Environmental Science
articles by Will Kemp

Research proposal: Measuring root respiration to locate the growing tips of tree roots

1 Aims

The aim of this project is to assess the viability of using soil respiration measurements to locate the outer extent of tree roots. There is currently no non-invasive method for doing this. It also aims to add to the understanding of soil respiration and its relationship with tree roots. A further aim is to assess the tree protection zone defined in AS4970 (Protection of trees on development sites) and determine its adequacy.

2 Background and knowledge gaps

There has been a considerable amount of research conducted into soil respiration over two centuries, and numerous studies have attempted to isolate its components – which include respiration by roots and organisms in the rhizosphere (Luo & Zhou, 2006).

Root respiration is considered to make up approximately half of total soil respiration, although this figure varies considerably with vegetation type, season, and geography (Luo & Zhou, 2006; Chen et al., 2010; Makita et al., 2012). Most of the remaining root respiration is produced by soil organisms and a small amount is contributed by oxidation of organic matter (Luo & Zhou, 2006; Kaur et al., 2010). A considerable proportion of total soil respiration comes from plant roots and their associated micro-organisms (Chen et al., 2009).

Soil respiration is sensitive to temperature – increasing as the soil gets warmer (Luo & Zhou, 2006). However, it appears that the two main components – root respiration and soil organism respiration – are affected to different extents, with root respiration being most strongly affected (Högburg, 2010; Kaur et al., 2010). Soil respiration is also sensitive to soil moisture content, highest in moderately moist soil and lowest in very wet or very dry soil (Adachi et al., 2009) – although the relationship between moisture and soil respiration is not straightforward (Luo & Zhou, 2006, Adachi et al., 2009).

Levels of soil respiration vary throughout the day, being lowest at night and highest around midday, which is correlated with changes in soil temperature (Luo & Zhou, 2006). Adachi et al. (2009) also found a depression around midday during the dry season in a tropical forest in Thailand, which correlates with midday depressions found in studies in other parts of the world, and is probably due to a time lag between photosynthesis and root respiration (Bekku et al., 2009).

An important component of soil respiration is that produced by soil organisms in the rhizosphere, particularly those organisms which are able to use the carbohydrates and other substances exuded by roots (Luo & Zhou, 2006; Chen et al., 2009). It is difficult to measure the effects of belowground processes (Valverde-Barrantes, 2007) and, therefore, to differentiate between root respiration and the respiration of associated soil organisms (Makita 2009), however a number of studies have found that the rate of root respiration is higher in finer roots than in coarser ones (Chen, 2009; Chen 2010; Makita, 2012).

Desrochers et al. (2002) found root respiration to be as much as 3.5 times higher in the fine roots of Populus tremuloides than in its coarse roots. Makita et al. (2012) found coarse root respiration rates between 0.025 and 5.6 nmol CO2 g−1 s−1 and fine root respiration rates between 0.52 and 24 nmol CO2 g−1 s−1 (based on dry root weight) in tropical rainforest trees.

In general, soil respiration rates appear to be positively correlated with fine root production and dependent on the traits of particular species (Valverde-Barrantes, 2007). Makita et al. (2012) found that most of the differences between species are explained by the diameter and tissue density of the roots, and that most of the interspecies variations in fine root respiration was explained by tissue density. The higher respiration rates found in finer roots can be explained by the physiological characteristics of the different types of roots (Makita et al., 2012).

Trees, particularly in urban environments, are important assets which produce a wide range of benefits for people and the environment. A city’s trees provide benefits which can be valued in the millions of dollars per year (Moore, 2009). The root systems of urban trees, however, are often implicated in damage caused to buildings and infrastructure (Ow & Sim, 2012). They also need to be protected from damage by construction and development (Standards Australia, 2009).

Locating and mapping tree roots without damaging them is not a simple matter and X-ray computed tomography was initially considered as a solution to this problem. However, it was not found to be practical outside the laboratory and was replaced by ground-penetrating radar (GPR) (Ow & Sim, 2012). GPR can determine the location, depth and size of objects buried in soil and it may be used for non-invasive mapping of tree roots. However, its resolution is limited (Hirano et al., 2009; Ow & Sim, 2012). Clay content and soil moisture can affect the resolution of GPR, but it has been used to identify tree roots down to about 50 mm diameter in a clay loam and 19 mm diameter in sand. (Hirano et al., 2009; Ow & Sim, 2012). It is not capable of detecting fine roots (Ow & Sim, 2012).

Australian standard AS4970 (“Protection of trees on development sites”) specifies a method for defining a “tree protection zone” (TPZ) around trees which may be affected by construction work. The TPZ is specified as twelve times the diameter of the tree, and any work that encroaches on that zone may require non-destructive root investigation. Specified non-destructive methods are “pneumatic, hydraulic, hand digging or ground penetrating radar” (Standards Australia, 2009). Of these, only ground penetrating radar is non-invasive – the others disturb the roots in some way and may be detrimental to the health of the tree. This leaves the arborist the option of either guessing at the topography of the fine roots, or digging them up to find out.

