REGULAR ARTICLE
Linking root morphology, longevity and function to root
branch order: a case study in three shrubs
Gang Huang&Xue-yong Zhao&Ha-lin Zhao&
Ying-xin Huang&Xiao-an Zuo
Received: 16 November 2009 / Accepted: 11 June 2010 / Published online: 13 July 2010 # Springer Science+Business Media B.V. 2010
Abstract Root branching order supports a powerful approach to understanding complex root systems; however, how the pattern of root morphological characteristics, tissue carbon (C) and nitrogen (N) concentrations, and root lifespan are related to anatomical features of variable root orders for mature shrubs (∼19 years old) in sandy habitats is still unclear. In this study, these relationships were investigated for three typical shrubs in Horqin Sand Land, Northeast China. Root diameter, individual root length, tissue carbon concentration, C:N ratio, root lifespan, root cross-sectional area (CSA), stele CSA, proportion of stele in root CSA, mean xylem vessel CSA and the number of xylem vessels all increased with root order for the three shrubs, while specific root length and nitrogen concentration decreased with
root order. The combined root biomass of the first two orders accounted for more than 63% of the first– fourth order root biomass for all the three shrubs. Proportion of stele to root CSA and number of xylem vessels of third-order root segments were significantly higher than that of the first two orders, and third-order roots showed secondary development with a contin-uous cork layer. All first-order and most second-order roots exhibited primary development, had an intact cortex, a lower proportion of stele to root CSA, and a smaller number of vessels. Our research suggests that the first two order roots of shrubs in sandy habitats are responsible mainly for absorption, and that they play a major role in root turnover and C and N flux in the soil organic matter pool due to their high proportion of biomass and N concentration, as well as their short lifespan.
Keywords C/N ratio . Minirhizotron . Nitrogen concentration . Root biomass . Root diameter . Vascular tissues
Introduction
Roots are very sensitive to environmental changes, and have become a fundamental subject of study for the comprehensive understanding of carbon allocation patterns and processes, especially given the current background of global climate change (Gill and Jackson
2000; Trumbore and Gaudinski2003; He et al.2004).
DOI 10.1007/s11104-010-0466-3
Responsible Editor: Tibor Kalapos. G. Huang (*)
Fukang Station of Desert Ecology,
Key Laboratory of Oasis Ecology and Desert Environment, Xinjiang Institute of Ecology and Geography,
Chinese Academy of Sciences, Urumqi 830011, China e-mail: [email protected]
G. Huang
:
X.-y. Zhao:
H.-l. Zhao:
Y.-x. Huang:
X.-a. ZuoNaiman Desertification Research Station, Cold and Arid Regions and Environmental and Engineering Research Institute, Chinese Academy of Sciences, 730000 LanZhou, China
In the past few decades, great progress has been made in understanding C allocation to roots, loss of C via roots, nutrient concentration of roots, and root longevity (Vogt et al.1986; Gower et al.1992; Knute and James1992; Hendrick and Pregitzer1993; Fahey and Hughes1994; Schoettle et al.1994; Wells et al. 2002), whereas the conclusions available on the relationship between these factors are still uncertain and even contradictory. To date, researchers estimating carbon sequestration or release face four main questions: (1) what kinds of roots tend to live for a shorter time; (2) do roots undertaking different functions possess different life histories; (3) what is the correlation between seques-tration and release; and, last but not least, (4) what factors are responsible for root longevity? In order to clearly address these issues, researchers need to consider the diversity of root types and their structure and function, before building a reliable and conven-tional category system corresponding to root function that will increase our ability to accurately measure root longevity.
Recent studies have found that root branching order exhibits high morphological heterogeneity (Pregitzer et al. 2002), and different order roots might have significantly different physiological activities, chemical composition and function (Pregitzer et al.1997; Guo et al.2008a,b,c). Characterizing roots by branching order has been considered a useful approach to identifying root function within root systems (Pregitzer et al. 2002; Guo et al. 2008c). Traditional categories defined according to root diameter mask the constructional and functional discrepancies among root groups. For instance, Withington et al. (2006) found that the median root longevity of Quercus robur was 1.6 times larger than that of Acer pseudoplatanus, although they had the same root diameter (D= 0.46 mm). In contrast, the diameter of distal roots varied with species, but almost all distal roots had the same function (Guo et al. 2008c). There is some research showing that branching order is closely related to root function and longevity (Wang et al. 2006); however, quantitative information on root lifespan of different order roots, relating to differences in root anatomical features, is still lacking. Specifically, different functional categories of rootstock could be distinguished according to their anatomical features. We do not know whether roots performing different functions have different root lifespans and nutrient concentrations or, more importantly, whether such
differences in root traits of functionally different root groups would induce different contributions to soil carbon (C) and nitrogen (N) storage and cycling.
In order to recognize general relationships between root traits and root order, we need to investigate morphological, chemical and anatomical root characteristics in relation to root order. How-ever, to date, most studies on this topic are from forest ecosystems (Pregitzer et al. 2002; Guo et al.
