Land-use effects on soil properties on the north-western slopes of New South Wales: Implications for soil condition assessment.
Abstract. In Australia, as elsewhere, there is a growing need for information relating to soil condition, its current status, and the nature and direction of change in response to management pressures. This information is required by land managers, and regional, State and, national agencies to inform modified land-use practices and investment to maintain and improve the soil resource. Here, we present data relating to soils under 3 land-use types at 6 properties the north-western slopes of New South Wales. We aimed to quantify the range of soil condition states that exist across the region and to test a range of potential soil condition indicators and their suitability to detect differences in soil condition between these land-use types. A range of soil properties showed no significant difference between land-uses and could be rejected as indicators. However, significant differences existed between the land-uses and soil depths for a range of the other soil parameters determined (bulk density, C, N, P, EC, and Na). Soil C, N, P, and Na concentration and total soil C were typically higher in woodland soils compared with other land-uses, while bulk density, pH, and EC were lower in the woodland soils. The depth at which these differences existed varied between soil parameters. Correlation and principal components analysis suggested that a minimum dataset of soil parameters including soil bulk density, pH, C, P, and Na would discern much of the difference in soil condition between the land-uses studied. It is proposed that these parameters be used as a minimum dataset of indicators for soil condition assessment on soils of the type across this region. Work continues under the New South Wales Land and Soil Condition Monitoring Program to further refine the selection of appropriate soil indicators in this and other regions of New South Wales.Additional keywords: soil condition, soil monitoring, soil indicators, NW Slopes, NSW.
Introduction
In Australia, as elsewhere, there is a growing need for information relating to soil condition, its current status, and the nature and direction of change in response to management pressures. This information is required by land managers, and regional, State, and national agencies to inform modified land management practices and investment to maintain and improve the soil resource. The concepts and definitions of soil health, quality, and condition are still being debated in the literature (e.g. Doran 1996; Karlen et al. 1997) and no consensus has yet been reached regarding the soil properties that might be most useful in defining and evaluating change in any or all of these. In the absence of such consensus, there is a need to at least identify those soil properties that most accurately reflect change in soils associated with a range of land-use pressures. Such information will be the foundation on which the evaluation of soil quality, health, or condition will be made.
A great deal of information currently exists in the scientific literature regarding the impact of specific management practices on soil properties in various regions of Australia. For example, in northern New South Wales (NSW) a large amount of information exists on the physical, chemical, and biological properties of soils under a range of farming and grazing practices (Harte 1984; Packer et al. 1998; Chan 2001; Greenwood and McKenzie 2001; Chan et al. 2002; Farquharson et al. 2003; Lodge et al. 2003a, 2003b; Harms et al. 2004; Zhang et al. 2007), while others have considered less intensively managed systems in this region such as native woodland and forests (Chilcott et al. 1997; Jackson and Ash 2001; Prober et al. 2002a, 2002b; Wilson 2002; Graham et al. 2004; Collard and Zammit 2006; Wilson et al. 2007). More recently, soil carbon status and storage under a range of these land-uses and management alternatives have also begun to attract attention (Hulugalle 2000; Chan et al. 2003; Young et al. 2005). These studies have employed a range of methodological approaches and examined a wide range of soil properties, but few have examined the relative effects on soil condition of a replicated suite of land-use types prevailing across the region, including relatively undisturbed sites. There is, however, an immediate need for information of this type to inform land management decisions and investment, to guide land use change for natural resource management, and to optimise soil condition outcomes.
Monitoring is now being promoted regionally, nationally, and internationally to assess and evaluate soil condition for the purposes of reporting and to prioritise investment in natural resource management. There is, as yet, no agreed process for soil monitoring, or an agreed suite of indicators by which to evaluate soil change. A defined, and limited, suite of soil condition indicators is urgently required in order to detect and report on the status and change in soils at a range of scales. A wide range of potential soil 'indicators' has been proposed for soil condition assessment (e.g. McKenzie et al. 2002; Sparling et al. 2002; Dalai et al. 2003; Schwenke et al. 2003; Commonwealth Government 2006) and typically includes indicators of physical, chemical, and biological soil condition. Few empirical studies in Australia have evaluated these, across a range of land-uses, to determine which indicators might be regionally appropriate and sensitive to differences between land-uses and management types.
