Fine root sampling is a highly intricate and challenging process, particularly for deep soils. The task of separating and morphologically scanning fine roots from the soil requires an enormous amount of effort. Consequently, deep root studies have been conducted only in a few regions globally, such as the Loess Plateau, the sandy regions of northern China, and south-western Australia (Drake et al., 2011; Li et al., 2019; Nan et al., 2019; Song et al., 2020; Wang et al., 2021; Zhou et al., 2019). Most studies have been limited to sampling depths above 1 m of soil layer (Finér et al., 2007; Finér et al., 2011; Jagodzinski et al., 2016; Yuan and Chen, 2012). In order to obtain deep (1–6 m) fine roots, our study was conducted with fine root sampling only at the end of the growing season. To investigate the seasonal dynamics of fine roots, micro root tubes provide a more convenient sampling method than root augers (Coleman and Aubrey, 2018). However, the use of micro root tubes does not allow for observations of roots at deeper soil depths. Although we conducted only one sampling during the growing season, the data we collected could reflect the general vertical distribution pattern of fine roots in that year. Specifically, our findings indicate that fine roots gradually decreased with increasing soil depth, and a higher abundance of fine roots was present in the shallow layer. This distribution pattern is representative of the global fine root distribution pattern, as shown in previous studies (Jackson et al., 1996; Zou et al., 2022). However, our investigation highlights that this distribution pattern is only representative of water uptake during specific times of the growing season. This finding underscores the significance of directly examining root uptake. Unfortunately, we were unable to establish the relationship between actual absorbing roots and root water uptake. This limitation is due to the fact that the commonly defined fine root diameter (≤ 2 mm) far exceeds that of absorbing roots (McCormack et al., 2015; Pregitzer et al., 2002). Therefore, our fine root samples contain a mix of large-diameter transport roots and absorbing roots.
Understanding the relationships between plant structure or morphology, function, and environment is an enduring area of ecological research (Freschet et al., 2021; Sack and Buckley, 2020). Researchers have long recognized that roots can alter resource uptake based on resource availability (Javot and Maurel, 2002; Lauter et al., 1996). However, in most field-based studies, root length density is commonly measured as a representative of water uptake function. For instance, research aimed at the subdivision of subsurface ecological niches relies on evaluating root length density at varying depths. This approach suggests that root distribution can reflect the ability to absorb resources (Brassard et al., 2011; Brassard et al., 2013; Sun et al., 2017; Zeng et al., 2021). Our results showed that fine root distribution did not match systematically the water uptake pattern over the whole growing season (Fig. 4). Recently, stable isotope methods have been increasingly used to determine root water uptake patterns (Dawson et al., 2020; Rothfuss and Javaux, 2017; Volkmann et al., 2016; Yang et al., 2017). The simultaneous use of an isotope in-situ measurement system and a liquid flow measurement system enables monitoring of root uptake patterns with high temporal resolution (Gessler et al., 2022; Kühnhammer et al., 2020). The resultant data could be directly used to develop temporally dynamic eco-hydrological models, rather than using root and/or soil water distribution data (Kühnhammer et al., 2020; Kulmatiski et al., 2017; Mazzacavallo and Kulmatiski, 2015; Seeger et al., 2020). In our current study, root water uptake patterns were determined using the stable isotope technology, and then were linked with fine root or soil water distribution. The results obtained in this study will provide data support and reference value for future development of related models.
When water deficit occurred at shallow soil layer, water uptake in deeper or wetter soil layers increased, resulting in mismatched distribution of fine roots and root water uptake patterns (Green and Clothier, 1999; Jarvis, 1976; Šimůnek and Hopmans, 2009; Thomas et al., 2020). Our results showed that the matching degree between the water uptake pattern and fine root distribution increased with soil water content significantly in high-density stands, but this correlation was insignificant in stands with low-density stand structures (Fig. 6A, B, C, D). This suggests that soil water content should be the primary factor responsible for variations in the degree of matching between water uptake patterns and fine root distribution in stands with high-density structures. However, other factors such as soil texture, soil compaction, and intra-forest interactions also contribute to the degree of matching in stands with low-density structures. When roots move to areas with higher soil moisture content due to water stress, the water uptake pattern may become more closely aligned with the distribution of soil water (Ellsworth and Sternberg, 2015; Gao et al., 2018; Kühnhammer et al., 2020; Kulmatiski et al., 2017). In the present study, however, the matching degree between the water uptake pattern and soil water distribution was found to increase gradually with increasing water stress (i.e., decreasing soil water) only in young stands (Fig. 7A, B, C, D). In contrast, in mature stands, the water uptake pattern deviated more from the soil water distribution pattern as water stress increased (Fig. 7A, B, C, D). This may be due to the fact that deeper soil water in mature stands has been over-consumed during stand development resulting in low soil water potential at deep soil layers (Liu et al., 2022), this can also be demonstrated by our soil water data (Fig. 2). It would be more difficult for mature stands to uptake water from deeper soil layers with increasing water stress (Christina et al., 2017; Davidson et al., 2011; Gardner, 1991; Yang et al., 2017), causing the stands to change their root uptake strategy and absorb water from relatively shallow soil layers alternatively.