2008c), and quantitative information from other ecosystems, especially shrubs from sandy habitats, are still lacking. Salix gordejevii Chang et Skv. (Salicaceae), Artemisia halodendron Turcz. ex Bess (Asteraceae), and Caragana microphylla Lam. (Fabaceae), are three native shrubs in the Horqin Sand Land in northeastern China, with an obviously different life form from shrubs found in forest ecosystems. We do not know whether the changing pattern of root morphology, tissue nutrient content and longevity related to root order is consistent with that of species from forest ecosystems, or, more importantly, which branch order can be used as a functional distinguishing criteria in shrub rootstock. The objective of this study was to examine how root morphological characteristics, N and C concen-tration, anatomical features and longevity of the three native shrubs were related to root branch order. We expected that the root system of the three shrubs adapted to sandy habitats would show a root order pattern similar to that reported for trees. Previous studies have proved that functional variations in transport and nutrient uptake are associated with anatomical traits based on branching order (Guo et al.
2008b, c; Valenzuela-Estrada et al. 2008), and mor-phological features showed a consistently changing trend with root order. Thus, we also intended to distinguish functional groups in roots of each shrub according to changing patterns of those traits with branch order.
Materials and methods Study site description
The study was conducted at the Naiman Station of Desertification and Farmland Research (NSDF), Chinese Academy of Sciences, in the eastern part of Inner Mongolia (42°58′N, 120°43′E; 360 m a.s.l.).
The soil, climate and vegetation have been reported in detail by Zhao et al. (2007).
Desertification is one of the most serious problems faced by arid and semi-arid regions. To rehabilitate desertified land and restore ecosystems, many artifi-cial measures, such as planting windbreak shrubs, fencing grass land and placing sand arresters, have been implemented in parts of the Horqin Sand Land in northeastern China since the mid-1970s (Zhao et al.
2007). S. gordejevii, A. halodendron and C. micro-phylla were planted widely in this area due to their strong wind-erosion protection ability. Top pruning has been performed several times to allow shrub plantation regeneration. Currently, a stable shrub-grass vegetation community has been established, and the three shrub species are distributed widely in the study area. In July 2006, three 50 m×50 m plots were selected for each plant, with the distance between two plots being more than 100 m.
Root excavation and dissection
At the growth peak of the aboveground plant parts (in July) in 2008, three plants for each species were selected randomly in each plot. Roots of the selected plants were excavated 5 cm away from the base with a sampling area of 20×50 cm2; the sampling depth was from 10 cm to 20 cm below the soil surface. The excavated roots and soil were packed in plastic bags, and all root segments were rinsed and cleaned with deionized water (4°C) after being transported to the laboratory. Roots were dissected into four branch orders with a 10×stereomicroscope (Motic SMZ-140, Xiamen, China) following the morphometric approach developed by Fitter (1982) and used widely by other researchers (Pregitzer et al. 2002; Guo et al. 2008c; Valenzuela-Estrada et al. 2008). The distal roots located at the end of the well-branched fine root system were defined as first-order roots; the root in which two first order roots linked was classified as a second-order root, and so on. In our samples, because not all soil blocks contained complete root orders, only the first four orders were analyzed. In addition, broken root segments were designated based on similarity in diameter and length between segments and complete root branch samples.
Root diameter and length were measured using CIAS software (CID-Bioscience, Camas, WA) for each order of roots after they were scanned on a high
resolution flat-bed scanner (1,200 dpi resolution, 19-μm pixel size, 256-level grayscale, TIFF format; Epson Scanner Perfection 4490, http://www.epson.com). Specifically, first-order roots of S. gordejevii were extremely fine and short. It was thus difficult and time-consuming to distinguish all of them from collected samples, so we chose at least ten sub-samples (with different quantities of root segments) of first-order roots to measure root diameter and length of each root segment under a compound microscope at 10×. After root diameter and length measurement, root mass was measured after roots were oven-dried at 68°C for 48 h, and the specific root length (SRL, m g−1) was estimated by dividing root length by root mass. Root length density (RLD, g m−2) of each order roots was calculated as root dry mass divided the area of sampling block (20 × 50 cm2). Due to difficulty in distinguishing first-order roots of S. gordejevii from whole root samples, first- and second-order roots of the three shrubs were lumped together, and the RLD of the combined first- and second-order roots was measured. Root N and C concentration measurements
Around 3–5 g roots for root morphological analysis were used to determine root N and C concentrations. Root N concentration was analyzed by micro-Kjeldahl analysis (UDK140 Automatic Steam Distill-ing Unit, Automatic Titroline 96, Italy). Root C concentration was determined by dichromate oxida-tion [Walkley-Black method (Nelson and Sommers
1982)].
Root anatomy
In August 2008, roots of each species were collected from three randomly chosen shrubs located separately in the three sampling plots, roots were dissected according to method developed by Fitter (1982). More than 20 root segments for each root order were fixed in formalin-acetic acid-alcohol (FAA) solution (90 ml 70 % ethanol, 5 ml 100 % glacial acetic acid, 5 ml 40% formalin). Root segments were then dehydrated in an ethanol series and embedded in paraffin. Finally, sections of 8 μm thick were prepared for analysis (de Neergaard et al. 2000; Guo et al. 2008c). For each root segment, cross sections were selected from root tip to about 500 μm away
from the root tip. All root cross sections were stained with the safranine-fast green method (Guo et al.2008c). These sections were measured for anatomical features, and photographed under a light microscope with a digital camera (Motic 3000). The cross-sectional areas (CSA) of roots, steles and xylem vessels, as well as the vessel number, were measured, and the proportion of stele in root CSA was also calculated.