Here, we present data relating to soils under 3 specific land-use types that are currently dominant across the north-western slopes of NSW to determine the ways in which their characteristics differ. In so doing, we seek to quantify the range of soil condition states that exist in this area and to test a range of potential soil condition indicators and their suitability to detect differences in soil condition between these land-use types. The information thus generated is intended to help inform on the practical assessment of soil condition and land-use options for these soil types in this region, to assist the prioritisation of investments in management actions to achieve desired soil condition outcomes.
Methods
Site descriptions
A series of 'site clusters' was established on 6 properties near the township of Bingara on the north-western slopes of NSW (Fig. 1). Each 'site cluster' was located on a separate property. In this region, mixed farming is typical, with the dominant land-uses being cultivation for cereals, oats, lucerne, etc., grazing of unimproved native pasture, and remnant woodland in isolated fragments, roadside reserves, and Travelling Stock Routes. For this reason, each site cluster consisted of 3 adjacent land-uses: (i) a cultivation paddock, (ii) unimproved native grass, and (iii) remnant woodland. Each of the individual land-uses was in an adjacent paddock located on common soil type, landscape position, aspect, and slope. Each land-use within a cluster was a maximum of 200 m from the other land-uses. Details of the locations, characteristics, geology, and soil types for each of the sites are summarised in Table 1.
Cultivation paddocks were selected such that they had each been cultivated, using traditional cropping techniques, for at least 30 years. This involved conventional tillage, no recorded adoption of 'conservation' tillage techniques (e.g. minimum tillage, stubble retention, etc.), and the use of long fallow periods to build soil moisture. Only the cultivation paddocks on 2 properties (Sites 3 and 4) had received recent applications of fertiliser, principally nitrogen and phosphorus. The grass sites on each property were selected in paddocks that had historically been grazed using set-stocking with sheep and cattle (more common in recent years) with minimal fertiliser application, legume addition, or adoption of controlled grazing. The woodland sites selected were relatively undisturbed roadside reserves or Travelling Stock Routes which are only periodically grazed by stock in transit. Only at one property (Property 5) had the woodland been regularly grazed. None of the native grass or woodland sites had any history of fertiliser applications.
Soil sampling
Within each land-use, 9 soil cores were collected from a 20 by 20 m plot in a regular grid pattern using a manual coring device with internal diameter 45 mm. Soil cores were taken to a total depth of 0.50 m wherever possible or as deep as could be collected where soil parent material was reached before this depth. Each soil core was subdivided into depth increments of 0-0.05 and 0.05-0.10 m, and then 0.10-m increments to 0.50 m (or solid rock, whichever was reached first).
Soil analysis
Soil samples were stored in cool, dry conditions until they could be processed (typically <48 h). Soil samples were then oven-dried at 40°C for 48 h, weighed, and an indicative bulk density calculated. We acknowledge that bulk density is normally calculated using oven-dry (105°C) soil weights. However, drying at this temperature yields a soil sample that is not suitable for subsequent analysis (particularly total carbon). Since it was our intention to link bulk densities from individual samples directly with other soil parameters, the lower drying temperature was used. This approach will result in a consistent but small overestimate of soil bulk density. Unpublished data suggest overestimates of <0.01 g/[cm.sup.3] for 0-0.05 and 0.05-0.10 m samples and <0.02g/[cm.sup.3] for 0.10-0.20 and 0.20-0.30 m samples.
Soil samples were crushed and sieved to <2 mm and each individual sample was then analysed for a suite of soil parameters that were intended to reflect physical, chemical, and biological properties of the soil (after Schipper and Sparling 2000; McKenzie et al. 2002). Soil parameters determined were pH (Ca[Cl.sub.2]), electrical conductivity (EC), organic carbon (Walkely and Black), total nitrogen (Kjeldahl), extractable phosphorus (Colwell), cations (ammonium acetate extractable), aluminium (KCI), sulfur (KCI), and chloride. All analyses were carried out at the INCITEC Werribee Laboratory in Melbourne, Victoria. Total soil carbon to 0.30 m depth (t/ha) was derived from soil depth, bulk density, and carbon concentration (%) acknowledging that the values calculated will represent small but consistent overestimates (<1 t/ha) due to bulk density calculations from 40°C dry soils.
