Individual-based measurement and analysis of root system development: Case studies for Larix gmelinii trees growing on the permafrost region in Siberia

We present results of individual-based root system measurement and analysis applied for Larix gmelinii trees growing on the continuous permafrost region of central Siberia. The data of root excavation taken from the three stands were used for the analyses; young (26 years old), mature (105 years old), and uneven-aged over-mature stand (220 years old). In this article, we highlight two topics: (1) factors affecting spatio-temporal pattern of root system development, and (2) interactions between aboveground (i.e., crown) and belowground (i.e., root) competition. For the first topic, the detailed observation of lateral roots was applied to one sample tree of the overmature stand. The tree constructed a superficial (<30cm in depth) and rather asymmetric root system, and each lateral root expanded mainly into elevated mounds rather than depressed troughs. This indicated that spatial development of an individual root system was largely affected by microtopography (i.e., earth hummocks). For these lateral roots, elongation growth curves were reconstructed using annual-ring data, and annual growth rates and patterns were compared among them. The comparison suggested that temporal root system development is associated with differences in carbon allocation among the lateral roots. For the second topic, we examined relationships between individual crown projection area (CA) and horizontal rooting area (RA) for the sample trees of each stand. RA was almost equal to CA in the young stand, while RA was much larger (three or four times) than CA in the mature and overmature stands. Two measures of stand-level space occupation, crown area index (aboveground: CAI; sum of CAs per unit land area) and rooting area index (belowground: RAI; sum of RAs), were estimated in each stand. The estimates of RAI (1.3-1.8m²m _-2) exceeded unity in all stands. In contrast, CAI exceeded unity (1.3m²m_-2) only in the young stand, and was much smaller (<0.3m²m_-2) in the two older stands. These between-stand differences in RAI-CAI relationships suggest that intertree competition for both aboveground and belowground spaces occurred in the young stand, but only belowground competition still occurred in the two older stands. Based on this finding, we hypothesized that competition below the ground may become predominant as a stand ages in L. gmelinii forests. Methodological limitations of our analysis are also discussed, especially for the analysis using the two indices of space occupation (CAI, RAI).