Root longevity
In July 2006, three mature shrubs with well developed canopies were selected for each species at each plot. A minirhizotron tube (clear plexiglass, 1 m long, 5 cm inner diameter) was installed for each shrub at an angle of 45°; the installed tube was 20 cm away from the base where 95% of the roots were distributed within a circle of 20 cm from the base, and the inserted depth was 60 cm. Images were taken at intervals of 20–30 days for each tube using a mini cylindrical scanner (CID-Bioscience, Camas, WA) from 13 April to 10 September 2007 and from 10 April to 10 September 2008. Because the air temperature was below the equipment limit and the soil was also frozen from October 2007 to March 2008, no observations were taken during this period. The mini cylindrical scanner is 33 cm in length and 5 cm in diameter and can provide a full colorized image (21.59 W×19.56 L cm) at 200 DPI. We used it to scan roots intersecting the minirhizotron wall at intervals of 20 cm along the tube; each tube was scanned three times. Images were analyzed by the CIAS root image analysis system, and root length, root order and diameter of new roots present on the minirhizotron screen were recorded and monitored until complete disappearance.
Root survival time was defined as the period from the first appearance of a root until its disappearance, which comprises root lifespan (root birth to death) and root decomposition time (root death to disappearance); roots living beyond the last observation timepoint were considered right censored. During the observation period, more than 92% of the roots appeared at the beginning of winter, and they were still alive in spring, root ages at death used in analysis ignored the inactive period of winter (174 days) (Andersson and Majdi 2005), which made the median longevity an unsatisfactory estimate of mean longevity. All root survival times
throughout the monitoring period were pooled, and the proportion of roots surviving as a function of their age was described by the Kaplan-Meier survival function, which is a non-parametric statistical technique that allows estimation of median lifespan (ML; Tierney and Fahey2001; West et al.2003). The log-rank test was used to determine the difference in fine root longevity between orders (Tierney and Fahey2001). During the whole observation period, only the first three orders of roots were well represented on the minirhizotron screens, and we combined the first- and second-order roots as one category due to the difficulty in discriminating between them. Survivorship curves and root lifespan of the combined order roots and third-order roots for each species were also calculated. Statistics
All data were managed and analyzed using SPSS software (SPSS, ver. 13.0 for Windows, Chicago, IL). Root CSA, stele CSA, mean xylem vessel CSA, numbers of xylem vessels, SRL, root tissue nitrogen concentration, root tissue carbon concentration and C: N ratio were transformed logarithmically to meet assumptions of normality and equal variance across root orders. Differences in these variables among root order and species were analyzed by two-way ANOVA procedure, and Tukey’s HSD test was applied to paired comparisons among root orders within each species, and among species within each order. All statistical tests were considered significant at P<0.05.
Results
Root morphology and biomass distribution
Mean root diameter, mean root length, SRL and root biomass density all differed significantly among root branch orders for the three shrub species (Table 1). Mean diameter of the three species increased pro-foundly with root order (Table 1). Root diameters of the first two orders were not significantly different, while root diameters of the third- and fourth-order roots were both significantly thicker than the first two orders. Root diameter of each order roots showed significant difference among species, and S. gordejevii had the thinnest roots among the three shrubs for the first three orders (Table1).
Mean root length also varied with branching order and species (Table 1), increasing systematically with ascending of root order for the three shrubs, except the fourth-order roots of A. halodendron, which were shorter than the third-order. Notably, for A. halodendron and C. microphylla, mean root length of first-order roots was significantly shorter than that of second-order roots, but for S. gordejevii, no significant difference in mean root length was seen between first- and second-order roots (Table1).
SRL decreased with ascending root order, and significance appeared across the first three orders for all shrubs (Table1). SRL decreased significantly from 191.96 m g–1 in first-order roots to 0.35 m g–1 in fourth-order roots of S. gordejevii, from 16.32 m g−1 to 0.52 m g−1 for A. halodendron, and from 25.04 m g−1to 0.16 m g−1for C. microphylla. When comparing among species within the same root order, S. gordejevii always had the largest SRL, with the exception of fourth-order roots (Table1).
The total root biomass density of the first four orders was 50.31 g m−2, 70.17 g m−2and 77.87 g m−2 for S. gordejevii, A. halodendron and C. microphylla, respectively. In terms of the same order roots, root biomass density had no significant difference among
species. Root biomass density varied with root order, and was significantly higher for combined first- and second-order roots than in the other two orders (Table1). The combined root biomass of the first two orders accounted for more than 63% of the total root biomass in the first four orders for the three shrubs; third- and fourth-order roots only took up 10–36%. Based on specific root length, the proportion of total length of the first four orders taken up by the first two orders was more than 80%.
Root C and N concentration, and C:N ratio
On the whole, root C concentration increased with root order for the three shrubs, especially for A. halodendron; the C concentrations of the third- and fourth-order were significantly higher than those of the first two orders, while no significant difference occurred among root orders for the other two species (Fig.1a). In contrast to C concentration, root N concentration decreased with root order, and differed significantly among root orders for S. gordejevii and C. microphylla (Fig.1b). The C:N ratio increased with root order for the three species, and differed significantly among root orders for S. gordeje-vii and C. microphylla (Fig.1c).