Statistical analysis
Differences in soil attributes between land-uses and depths across all sample sites were tested with mixed effects analysis of variance (ANOVA), with land-use and depth defined as fixed factors and property as a random factor. Prior to analyses, histograms of the raw data were used to examine for normal error distributions and data were transformed where appropriate. Significant differences between land-uses at various depths were tested using Tukey's post-hoc tests for each soil attribute. Principal components analysis (PCA) was used to demonstrate similarity of various soil components between land-uses and sites using the average values of each soil attribute for the 0-0.05 m depth for each sampling site and soil type.
Results
Several significant effects were found with land-use and soil depth for the soil properties determined (Table 2). There was a strong (P < 0.005) and significant land-use effect for all soil properties determined with the exception of chloride, calcium, magnesium, and potassium (Table 2). A strong (P < 0.0001) and significant depth effect was also found for all soil properties except chloride, indicating a significant change in the various properties with depth in the soil. There was also a strong land-use x depth interaction for most soil properties (except for bulk density, magnesium, and potassium), indicating that the nature of change in soil properties with depth differed between the 3 land-uses.
Where there was a significant land-use effect (bulk density, pH, carbon, nitrogen, phosphorus, EC, and sodium), Tukey's post-hoc test was used to examine the nature of differences between the soils under different land-uses (Table 3). Soil properties showing no significant land-use effect were omitted from further analysis. Where significant differences existed between the land-uses, values for soil properties (carbon, nitrogen, phosphorus, sodium) were typically higher (P < 0.05) under woodland, while soils under cropping and grass were statistically similar to each another. For soil carbon (%), values were in the order cultivation < grass < woodland (Table 3) at the soil surface. Where they were significantly different between land-uses, bulk density, pH, and EC were lower (P < 0.05) under woodland.
Differences between the soils of the 3 land-uses were, however, found at different depths depending on the soil parameter, which explains the interaction found in the ANOVA analysis. For example, differences in soil bulk density, carbon, and nitrogen between the land-uses were found largely at the soil surface (Table 3, Fig. 2), whereas for the remaining soil parameters, differences were typically found at depth in the soil. Concentrations of carbon and nitrogen were significantly larger under woodland in the surface soil layers (0-0.05, 0.05-0.10, 0.10-0.20 m) compared with the 2 land-uses, which were generally statistically similar. In these surface layers, bulk density was lower under woodland compared with other land-uses. Soil pH differed between the land-uses mainly in the subsurface layers. Here, woodland soils were more acid at all depths below 0.10 m compared with the other land-uses.
Soil EC did not differ a great deal between the various land-uses. Only at 0.40 0.50 m was EC slightly but significantly (P < 0.05) lower under woodland. Soil Na increased under all land-uses with increasing soil depth. Although cultivation and grass sites had statistically similar values, both had higher Na concentrations than woodland in the deeper soil layers (0.30-0.50 m).
For those soil parameters showing a significant land-use effect, a Pearson's correlation matrix was constructed (Table 4). Significant correlations (P < 0.01) were found between most of the soil parameters. Carbon (%) was significantly correlated with most other soil parameters. Total nitrogen (r = 0.84) and extractable phosphorus (r = 0.48) were strongly and positively correlated with carbon. Bulk density was strongly and negatively correlated with carbon (r = -0.46). Soil pH was strongly and positively correlated with EC (r = 0.49) and sodium (r = 0.31) but was negatively correlated with extractable phosphorus (r = 0.39). Other correlations between soil properties, although significant, were weaker.
The projection of variables using a PCA indicated that sample sites were grouped according to land-use using the range of soil properties selected (Fig. 3a). This diagram confirms that woodland soils bad generally higher values for the properties determined compared with the other 2 land-uses. The PCA (Fig. 3b) also indicated that, not only was the separation of land-uses into groups influenced by the relative strength of the various soil properties, but also the direction component in the data collected. There was considerable overlap in the cultivation and grazing data groupings. Woodland was separated from the other 2 land-uses on the Factor I axis by virtually all of the parameters investigated. On the Factor 2 axis, woodland was separated from the other 2 land-uses largely on the basis of carbon, nitrogen, phosphorus, pH, and EC, and most strongly by Na.
Three distinct groups of soil parameters appeared to exist in the data. Bulk density was alone in the top left quadrant of the plot and was oriented in an opposite direction to carbon, nitrogen, phosphorus, and potassium. These patterns support the information generated from the correlation matrix and indicate that, of these latter variables, carbon and nitrogen had the strongest influence. In the top right quadrant of the plot, several variables grouped together including pH, EC, calcium, chloride, and magnesium. Most of these variables had similarly strong influence with the exception of chloride, which was of lesser importance. Sodium also had a strong influence but this appeared to be largely independent of all other variables.