A tree’s fine roots are generally mostly concentrated at the outer edge of the root zone (Kozlowski, 1971) and one implication of the studies which show a correlation between soil respiration and the presence of fine tree roots is that an increase in soil respiration around the outer edge of a tree’s root zone could be observable. I found evidence to suggest this may be the case in field work conducted in far north Queensland in 2013. However, we did not collect enough data to come to any reliable conclusion (Kemp, unpubl.) and there appears to be a lack of literature exploring how soil respiration varies with distance from the base of a tree.

If it can be shown that the outer edge of the root zone may be detected by changes in soil respiration along a transect radiating from the base of a tree, this method could be employed to fill in the gap left by GPR, allowing arborists to accurately and non-invasively determine the true extent of the root zone. This study will evaluate the potential of such a method by conducting field tests on a range of tree species in a range of different soils to find out if a readily identifiable peak in soil respiration does, in fact, occur at the farthest extent of the root system.

If it is found that this method is able to reliably locate the outer edge of the root zone, the study will continue to measure the root zones of a range of trees and use the data collected to develop a model of the relationship between the distance from the trunk and the level of soil respiration. It will also evaluate to what extent the TPZ defined in AS4970 correlates with the actual diameter of the root zone.

3 Importance of the project

Amenity trees bring many important benefits to people and the environment – particularly in urban environments, where they may be crucial to long term sustainability (Moore, 2009). Encroachment of construction work on tree root zones can terminally damage trees and must be prevented. There is currently no non-invasive method for accurately determining the outer limits of root growth and the development of a relatively simple method of achieving this could benefit arborists and the trees they care for.

Although soil respiration has been studied for 200 years (Luo & Zhou, 2006) there are still some gaps in knowledge. Accurately modelling soil carbon flux is becoming increasingly important as part of the drive to understand the effects of global climate change and develop potential amelioration measures. Understanding spatial variation in soil respiration and other factors influencing its rate is critical to development of those models (Luo & Zhou, 2006; Adachi et al., 2009; Chen et al., 2010; Bekku et al., 2011).

Trees play a major role in the global carbon cycle and are important reservoirs for atmospheric carbon (Stephenson et al., 2014). However, they are also a major contributor to soil respiration in many ecosystems, and their roots influence respiration rates over considerable areas of soil. Understanding how soil respiration varies over the root zone of a tree may help increase the accuracy of soil respiration and carbon budget models.

4 Hypotheses

I hypothesise that there is a clearly detectable peak in soil respiration at the outer edge of a tree’s root zone when compared with other points along a transect radiating outwards from the base of the trunk. Current knowledge suggests this may be the case.

5 Methodology

5.1 Procedure

As this study is attempting to find a general pattern, it will be conducted in as wide a range of locations as possible. Soil respiration rates will be tested at 1 m intervals along a number of transects radiating from the base of individual trees. The trees selected for testing and the number of transects around each tree will depend on the surrounding vegetation. The minimum number of transects per tree will be two, with five transects being the preferred number. The transects will be spaced as widely as the surrounding vegetation allows, keeping them as far away from other trees as possible.

The aim is to study the effect of tree roots on soil respiration rate, so an absence of other trees or shrubs whose proximity to the transect is likely to affect the results is essential. In order to reduce the variability introduced by types of ground cover that may or may not be present at other sites, grass and leaf litter will be cleared from the transects, leaving bare soil, 24 hours before soil respiration rates are measured.

The study will be conducted in two phases. The first phase will be a preliminary study to find out if the hypothesised respiration pattern can be detected. The preliminary study will be carried out on twenty trees of different species at different locations around Adelaide. If the hypothesised respiration pattern is found to occur in more than one of these trees, the second phase will be carried out with a larger number of trees in as wide a number of locations as possible.

Suitable locations (open forest, areas with widely spaced trees) will be found and suitable trees will be identified, marked, and given numbers. In order to attempt to eliminate bias and randomise the sample, five of the suitable trees will be selected using the Android “Random number generator” app (by Muse Guy Productions). Suitable trees will be defined as any tree whose canopy is at least half the canopy diameter away from the canopy of another tree for at least half the canopy circumference, where there are no large shrubs under the canopy. If there are less than ten suitable trees in a location, half the suitable trees will be randomly selected.

If a respiration peak is found as hypothesised, a 300 mm x 300 mm x 300 mm hole will be dug at the position on the transect where the peak was found, and the roots found there will be inspected to determine if they are fine tree roots. This hole will be dug with care and any roots thick enough to leave in place will be left undamaged.

At each site, an equal number of areas without trees will be selected using the same selection method used for the trees themselves. These areas will be used as a control. They will be approximately the same diameter as the mean canopy diameters of the trees being tested, and would meet the specifications for suitable trees if there was a tree in the centre. Measurements will be taken on transects radiating out from the center of this area to the average tree canopy radius.

5.2 Duration

This study will be conducted over a six month period in order to collect data at different times of year, with different soil temperatures.