Order Salix gordejevii Artemisia
halodendron Caragana microphylla Root diameter (mm) 1 0.14±0.01 Aa 0.36±0.24 Ab 0.41±0.29 Ac 2 0.23±0.01 Aa 0.64±0.05 Ab 0.66±0.05 Ab 3 0.61±0.03 Ba 1.74±0.16 Bb 2.03±0.46 Bc 4 6.91±0.31 Ca 3.35±0.18 Cb 10.23±1.22 Cc
Individual root length (mm)
1 0.73±0.05 Aa 91.35±8.5 Ab 37.70±1.85 Ac
2 6.86±0.59 Aa 248.12±44.8 Bb 211.25±15.47 Bb
3 776.3±20.44 Ba 944.00±116.3 Cb 523.52±34.12 Cc
4 3198.2±68.2 Ca 208.84±1.97 Bb 2082.66±132.08 Da
Specific root length (m g−1)
1 191.96±23.46 Aa 16.32±1.17 Ab 25.04±4.03 Ac
2 48.79±5.99 Ba 8.93±1.22 Bb 5.52±3.68 Bc
3 4.08±0.52 Ca 2.26±0.78 Cb 0.51±0.18 Cc
4 0.35±0.04 Cab 0.52±0.04 Ca 0.16±0.02 Cb
Root biomass density (g m−2)
1, 2 34.28±3.70 Aa 63.02±11.80 Aa 49.65±19.90 Aa
3 6.92±2.26 Ba 2.18±0.47 Ba 20.43±8.52 Aa
4 9.11±3.29 Ba 4.97±1.07 Ba 7.79±3.42 Ba
Table 1 Root morphologi-cal and biomass parameters among root orders within species. Data were pooled for each plot and values represent means and their associated standard errors. Different capital letters indicate significant difference among root orders within each species (P<0.05), different lowercase letters indicate significant difference among species within each root order (P<0.05)
Root lifespan
Survival curves differed among the three shrubs (Fig.2). S. gordejevii had the longest root longevity among the three shrubs in terms of the first two orders. Median root lifespan changed consistently with root order for the three shrubs, it was 34, 113 and 100 days for the combined first two orders, and 47, 198 and 221 days for third-order roots of A. halodendron, S.
gordejevii and C .microphylla, respectively . The median root lifespan was significantly longer for the third-order than lower orders (Fig.2).
Root anatomy
First-order roots exhibited primary development, and the cortical layers were continuous and complete. In contrast, third-order roots had secondary development,
0 100 200 300 400 500 Root C (mg/g) S.gordegevii A.halodendron C.microphylla
a
Aab Aab Bb Ba Aa Ab Aa Ab Aa Ab Aa Aa 0 8 16 24 Root N (mg/g)b
Aa Ba Ca Ca Ab Ab Bb Bc Aa Aa Aa Ab 1 2 3 4 0 50 100 150 200 Root C:N ratio Root orderc
Aab Aab ABa Ba Ab Ab Bb Bb Aa Aa Aa Ab Fig. 1 Total root carbon (C)concentration, root nitrogen (N) concentration and C:N ratio for the first four order roots of three shrubs. Error bars Standard error of the mean (n=9). Different capital letters indicate significant difference among root orders within each species (P<0.05), different lowercase letters indicate significant difference among species within each order (P<0.05)
and the cortex disappeared completely. Secondary xylem development and continuous cork layer occurred in third-order roots for all investigated individual root. A part of the second-order roots also showed secondary xylem development, with a proportion of 37%, 12% and 25% for A. halodendron, S. gordejevii and C. micro-phylla, respectively.
Root CSA, stele CSA, mean xylem vessel CSA, proportion of stele in root CSA and the number of xylem vessels increased with root order for the three shrubs (Table2). All these parameters differed signif-icantly among root orders for A. halodendron and S. gordejevii, as well as between the first two orders and the third-order for C. microphylla. No difference of root CSA, stele CSA, mean xylem vessel CSA and number of xylem vessels was observed between first-and second-order for C. microphylla (Table2).
Discussion
Root anatomy and function classification
Anatomical features are linked to root morphological characteristics, and chemical and physiological activi-ties (lifespan) both showed systematic changes with root order (Pregitzer et al. 2002; Guo et al. 2004, 2008c). Cortical cells took up a large proportion of the whole root cross-sections for first-order roots, which made first-order roots have a higher metabolic rate and require a higher nitrogen concentration to support their physiological activities (Lux et al. 2004). First-order roots of all three shrubs have a higher N concentration than higher-order roots, and this supported their anatomical traits. Moreover, cortical cells were dam-aged or lost easily under stressed conditions (Brundrett 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 0 50 100 150 200 250 0.00 0.25 0.50 0.75 1.00 1st and 2nd order (113d,n=3471) 3rd order (198d,n=887)
(a)
S.gordegevii 1st and 2nd order (34d,n=2641) 3rd order (47d,n=632)(b)
A.halodendron Proportion Surviving Proportion Surviving Proportion SurvivingTime(growing season days)
1st and 2nd order (100d,n=1210) 3rd order (221d,n=980)
(c)
C.microphyllaFig. 2 Root survivorship curves from minirhizotron data for first- to third-order roots of three shrubs. Median life spans and root number used for analysis are indicated in parentheses. Data are for all roots observed through minirhizotron windows from a total of nine plants for each species. Log-rank test showed that first- and second-order roots had shorter root lifespans than third-order roots for all three shrubs
2002; Wells and Eissenstat2003; Soukup et al.2004), thus implying that first-order roots had the shortest lifespan and were more sensitive to environmental changes. In our study, lifespan of the combined first-and second-order roots was significantly shorter than the third-order roots, which also partly supports the conclusion mentioned above. In contrast, third- and higher-order roots were composed mainly of secondary vascular tissues. Secondary vascular tissues limited root absorptive ability and lowered root nitrogen concentration, but prolonged root lifespan due to the protection of a continuous cork layer (Eissenstat and Achor1999; Brundrett2002).