Values for total carbon (t/ha) to 0.30 m depth were derived for the various land-uses using soil depth, bulk density, and carbon concentration (Fig. 4). There was a strong, significant property effect (F = 31.55, P < 0.001), land-use effect (F = 64.55, P < 0.001), and also a strong property x land-use interaction (F = 14.03, P < 0.001) for total soil carbon. Overall, total soil carbon under cultivation and grass were statistically similar but total carbon was significantly higher (P < 0.001) under woodland. This pattern was consistent for all properties studied except for properties 2 and 5, where little difference existed in total carbon between the land-uses.
Discussion
A number of the soil parameters examined showed strong land-use effects. However, chloride, calcium, magnesium, and potassium showed little difference between land-uses on the soils studied. These latter parameters would therefore seem to be of limited utility in distinguishing between soils under different land-use types in this environment.
For those properties that did show a strong land-use effect, soil carbon, nitrogen, phosphorus, and sodium were all higher under woodland, while soil bulk density, pH, and EC were lower under woodland, compared with the other land-uses. The greater accumulation of carbon and nitrogen under trees compared with other land-uses would seem to be a common feature in this environment and is well documented. For example, Chilcott et al. (1997), Murphy et al. (2002), Young et al. (2005), and Collard and Zammit (2006) have all observed higher organic carbon concentrations in soils under woodland than equivalent cleared sites and that soil organic carbon is rapidly lost following the removal of trees in this environment. Higher concentrations of carbon and nitrogen in the near surface layers under woodland reflect the larger quantity of biomass input to the soil surface under woodland, slower rates of decomposition, and hence organic matter accumulation.
Other work has reported higher concentrations of phosphorus under trees than cleared sites in this environment where soils have not been fertilised (Chilcott et al. 1997; Wilson 2002; Graham et al. 2004; Wilson et al. 2007). These results would therefore seem to suggest that, overall, soils under trees have higher organic matter and nutrient status with lower bulk density compared with grazed or cultivated soils and represent 'islands' of higher soil quality in this environment in a similar way to that proposed by Dean et al. (1999), among others. What remains unclear is whether the higher concentrations of elements such as nitrogen and phosphorus under trees represent a preferential accumulation under trees, a redistribution of material from deeper soil layers, or simply a conservation of material that has been lost elsewhere. Work is continuing to elucidate these processes.
The depth in the soil at which these differences between land-uses occurred was not consistent between the soil parameters and land-uses. For example, differences in those properties most affected by organic matter status (bulk density, carbon, nitrogen) were largely restricted to the surface layers and differences between the land-uses diminished with increasing soil depth. This pattern suggests that changes in soil properties associated with organic matter are driven largely from the soil surface. For soil pH, EC, phosphorus, and sodium, however, differences were largely restricted to the subsurface layers. These various properties might therefore respond to, and indicate the operation of, a range of processes that are not directly related to organic matter inputs at the soil surface. Soil pH was slightly elevated under woodland at the soil surface, although this was not statistically significant. Wilson (2002), Graham et al. (2004), and Wilson et al. (2007) observed higher pH at the soil surface under trees and concluded that this resulted from a 'biological pumping' process as proposed by Noble et al. (1996), whereby trees draw alkalinity from deep in the soil and deposit this on the soil surface in litter. Wilson et al. (2007) also demonstrated that a balancing soil acidification took place at depth under trees which was not present under pastures. In this current work, soil acidity was indeed higher under the woodland in the deeper soil layers, indicating the probable operation of this process in the sites studied.
It was interesting that under woodland, unlike the other land-uses, soil phosphorus was higher through much of the deeper soil profile. This might reflect the limited removal and hence the effective 'conservation' of soil phosphorus in the woodland soil profile where the soils of the other 2 land-uses have been subject to continued historical removal of phosphorus as a result of agricultural activity. Alternatively, the availability of phosphorus in the woodland profile might be a response to the acidity profile under this land-use, since there was a strong relationship between soil pH and extractable phosphorus across all land-uses.
