Machine learning methods for assessing photosynthetic activity: environmental monitoring applications

Abstract

Monitoring of the photosynthetic activity of natural and artificial biocenoses is of crucial importance. Photosynthesis is the basis for the existence of life on Earth, and a decrease in primary photosynthetic production due to anthropogenic influences can have catastrophic consequences. Currently, great efforts are being made to create technologies that allow continuous monitoring of the state of the photosynthetic apparatus of terrestrial plants and microalgae. There are several sources of information suitable for assessing photosynthetic activity, including gas exchange and optical (reflectance and fluorescence) measurements. The advent of inexpensive optical sensors makes it possible to collect data locally (manually or using autonomous sea and land stations) and globally (using aircraft and satellite imaging). In this review, we consider machine learning methods proposed for determining the functional parameters of photosynthesis based on local and remote optical measurements (hyperspectral imaging, solar-induced chlorophyll fluorescence, local chlorophyll fluorescence imaging, and various techniques of fast and delayed chlorophyll fluorescence induction). These include classical and novel (such as Partial Least Squares) regression methods, unsupervised cluster analysis techniques, various classification methods (support vector machine, random forest, etc.) and artificial neural networks (multilayer perceptron, long short-term memory, etc.). Special aspects of time-series analysis are considered. Applicability of particular information sources and mathematical methods for assessment of water quality and prediction of algal blooms, for estimation of primary productivity of biocenoses, stress tolerance of agricultural plants, etc. is discussed.

Keywords: Ecological monitoring; Machine learning; Photosynthesis; Phytoplankton; Primary productivity; Stress tolerance.

Fig. 1

An example regression tree (Dalaka et al. 2000). Reprinted from Ecological Modelling, Vol. 129, A. Dalaka, B. Kompare, M. Robnik-Sikonja, S.P. Sgardelis, Modelling the effects of environmental conditions on apparent photosynthesis of Stipa bromoides by machine learning tools, Pages 245–257, © 2000, with permission from Elsevier

Fig. 2

The workflows of regression stacking for phenotyping photosynthetic capacities. ANN, artificial neural network; SVM, support vector machine; LASSO, least absolute shrinkage and selection operator; RF, random forest; GP, Gaussian process; and PLS, partial least squares. P and p are model predictions at different modeling stage. The regression models are trained with a leave-one-out cross validation approach (the Nth fold is reserved) to form the out-of-sample predictions matrix. The final predictions of each fold were made using the LASSO model based on the out-of-sample predictions matrix (no data normalization). Fu et al. (2019), licensed under CC-BY 4.0

Fig. 3

Working principle of an artificial neuron (Decaro et al. 2019). Licensed under CC-BY 4.0

Fig. 4

Typical schemes of ANN, CNN, and RNN models. a General ANN architecture, b general CNN architecture, and c general RNN architecture (Yu et al. 2022). © 2021 Wiley Periodicals LLC

Fig. 5

Overview of the proposed SSTNN. Reprinted from International Journal of Applied Earth Observation and Geoinformation, Vol. 102, Qiao et al., Crop yield prediction from multi-spectral, multi-temporal remotely sensed imagery using recurrent 3D convolutional neural networks, 102,436, © 2021, with permission from Elsevier

Fig. 6

Schematic block diagram illustrating the main components of a mixture density network (MDN), a class of neural networks that estimates multivariate probability density functions with their corresponding parameters (μ, Σ) and mixing coefficients (α) to arrive at an optimal Chla retrieval. Note that a covariance matrix (Σ) is reduced down to standard deviation (σ) when a single target variable (e.g., Chla) is sought. Reprinted from Remote Sensing of Environment, Vol. 240, Pahlevan et al., Seamless retrievals of chlorophyll-a from Sentinel-2 (MSI) and Sentinel-3 (OLCI) in inland and coastal waters: a machine-learning approach, 111,604, © 2020, with permission from Elsevier