Previous studies have proved that functional variations in transport and nutrient uptake are associ-ated with anatomical traits based on branching order (Guo et al.2008c; Valenzuela-Estrada et al.2008). In our study, primary development occurred mainly in the first two orders, while secondary development occurred in all root segments of the third-order. This suggested that roots could be divided into two functional groups, one for absorption, and the other for transport or storage. The first-order and most of the second-order roots had intact cortex, a low proportion of stele in root CSA, a small number of
xylem vessels and no sign of secondary growth— they served for absorption. In contrast, the third-order roots had no cortex, but had a high proportion of stele in root CSA and a large number of xylem vessels—these traits limit uptake capacity, and these roots serve primarily for transport or storage. For instance, the number of vessels increased markedly from 13, 12 and 11 in the first-order roots to 294, 590 and 333 in the third-order roots for S. gordejevii, A. halodendron and C. microphylla, respectively (Table2), and this suggests that hydraulic transport capacity increases dramati-cally from first-order to third-order roots.
Root morphology, chemical characteristics and lifespan
Root branching is a very efficient organizing con-struction for root systems, advancing nutrient and water absorption efficiency and optimizing C distri-bution in root systems by making roots with different morphologies fulfill different functions (Kenrich
2002). Previous studies have found that certain root morphological parameters (root diameter, length of individual root and specific root length, etc.) and
Order S. gordejevii A. halodendron C. microphylla
Root CSA (μm2) 1 40,121.43±2,438.28 Aa 40,395.05±3,481.62 Aa 62,817.96±4,457.92 Ab 2 34,220.95±3,799.29 Ba 66,120.9±3,717.39 Bb 195,187.86±4,497.84 Aa 3 476,936.80±12,434.86 Ca 358,438.6±6,601 Cb 875,964.69±114,435.5 Bc Stele CSA (μm2) 1 8,842.58±985.70 Aa 10,877.16±1,538.91 Aa 17,667.65±1,676.14 Ab 2 22,730.60±3,957.93 Ba 32,983.97±1,320.63 Bb 60,660.77±1,682.31 Ac 3 390,450.30±10,099.11 Ca 195,295.90±9,533.96 Cb 763,119.10±7,810.69 Bc Mean xylem vessel CSA (μm2)
1 25.78±4.29 Aa 63.83±5.30 Ab 81.57±10.47 Ac
2 85.76±8.78 Ba 108.72±21.67 Bb 94.91±8.31b Aa
3 255.74±10.27 Ca 132.26±10.3 Cb 275.93±19.09 Ba
Proportion of stele in root CSA
1 0.22±0.02 Aa 0.25±0.02 Aa 0.25±0.01 Aa
2 0.69±0.01 Ba 0.49±0.02 Bb 0.27±0.01 Ac
3 0.81±0.00 Ca 0.54±0.01 Cb 0.87±0.00 Ba
No. of xylem vessels (n)
1 13±1 Aa 12±1 Aa 11±1 Aa
2 47±3 Ba 67±7 Bb 40±4 Aa
3 294±16 Ca 590±57 Cb 333±24 Cc
Table 2 Root anatomical characteristics among root orders for three shrub species. Data are means and their standard errors. Different capital letters indicate significant difference among root orders within each species (P<0.05), different lowercase letters indicate significant difference among species within each root order (P<0.05). CSA Cross-sectional area
chemical characteristics correlated well with root order (Eissenstat et al. 2000; Hishi and Takeda
2005; Wang et al. 2006; Guo et al. 2008c). For instance, in a report on nine North American trees, Pregitzer et al. (2002) showed that the diameter and mean length of individual roots increased while SRL decreased within the first three root orders. Both Pregitzer et al. (1997) and Guo et al. (2004) found root N concentration increased with root order in trees. In agreement with the present study, another study on an ericoid shrub (Valenzuela-Estrada et al.
2008) showed that shrubs followed the same rule with regards to the relationship between root morphology, chemical characteristics and branching order. We assume that the relationship of root order with root morphology and chemical properties could be univer-sal for woody species, despite the marked differences in morphologies and habitats among species. In addition, root diameter, organic C concentration, N concentration and C:N ratio were similar in first- and second-order roots, differing significantly from third-and fourth-order roots. Because high N concentration indicates high metabolic activity and C sink strength (Pregitzer et al. 1997), roots of the first two orders could be considered as the major organ for water and nutrient absorption. This partly proved that the first two order roots play different roles from those of high-order roots for shrubs in sand land, and is consistent with a root functional classification based on anatomical variance across root order.
