengths to measure each nutrient analyzed using partial least squares regression (PLSR). PLSR is particularly well suited to managing highly correlated independent variables, such as individual reflectivity over a continuous spectrum.

“The developed PLSR model predicted plant nutrients with moderate to high accuracy for macro- and micronutrients in temperate hardwood forests in the northeastern United States. Spectral measurements combined with wavelength selection and PLSR models can be used to quantify macro- and micronutrients in leaves at a regional scale and can be further improved by incorporating local geological materials and tree genera,” said Qian Yu, associate professor of earth, geography and climate sciences at the University of Massachusetts-Amherst and corresponding author of the research paper.

The researchers realized that they could use spectra to determine nutrient concentrations in tree foliage much better when they took into account the type of soil the trees were growing in. Notably, this variable is often overlooked when assessing the nutrient composition of tree leaves.

The team measured tree foliage nutrients Ca and P in four soil types: coarse glacial moraine soil, glacial fluvial soil, melt moraine soil and outwash soil. By taking soil type into account, the researchers improved the accuracy of their PLSR nutrient analysis of R² = 0.66 to R² = 0.87 (one R²or coefficient of determination, which is 1.0, means that all observed deviations are due to differences in the independent variable or to spectral measurements).

Different soils had greater or lesser effects on Ca and P. For example, soil had little effect on Ca concentrations but tended to affect P to a greater extent. Alluvial soils also provided the most accurate predictions for Ca and P nutrients. Crucially, Ca is a limiting nutrient in forests, playing important roles in plant structure, chemical signaling, and as an enzyme cofactor that alters enzyme efficiency.

The tree genus also influenced the accuracy of foliage nutrient concentration predictions. In fact, analysis of spectral data based on tree genus improved the accuracy of nutrient predictions even better than soil composition. Tree genus improved the Ca prediction accuracy of R² = 0.66 to R² = 0.91 and R² = 0.93 for the genera Fagus (beech) and Quercus (oak), respectively. The team also analyzed spectral data for the genera Acer (maple) and Betula (birch).

The research team suspects that certain tree genera are likely to influence leaf nutrients because the physiological mechanisms responsible for nutrient uptake and transport are genetically controlled and unique to each genus. In addition, some tree genera may require certain nutrients more than other genera that are better adapted to a particular soil, leading to differences in nutrient uptake.

Ultimately, the research team hopes that the new spectral assessment technique can be successfully applied to other deciduous forests, supporting important forest management decisions.

“The method presented in this article shows promise for large-scale plant nutrient assessment and can reduce the costs of traditional field-based approaches,” said Wenxiu Teng, a doctoral student in earth, geography and climate sciences at the University of Massachusetts-Amherst and lead author of the article.