Soil EC was similar across most land-uses but slightly lower under woodland deep in the soil. Soil sodium, however, increased under all land-uses with increased depth. The higher sodium under trees perhaps reflects the more limited deep drainage under this land-use compared with others. Other work (e.g. Young and McLeod 2001) suggests that chloride is a valuable indicator of subsurface processes such as deep drainage and hence dryland salinity processes. However, no clear land-use effect was found for chloride across the sites that we studied. One possible explanation for this is the dominance of sodium bicarbonate rather than sodium chloride in our region of study. In this instance, sodium content would seem to be a more sensitive indicator of these processes.
When total carbon quantities were determined to 0.30 m across all the land-uses, woodland soils had a significantly larger quantity than cultivation and grass land-uses, with these latter being very similar. However, this relative difference was not consistent across the study sites. At 2 properties (2 and 5), there was no statistical difference between the land-uses. At property 5 (Myall Creek) the woodland had been grazed at an atypically high rate compared with the other woodland sites, which probably resulted in the lower overall carbon content. Removal of biomass through grazing would therefore seem to have a very significant effect on overall soil carbon stock. At property 2, however, the relatively high value for total carbon under cultivation resulted from an unusually high carbon concentration in the 0.05-3.10 m layer compared with other cultivation paddocks. This was consistent across the sample area. Reasons for this result are unclear but might result from specific management actions on this property such as stubble incorporation into this layer increasing soil carbon. This does illustrate that, although some broad patterns can be identified across a range of sites and land-uses, local changes that diverge from this general pattern might result from the actions of specific land managers.
Doran et al. (1994) and Doran and Jones (1996) suggested that soil condition indicators are not useful if they can be estimated from other indicators, are of insufficient precision, or are difficult to interpret. In this study, we found significant correlations between many of the soil parameters, and a limited suite of indicators would probably yield sufficient information to inform on relative soil condition among the land-use types examined. The strongest correlations suggested groupings of the parameters. For example soil carbon, nitrogen, phosphorus, and bulk density were strongly correlated, while strong positive correlations existed separately between pH, EC, and sodium concentration. PCA indicated that these variables could be grouped not only by the strength, but also the direction, of their influence. These groups included one containing carbon, nitrogen, and phosphorus of which carbon had the strongest influence. Another group contained pH, EC, calcium, and magnesium within which each had similar strength of influence. The importance of magnesium to physiological or other processes in this environment is difficult to interpret, while the other properties were strongly correlated. For this reason pH or EC would be sufficient indicators to reflect this group of properties. The 2 other parameters, bulk density and sodium, had strong influence but in independent directions. These results are similar to those of Schipper and Sparling (2000) for soils in New Zealand.
A key challenge facing regional, State, and national agencies is to efficiently and effectively assess and monitor changes in soil condition, but uncertainty still exists as to which indicators might be most appropriate. McKenzie et al. (2002) and Schipper and Sparling (2000) have suggested a list of parameters that might potentially serve as indicators of soil condition. Our results suggest that there are several parameters that can discern differences between land-uses in the environment studied and that different parameters might indicate the operation of distinct sets of soil processes. Due to its strength of influence and correlation with other parameters, soil carbon is undoubtedly a key indicator, a finding that concurs with that of Dalal et al. (2003). However, the other key variables bulk density, pH (or EC), sodium, and phosphorus would also seem to offer useful information with respect to processes that are not directly linked with surface organic inputs. We therefore conclude that the assessment of this limited suite of parameters would indicate much relating to the key differences between soils and dominant land-uses in this region. This suite of parameters would therefore seem to represent an adequate and indicative minimum dataset for soil condition assessment in this region.
Evaluating and interpreting differences in these soil indicators from the perspective of soil quality and health, and predicting the extent to which soil function might be impaired as soil properties reach specific thresholds, is a complex task on which information and consensus are limited. For some soil properties, threshold values are more easily defined than others. For example, it is widely acknowledged that critical soil pH (Ca[Cl.sub.2]) values exist (i.e. pH 5.5 and pH 4.8) below which a progressively stronger inhibition of plant growth takes place (Lockwood et al. 2003). All of the sites we studied had soil pH values well above these critical thresholds. For most other soil properties though, such thresholds are less certain and are complicated by local factors such as climate, parent materials, historical or contemporary management, and the relative stability of soil properties.