The relationship between the C cost of constructing and maintaining roots and the benefits roots provide for plant growth is tightly linked to SRL. Studies on SRL have been reported for many woody species. For instance, Shi et al. (2008) studied root morphology across the first five orders of 20 hardwood tree species, and showed that SRL of the first-order roots ranged from 32.71 m g−1 to 203.83 m g−1. In a mesic temperate forest in North America, the SRL of the finest two order roots of 25 coexisting woody species ranged from 10.9 m g−1to 115.1 m g−1 (Comas and Eissenstat2009). We also found that shrubs had higher SRL values than trees in the woody species investi-gated above. Valenzuela-Estrada et al. (2008) reported a very high SRL (831 m g−1) for first-order roots of a highbush blueberry. In our study, SRL was 191.96, 16.32 and 25.04 m g−1for S. gordejevii, A. haloden-dron and C. microphylla, respectively. Compared to SRL of shrubs in the mesic habitat, shrubs adapted to
the sandy habitat had lower SRL. This was in agreement with the conclusion that species living in a resource-poor ecosystem are characterized by low SRL, high root tissue density and low absorptive capacity (Eissenstat 1992; McCrady and Comerford
1998; Taylor and Peterson 2000; Comas et al. 2002; Hishi and Takeda 2005). However, roots of S. gordejevii had significantly thinner diameter and larger SRL than the other two shrubs and even some forest trees. This might be explained by the microsite that S. gordejevii inhabits; this shrub is generally settled down in the lowland between dunes and the shifting sand land where soil water content is high and their growth could be favored.
Yanai et al. (1995) modeled the relationship of root diameter and nutrient uptake rate with root longevity, and reported that the optimal root for nutrient absorption was infinitely fine. Some previous research has shown the diameter of individual fibrous root to be positively correlated with root longevity (Eissenstat et al. 2000), while our results do not totally support this. In root order level, the first two orders of roots had significantly shorter lifespans than third-order roots for all three shrubs due to their smaller diameters, while at the species level, the relationship between root diameter and lifespan was inconsistent with branch order. In the first instance, root diameter was similar between A. halodendron and C. microphylla (0.64 vs 0.66 for the second-order; 1.74 vs 2.03 for the third-order), but root lifespan of C. microphylla was about three times that of A. halodendron. In the second instance, the root diameters of the investigated branch-ing orders of S. gordejevii were lowest in the three shrubs, while their lifespans were considerably longer than those of A. halodendron.
Numerous studies have proved that leaf longevity is inversely correlated with nitrogen concentration for a wide range of species (Reich et al. 1992). Valenzuela-Estrada et al. (2008) also found that low-order roots (Vaccinium corymbosum L.) had low C:N ratio, and their lifespan was shorter than that of high-order roots. Although the relationship between N concentration and root lifespan related to root order was consistent within a shrub, this was not true among species within the same order roots. The N concentration of the first two order roots of S. gordejevii was slightly higher than A. halodendron, but root lifespans of S. gordejevii were 3.32-fold higher than that of A. halodendron. This might result
from different growth strategies and special root functions for species. For instance, as a nitrogen fixer, C. microphylla had the highest nitrogen con-centration in the three species, but this does not follow the hypothesis that roots with higher nitrogen concentration should have a shorter lifespan.
Biomass allocation and C and N returning by different functional roots
The first two order roots shared the major part of the total root biomass density of the first four orders, and root biomass density did not systematically increase with ascending root order. This is in line with some previous research, for example, biomass of first-order roots was greater than that of second-order roots for nine North American temperate trees (Pregitzer et al.
2002) and longleaf pine (Guo et al.2004). Wang et al. (2006) reported that the first two order roots of two temperate tree species (Larix gmelinii and Fraxinus mandshurica) represented up to 32–45% of total root biomass of the five orders combined. In an ericoid root system, Valenzuela-Estrada et al. (2008) also found that lower-order roots (first three orders) comprised most of the root biomass in a five order root system. Wang et al. (2006) deduced that a high branching ratio between the first two orders was responsible for the relatively higher biomass propor-tion of the first two orders in total root biomass. Due to heterogeneity of root distribution in vertical and horizontal directions, our sampling area might not completely reflect root distribution for the whole root system. In particular, biomass of the higher-order roots was underestimated in the present study. However, root biomass density of the first two orders was similar to fine root (diameter <1 mm) biomass density estimated by soil cores (Zhao1994) and, according to diameter variation with order, the first two order roots simply represented the majority of the fine roots.
The morphological, chemical and lifespan differ-ences among branching orders in our study suggest that different order roots play different roles in ecosystem carbon and nutrient cycling (Valenzuela-Estrada et al.