In general, as the soil water content increased, the root water uptake pattern in the young stands tended to be closer to the fine root distribution pattern and far away from the soil water distribution, while the matching of the root water uptake pattern with both the fine root and soil water distributions became closer in the mature stands (Fig. 6; Fig. 7). Therefore, our first hypothesis was partially supported by the results.
The matching degree between water uptake pattern and soil water distribution was higher than the matching degree between the water uptake pattern and fine root distribution in mature stands (Fig. 5), suggesting that water uptake pattern matched more closely with soil water distribution than with fine root distribution in mature stands. However, the water uptake pattern matched closely with both soil water distribution and fine root distribution in young stands (Fig. 5). This evidence indicates that water uptake functions in young stands are driven by both root structure and available water resources, while water uptake in mature forests with larger and more complex root system is better associated with soil water availability than root structure. Mature trees have deeper roots and more complex root system than young trees (Coleman and Aubrey, 2018; Finér et al., 2007; Geng et al., 2022; Yuan and Chen, 2010, 2012). Therefore, mature trees are able to absorb soil water from deeper, wetter, and more stable water sources to meet transpiration requirements (Drake et al., 2011; Feild and Dawson, 1998; Finér et al., 2007; Geng et al., 2022; Huo et al., 2018), resulting in their water uptake pattern being closer to the soil water distribution and farther away from the fine root distribution. Given the stand age, the matching degree between the water uptake pattern and fine root distribution was lower in high-density stands than in low-density stands during the growing season, but the difference in the matching degree between the water uptake pattern and soil water distribution was minimal (Fig. 5). This indicates that the relationship between root water uptake pattern and fine root distribution is more affected by stand structure compared to the relationship between root water uptake pattern and soil water distribution.
Previous studies have found that root uptake patterns may be closer to soil water distribution patterns than to fine root distribution patterns (Kühnhammer et al., 2020; Kulmatiski et al., 2017). This holds true only for mature stands in our study. However, in young stands, our results showed that root water uptake patterns closely matched both soil water distribution and fine root distribution patterns, as indicated by their high matching degrees with little difference (Fig. 5). The reason for the variability between the current and previous results may be due to the fact that the material of previous study was herbaceous perennial while this study was based on trees in stands, as well as their neglect of the seasonal dynamics of root uptake. Additionally, it is interesting that these two matching degrees were significantly related in high-density stands but positively in the mature stand and negatively in the young stand (Fig. 8). In low-density stands, regardless of young or mature stands, their correlations were negative, although not significant (Fig. 8). This intriguing result implies that while the fine root distribution drives root uptake, the soil water distribution also drives root uptake in a synergistic or trade-off manner.
In addition to the variables related to soil water content, other factors that influence the matching degree between water uptake pattern and fine root or soil water distributions were also identified (Fig. 6; Fig. 7). For example, we found that higher soil water heterogeneity (as represented by the coefficient of variation of soil water content across the soil profile) significantly decreased the matching degree between the water uptake pattern and fine root distribution in the mature high-density stand (Fig. 6E) and the matching degree between the water uptake pattern and soil water distribution in the mature low-density stand (Fig. 7E). Previous studies have shown that the soil water heterogeneity was the cause of compensatory root water uptake, which changed the matching degree between root water uptake and fine root distribution or soil water distribution (Ellsworth and Sternberg, 2015; Green and Clothier, 1999). Additionally, of the three meteorological factors, only temperature was found to have a significant inhibitory effect on the matching degree between fine root distribution and root water uptake in the mature low-density stand. This indicates that the meteorological factors have a relatively weak modulation on the matching degree of between water uptake pattern and fine root or soil water distribution, compared to the direct effect of soil water content. That is, our second hypothesis was also supported by the results.