Hazelton and Murphy (2007) attempted to identify critical thresholds for a range of soil properties in NSW, including soil carbon. Using their categorisation though, the surface soils of all the sites and land-use types that we studied would be categorised as having high (H1) soil carbon contents. In comparison with other soils across NSW, this assessment might be accurate, but perhaps of greater significance is that the soils we studied in this region varied considerably in their carbon content between land-uses. The implication therefore is that although all the soils studied were high in carbon by NSW standards, some of the soils were in better condition than others. For many soil properties, critical threshold values might therefore be difficult to define in a comprehensive way and will vary locally, regionally, and by specific enterprise. Judgements regarding soil quality and health might therefore need to be made at a more local level and be based, at least initially, on the quantification of a range of values typical of a region, as we have done here.
It is acknowledged that this study was undertaken at only 6 properties across a reasonably large region. However, the properties and sample sites selected were believed to be representative of the region. A study of a wider range of sites, soil types, and land-uses would be necessary to determine a definitive range of suitable indicators. This process has begun under the NSW Land and Soil Condition Monitoring Program, coordinated by the NSW Department of Environment and Climate Change.
Acknowledgments
This work was funded by the NSW Department of Natural Resources (now the Department of Environment and Climate Change), as an early component of its Statewide Land and Soil Condition Monitoring Program. The authors would like to thank the landholders at the properties studied around Bingara for their assistance and for access to study sites. Thanks also to Melinda McHenry and to the anonymous referee for constructive comments on earlier drafts of this article.
Manuscript received 14 December 2007, accepted 5 May 2008
References
Chan KY (2001) Soil particulate organic carbon under different land use management. Soil Use and Management 17, 217-221. doi: 10.1079/ SUM200180
Chan KY, Heenan DP, Oates A (2002) Soil carbon fractions and relationship to soil quality under different tillage and stubble management. Soil & Tillage Research 63, 133-139. doi: 10.1016/ S0167-1987(01)00239-2
Chan KY, Heenan DP, So HB (2003) Sequestration of carbon and changes in soil quality under conservation tillage on lighter-textured soils in Australia. Australian Journal of Experimental Agriculture 43, 325-334. doi: 10.1071/EA02077
Chilcott C, Reid N, King K (1997) Impact of trees on the diversity of pasture species and soil biota in grazed landscapes on the northern Tablelands, NSW. In 'Conservation outside nature reserves'. (Eds P Hale, D Lamb) (Centre for Conservation Biology, The University of Queensland: Brisbane)
Collard SJ, Zammit C (2006) Effects of land-use intensification on soil carbon and ecosystem services in Brigalow (Acacia harpophylla) landscapes of southeast Queensland, Australia. Agriculture, Ecosystems & Environment 117, 185-194. doi: 10.1016/j.agee.2006.04.004
Commonwealth Government (2006) Natural Resource Management Resource Condition Indicators. www.nrm.gov.au/monitoring/indicators/index.html#1ist, Canberra.
Dalal RC, Eberhard R, Grantham T, Mayer DG (2003) Application of sustainability indicators, organic matter and electrical conductivity, to resource management in the northern grains region. Australian Journal of Experimental Agriculture 43, 253 259. doi: 10.1071/EA00186
Dean WRJ, Milton SJ, Jeltsch F (1999) Large trees, fertile islands and birds in arid savanna. Journal of Arid Environments 41, 61-78. doi: 10.1006/ jare. 1998.0455
Department of Mines (1971) Inverell 1:250 000 Geological Series Sheet, Sydney, NSW.
Doran JW (1996) Soil quality and health: the international situation and criteria for indicators. In 'Proceedings, Workshop on Soil Quality Indicators for New Zealand Agriculture'. 8-9 February 1996, Lincoln University, Christchurch, New Zealand.