2008). Consistent with previous studies, the first two orders roots of the three shrubs were significantly shorter and thinner than third- and fourth-order roots, but the first two order roots accounted for more than 60% of total root biomass, and had the highest nitrogen concentration and shortest lifespan compared with
third-and fourth-order roots. Through calculating the amount of organic C and N return by the first two orders based on root C, N concentration, root biomass density and turnover rate (reciprocal of lifespan), we found the organic C values ranged from 33.58 g m−2 year−1 to 245.42 g m−2year−1, and nitrogen values ranged from 0.93 g m−2year−1to 5.06 g m−2year−1. These amounts are slightly higher than extra C and N returning to soil via root mortality of sweetgum (Iversen et al. 2008). This resulted mainly from the quick turnover rate of the three species investigated, especially for A. halodendron, whose lifespan was even shorter than that of some grasses (Watson et al. 2000; Gill et al.
2002). In addition, the absorptive roots (first two order roots) accounted for more than 84% of organic carbon and 87% of nitrogen input to soil of the first three orders, while higher-order roots contributed only a small proportion due to their low biomass, low nitrogen concentration and long lifespan.
In summary, root anatomical, morphological and chemical features were investigated in terms of root order for three shrubs inhabiting sandy habitats. The results indicated that first- and second-order roots were quite similar and served mainly absorption, while third-and higher-order roots had secondary development, third-and they served primarily conduction, anchorage or storage. Although root diameter, SRL, N concentration, C:N ratio, root lifespan and anatomical features all changed systematically with ascending of root order, the first two order roots did not exhibit any significant difference in morphological and chemical properties. The first two order roots had thinner root diameter, higher SRL value and N concentration, and shorter lifespan than higher-order roots, which is consistent with a functional classification based on anatomical characteristics. In addition, the first two order roots accounted for more than 60% of the first-four order root biomass for the three shrubs. Considering the short lifespan and high nitrogen concentration of the first two order roots, the absorptive roots (first two orders) accounted for a profound proportion of the C and N inputs to soil in the whole root system.
Acknowledgments We are grateful to two anonymous reviewers for valuable comments on our manuscript. We also appreciate our colleagues and several other research workers at the NMDS for constructive criticism and help. This study was supported financially by the Chinese National Natural Science Foundation (2009CB421303-4, 40871004 and 30972422), and we also appreciate this financial support.
References
Andersson PI, Majdi H (2005) Estimating root longevity at sites with long periods of low root mortality. Plant Soil 276:9–14 Brundrett MC (2002) Coevolution of roots and mycorrhizas of
land plants. New Phytol 154:275–304
Comas LH, Eissenstat DM (2009) Patterns in root trait variation among 25 co-existing North American forest species. New Phytol 182:919–928
Comas LH, Bouma TJ, Eissenstat DM (2002) Linking root traits to potential growth rate in six temperate tree species. Ecophysiology 132:34–43
de Neergaard E, Lyshede OB, Gahoonia TS, Care D, Hooker JE (2000) Anatomy and histology of roots and root-soil boundary. In: Smit AL, Bengough AG, Engels C, Noordwijk M, Pellerin S, Geijn SC (eds) Root methods: a handbook. Springer, Berlin, pp 33–74
Eissenstat DM (1992) Costs and benefits of constructing roots of small diameter. J Plant Nutr 15:763–782
Eissenstat DM, Achor DS (1999) Anatomical characteristics of roots of citrus rootstocks that vary in specific root length. New Phytol 141:309–321
Eissenstat DM, Wells CE, Yanai RD, Whitbeck JL (2000) Building fine roots in a changing environment: implica-tions for root longevity. New Phytol 147:33–42
Fahey TJ, Hughes JW (1994) Fine root dynamics in a northern hardwood forest ecosystem, Hubbard Brook experimental forest. J Ecol 82:533–548
Fitter AH (1982) Morphometric analysis of root systems: application of the technique and influence of soil fertility on root system development in two herbaceous species. Plant Cell Environ 5:313–322
Gill RA, Jackson RB (2000) Global patterns of root turnover for terrestrial ecosystems. New Phytol 147:13–31 Gill RA, Burke IC, Lauenroth WK, Milchunas DG (2002)
Longevity and turnover of roots in the shortgrass steppe: influence of diameter and depth. Plant Ecol 159:241–251
Gower ST, Vogt KA, Grier CC (1992) Carbon dynamics of Rocky Mountain Douglas-fir: influence of water and nutrient availability. Ecol Monogr 62:43–65
Guo DL, Mitchell RJ, Hendricks JJ (2004) Fine root branch orders respond differentially to carbon source-sink manipulations in a longleaf pine forest. Oecologia 140:450–457
Guo DL, Li H, Mitchell RJ, Han W, Hendricks JJ, Fahey TJ, Hendrick RL (2008a) Fine root heterogeneity by branch order: exploring the discrepancy in root turnover estimates between minirhizotron and carbon isotopic methods. New Phytol 177:443–456
