Abstract:

The lack of efficient phenotyping capability has become a bottleneck to the studies of forest tree breeding, and the key difficulty of phenotyping is to capture accurate phenotypic data so that the results acquired can be correctly explained. The central challenge of precision forestry is automatic, large-scale and real-time analysis of phenotypic traits. Conventionally, evaluation of tree phenotypic traits is manually-performed, subjective, inefficient, destructive and error-prone. The manual analysis is unable to meet the needs of exploring the internal relationships among "genotype-phenotype-environment" and unraveling the formation mechanism of specific biological traits in a comprehensive manner. Modern phenotyping technology uses a system equipped with various types of imaging sensors to automatically collect a large number of phenotypic data, such as forest morphology, structure, physiology and biochemistry, so as to monitor the growth of individual trees. In addition, the nondestructive measurements enable continuous monitoring of the same individual trees to obtain the phenotypic traits related to growth and development. For example, in stress studies, phenotyping technology can clarify the model of response and resistance of the trees to the stress. Effective implementation of remote sensing to phenotype trees that facilitates accurate, high-throughput, automatic and non-destructive screening of trees in genetic tests is critical for accelerating tree improvement and breeding strategies for higher yield and stress-resistance in precision forestry. We present a review of recent advances in forestry phenotyping and an introduction of individual-based and stand-based forestry phenotyping. Thus, various techniques for tree phenotyping were presented together with applications of these techniques. The measured parameters, spectral range, imaging principle, advantages and disadvantages of visible camera, fluorescence imaging sensor, near-infrared camera, hyperspectral imaging sensor, thermal infrared imaging sensor, and LiDAR scanner, as well as the development of tree phenotype information collection were analyzed. Furthermore, faster and higher-resolution tree data collection would lead to the improvement of precision-forestry practices. Future application of phenotyping platforms requires:1) establishing new platforms to obtain the key phenotypic traits of stands and individuals so as to improve the accuracy and throughput; 2) using environmental monitoring techniques to analyze the phenotypic responses of trees under abiotic stress so as to carry out breeding for high resistance varieties; 3) using the phenotypic changes under biological stress to promote the accurate monitoring, classification, identification and control of pests and diseases; 4) using high throughput phenotyping technology to combine with genome wide selection, quantitative trait loci and genome wide association study to identify the function of genes. In this paper, it was described that advances in phenotyping technologies lead to real-time, high-quality and high-throughput data on trees to accelerate tree improvement through breeding and to optimize precision forestry practices.

Key words: phenotype, imaging sensor, precision forestry, forest tree genetics and breeding, biotic stress