5.3 Sampling

Soil CO2 flux will be measured at 1 m intervals from the base of trees, along transects radiating from the trunk. Angles between transects will be approximately equal and only sited where influence of other trees or shrubs is at a minimum. Up to 5 transects will be measured at each tree. Sample points will only be measured once, although several gas samples will need to be taken for each measurement.

The diameter of the canopy and trunk diameter at breast height (DBH, measured at 1.4 m above the soil surface) will be recorded for each sample tree.

5.4 Methods

Soil CO2 flux will be measured using a portable infrared or photoacoustic gas analyser, depending on availability.

5.5 Analysis

As CO2 flux levels vary considerably with temperature, season, time of day, soil moisture, ground cover, soil type, tree size, and tree species, etc., absolute values for CO2 flux will have no intrinsic meaning. To find and compare patterns for different trees and locations etc., the data will be normalised for each transect – giving a set of results between 0 and 1, where 1 = the maximum CO2 flux measured on that transect.

As tree sizes vary considerably too, the distance from the base of the tree to any CO2 flux peak will also be normalised in two different ways – as a proportion of the radius of the canopy and as a proportion of the tree’s DBH.

The tree results will be compared with the control results to establish whether or not there is a peak in respiration that corresponds with the outer extent of the tree roots and does not occur where there are no trees.

5.6 Materials

A portable gas analyser and power supply (generator or battery/inverter), vehicle for travel, tape measure.


Adachi, M., Ishida, A., Bunyavejchewin, S., Okuda, T., & Koizumi, H. (2009). Spatial and temporal variation in soil respiration in a seasonally dry tropical forest, Thailand. Journal of Tropical Ecology, 25, 5, 531-539.

Bekku, Y., Sakata, T., Nakano, T., and Koizumi, H. (2009). Midday depression in root respiration of Quercus crispula and Chamaecyparis obtusa: its implication for estimating carbon cycling in forest ecosystems. Ecological Research, 24, 4, 865-871.

Chen, D., Zhou, L., Rao, X., Lin, Y., and Fu, S. (2010). Effects of root diameter and root nitrogen concentration on in situ root respiration among different seasons and tree species. Ecological Research, 25, 5, 983-993.

Desrochers, A., Landhäusser, S.M., and Lieffers, V.J. (2002). Coarse and fine root respiration in aspen (Populus tremuloides). Tree Physiology, 22, 10, 725-732.

Hirano, Y., Dannoura, M., Aono, K., Igarashi, T., Ishii, M., Yamase, K., Naoki, M. and Kanazawa, Y. (2009). Limiting factors in the detection of tree roots using ground-penetrating radar. Plant & Soil, 319 1/2, 15-24.

Högberg, P. (2010). Is tree root respiration more sensitive than heterotrophic respiration to changes in soil temperature? New Phytologist, 188, 1, 9-10.

Kaur, K., Jalota, R. K., & Midmore, D. J. (2010). Soil respiration rate and its sensitivity to temperature in pasture systems of dry-tropics. Acta Agriculturae Scandinavica, Section B – Soil & Plant Science, 60, 5, 407-419.

Kozlowski, T.T. (1971) Growth and Development of Trees – Volume II, Cambial Growth, Root Growth, and Reproductive Growth. New York, USA: Academic Press Inc.

Luo, Y. and Zhou, X. (2006). Soil Respiration and the Environment. Burlington, USA: Elsevier.

Makita, N., Hirano, Y., Dannoura, M., Kominami, Y., Mizoguchi, T., Ishii, H., and Kanazawa, Y. (2009). Fine root morphological traits determine variation in root respiration of Quercus serrata. Tree Physiology, 29, 4, 579-585.

Makita, N., Kosugi, Y., Dannoura, M., Takanashi, S., Niiyama, K., Kassim, A.R., and Nik, A.R. (2012). Patterns of root respiration rates and morphological traits in 13 tree species in a tropical forest. Tree Physiology, 32, 3, 303-312.

Moore, G.M. (2009). Urban trees: worth more than they cost. Proceedings of the 10th National Street Tree Symposium 2009. Glen Osmond, SA: Treenet.

Ow, L.F. and Sim, E.K. (2012). Detection of urban tree roots with the ground penetrating radar. Plant Biosystems, 146, sup1, 288-297.
Standards Australia (2009). AS4970-2009: Protection of trees on development sites. Sydney, NSW: Standards Australia.

Stephenson, N.L., Das, A.J., Condit, R., Russo, S.E., Baker, P.J., Beckman, N.G., Coomes, D.A., Lines, E.R., Morris, W.K., Ruger, N., Alvarez, E., and 27 others (2014). Rate of tree carbon accumulation increases continuously with tree size. Nature, 507, 7490, 90-93.

Valverde-Barrantes, O.J. (2007). Relationships among litterfall, fine-root growth, and soil respiration for five tropical tree species. Canadian Journal of Forest Research, 37, 10, 1954-1965.