Guo DL, Mitchell RJ, Withington JM, Fan PP, Hendricks JJ (2008b) Endogenous and exogenous controls of root life span, mortality and nitrogen flux in a longleaf pine forest: root branch order predominates. J Ecol 96:737– 745
Guo DL, Xia M, Wei X, Chang W, Liu Y, Wang Z (2008c) Anatomical traits associated with absorption and mycor-rhizal colonization are linked to root branch order in twenty-three chinese temperate tree species. New Phytol 180:673–683
He JS, Wang ZQ, Fang JY (2004) Underground ecology with global changes: problems and prospects (in Chinese). Chin Sci Bull 49:1226–1233
Hendrick RL, Pregitzer KS (1993) The dynamics of fine root, length, biomass, and nitrogen content in two northern hardwood ecosystems. Can J For Res 23:2507–2520 Hishi T, Takeda H (2005) Dynamics of heterorhizic root
systems: protoxylem groups within the fine-root system of Chamaecyparis obtuse. New Phytol 167:509–521 Iversen CM, Ledford J, Norby RJ (2008) CO2 enrichment
increases carbon and nitrogen input from fine roots in a deciduous forest. New Phytol 179:837–847
Kenrich P (2002) The origin of roots. In: Waisel Y, Eshel E, Kafkafi U (eds) Plant root: the hidden half. Dekker, New York, pp 1–15
Knute JN, James WR (1992) Fine root production estimates and below-ground carbon allocation in forest ecosystems. Ecology 73:1139–1147
Lux A, Luxova M, Abe J, Morita S (2004) Root cortex: structural and functional variability and responses to environmental stress (in Japanese). Root Res 13:117–131 McCrady RL, Comerford NB (1998) Morphological and anatomical relationships of Loblolly pine fine roots. Trees 12:431–437
Nelson DW, Sommers LE (1982) Total carbon, organic carbon and organic matter. In: Page AL, Miller RH, Keeney DR (eds) Methods of soil analysis. ASA, Madison, pp 539– 577
Pregitzer KS, Kubiske ME, Yu CK, Hendrick RL (1997) Relationships among root branch order, carbon, and nitrogen in four temperate species. Oecologia 111:302– 308
Pregitzer KS, DeForest JL, Burton AJ, Allen MF, Ruess RW, Hendrick RL (2002) Fine root architecture of nine North American trees. Ecol Monogr 72:293–309
Reich PB, Walters MB, Ellsworth DS (1992) Leaf life-span in relation to leaf, plant, and stand characteristics among diverse ecosystems. Ecolo Monogr 62:365–392
Schoettle AW, Fahey TJ, Shoettle AW (1994) Foliage and fine root longevity in pines. Ecol Bull 43:136–153
Shi W, Wang ZQ, Liu JL, Gu JC, Guo DL (2008) Fine root morphology of twenty hardwood species in Mao-er Shan natural secondary forest in northeastern China (in Chinese). J Plant Ecol 32:1217–1226
Soukup A, Malá J, Hrubcová M, Kálal J, Votrubová O, Cvikrová M (2004) Differences in anatomical structure and lignin content of roots of pedunculate oak and wild cherry-tree plantlets during acclimation. Biol Planta 48:481–489
Taylor JH, Peterson CA (2000) Morphometric analysis of Pinus banksiana Lamb. root anatomy during a 3-month field study. Trees 14:239–247
Tierney GL, Fahey TJ (2001) Evaluating minirhizotron esti-mates of fine root longevity and production in the forest floor. Plant Soil 229:167–176
Trumbore SE, Gaudinski JB (2003) The secret lives of roots. Science 302:1344–1345
Valenzuela-Estrada LR, Vera-Caraballo V, Ruth LE, Eissenstat DM (2008) Root anatomy, morphology, and longevity among root orders in Vaccinium corymbosum (Ericaceae). Am J Bot 95:1506–1514
Vogt KA, Grier CC, Sprugel DG, Vogt DJ (1986) Overestima-tion of net root producOverestima-tion: a real or imaginary problem? Ecology 67:577–579
Wang ZQ, Guo DL, Wang XR, Gu JC, Mei L (2006) Fine root architecture, morphology, and biomass of different branch orders of two Chinese temperate tree species. Plant Soil 288:155–171
Watson CA, Ross JM, Bagnaresi U, Minotta GF, Roffi F, Atkinson D, Black KE, Hooker JE (2000) Environment-induced modifications to root longevity in Lolium perenne and Trifolium repens. Ann Bot 85:397–401
Wells CE, Eissenstat DM (2003) Beyond the roots of young seedlings: the influence of age and order on fine root physiology. J Plant Growth Regul 21:324–334
Wells CE, Glenn DM, Eissenstat DM (2002) Changes in the risk of fine-root mortality with age: a case study in peach, Prunus persica (Rosaceae). Am J Bot 89:79–87
West JB, Espeleta JF, Donovan LA (2003) Root longevity and phenology differences between two co-occurring savanna bunchgrasses with different leaf habits. Funct Ecol 17:20–28 Withington JM, Reich PB, Oleksyn J, Eissenstat DM (2006) Comparisons of structure and life span in roots and leaves among temperate trees. Ecol Monogr 76:381–398 Yanai RD, Fahey TJ, Miller SL (1995) Efficiency of nutrient
acquisition by fine roots and mycorrhizae. In: Smith WK, Hinkley TM (eds) Resource physiology of conifers: acquisition, allocation, and utilization. Academic, New York, pp 75–103
Zhao AF (1994) Morphology, distribution and dynamics of root systems of Artemisia halodendron and Caragana micro-phylla (in Chinese). Grassland of China 3:15–19 Zhao HL, Zhou RL, Su YZ, Zhang H, Zhao LY, Drake S (2007)
Shrub facilitation of desert land restoration in the Horqin Sand Land of Inner Mongolia. Ecol Eng 31:1–8
