How drought stress becomes visible upon detecting tree shape using terrestrial laser scanning (TLS)

Due to climate change, the occurrence of drought events with essential effects on trees will arise. The impact of severe drought stress on trees’ vitality with regard to growth has often been analysed using traditional, easy-to-measure variables, such as diameter at breast height (d_1.3). Another commonly used tree-vitality indicator is crown transparency, which is not directly measurable and has to be determined qualitatively by well-trained field experts. In this study, we focused on tree dimensions, as potential vitality indicators, that are difficult to measure. The new approach for the efficient monitoring of tree vitality introduced here revealed three-dimensional change of tree shape due to drought stress. The unique drought stress experiment “Kranzberg Forest Roof Experiment” (KROOF) was used as a basis for scanning and analysing the growth of Norway spruce (Picea abies (L.) H. Karst.) and European beech (Fagus sylvatica L.) under progressively limiting water reserves. Before the start of the experiment in the winter of 2012/2013, terrestrial laser scanning (TLS) was performed and repeated in the winter of 2018/2019. One sample of 21 trees was trenched and roofed (treatment), while additional 26 trees served as untreated reference (control). Using the TLS-point clouds of the two subsequent surveys, structural tree modifications within the 6-year period can be directly visualised, computed and linked to drought stress. Drought stress led to significantly smaller crown size and lower height growth for both tree species. The crowns of Norway spruce trees increased significantly in transparency and roughness. In addition, high competition combined with drought stress significantly reduced the roughness and increased the compactness of the crown. The periodic annual change in crown projection area (pac_cpa) as well as the periodic annual height increment (pai_height) differed significantly between control and treatment for both tree species. Under drought conditions, pac_cpa changed by –0.74 m² yr⁻¹ and –0.42 m² yr⁻¹ for spruce and beech trees respectively, whereas the control trees showed a growth of 0.17 m² yr⁻¹ and 0.62 m² yr⁻¹ respectively. This means that crowns became considerably smaller under dry conditions. Under drought, the pai_height was 0.09 m yr⁻¹ less for spruce and 0.17 m yr⁻¹ less for beech compared with normal growing conditions. The periodic annual change in crown roughness (pac_roughness) was −9.5% yr⁻¹ if local competition increased by one. Our results show that TLS can offer new opportunities for identifying structural features in trees. Iterative TLS-surveys may extend existing measuring campaigns on common long-term experimental plots, in order to analyse general changes or monitor tree vitality.