Doran JW, Coleman DC, Bezdicek DE, Stewart BA (1994) "Defining soil quality for a sustainable environment.' Special Publication 35. (Soil Science Society of America: Madison, WI)
Doran JW, Jones AJ (1996) 'Methods for assessing soil quality.' Special Publication 49. (Soil Science Society of America: Madison, WI)
Farquharson RJ, Schwenke GD, Mullen JD (2003) Should we manage soil organic carbon in Vertosols in the northern grains region of Australia? Australian Journal of Experimental Agriculture 43, 261-270. doi: 10.1071/EA00163
Graham S, Wilson BR, Reid N (2004) Scattered paddock trees, litter chemistry and surface soil properties in pastures of the New England Tablelands, New South Wales. Australian Journal of Soil Research 42, 905-912. doi: 10.1071/SR03065
Greenwood KL, McKenzie BM (2001) Grazing effects on soil physical properties and the consequences for pastures: a review. Australian Journal of Experimental Agriculture 41, 1231-1250. doi: 10.1071/ EA00102
Harms B, Dalai R, Wang W (2004) Soil carbon and soil nitrogen changes after clearing of mulga vegetation. In 'Supersoil. Annual Conference of the Australian Soil Science Society and the New Zealand Soil Science Society'. Sydney, 5-9 December 2004. CD-ROM (ASSSI: Sydney)
Harte AJ (1984) Effect of tillage on the stability of three red soils of the northern wheatbelt. Journal of Soil Conservation 40, 94-101.
Hazelton P, Murphy B (2007) 'Interpreting soil test results: what do all the numbers mean?' (CSIRO Publishing: Melbourne, Vic.)
Hulugalle NR (2000) Carbon sequestration in irrigated Vertisols under cotton-based fanning systems. Communications in Soil Science and Plant Analysis 31, 645-654.
Isbell RF (1996) 'The Australian Soil Classification.' (CSIRO Publishing: Melbourne, Vic.)
Jackson J, Ash AJ (2001) The role of trees in enhancing soil nutrient availability for native perennial grasses in open eucalypt woodlands of north-east Queensland. Australian Journal of Soil Research 52, 377-386.
Karlen DL, Mausbach MJ, Doran JW, Cline RG, Harris RF, Schumann GF (1997) Soil quality: a concept, definition and framework for evaluation. Soil Science Society of America Journal 61, 4-10.
Lockwood PV, Wilson BR, Daniel H, Jones M (2003) 'Soil acidification and natural resource management: directions for the future.' (University of New England: Annidale, NSW)
Lodge GM, Murphy SR, Harden S (2003a) Effects of grazing and management on herbage mass, persistence, animal production and soil water content of native pastures. 1. A redgrass-wallaby grass pasture, Barraba, North West Slopes, New South Wales. Australian Journal of Experimental Agriculture 43, 875-890. doi: 10.1071/ EA02188
Lodge GM, Murphy SR, Harden S (2003b) Effects of grazing and management on herbage mass, persistence, animal production and soil water content of native pastures. 1. A mixed native pasture, Manilla, North West Slopes, New South Wales. Australian Journal of Experimental Agriculture 43, 891-905. doi: 10.1071/ EA02189
McKenzie N, Henderson B, McDonald W (2002) 'Monitoring soil change: principles and practices for Australian conditions.' (CSIRO Land and Water: Canberra, ACT)
Murphy BW, Rawson A, Ravenscroft L, Rankin M, Millard R (2002) Paired site sampling for soil carbon estimation--NSW. Australian Greenhouse Office Technical Report No. 34, Canberra.
Noble AD, Zenneck I, Randall PJ (1996) Leaf litter alkalinity and neutralisation of soil acidity. Plant and Soil 179, 293 303. doi: 10.1007/BF00009340
Packer 1, Medway J, Jones B, Koen T (1998) Are 'conservative' cropping systems improving soil infiltration, organic carbon and bulk density in southern NSW. In 'Proceedings of the 9th Australian Agronomy Conference'. Charles Sturt University, Wagga Wagga, NSW. (ASA and Charles Sturt University: Wagga Wagga, NSW)
Prober SM, Lunt ID, Thieve KR (2002a) Determining reference conditions fur management and restoration of temperate grassy woodlands: relationships among trees, topsoil's and underscore flora in little-grazed remnants. Australian Journal of Botany 50, 687-697. doi: 10.1071/BT02043
Prober SM, Thieve KR, Lunt ID (2002b) Identifying ecological barriers to restoration in temperate grassy woodlands: soil changes associated with different degradation states. Australian Journal of Botany 50, 699-712. doi: 10.1071/BT02052
Schipper LA, Spading GP (2000) Performance of Soil Condition Indicators across taxonomic groups and land uses. Soil Science Society of America Journal 64, 301-311.
Schwenke G D, Reuter D J, Fitzpatrick RW, Walker J, O'Callaghan P (2003) Soil and catchment health indicators of sustainability. Australian Journal of Experimental Agriculture 43, 205 222. doi: 10.1071/ EA01012
Sparling GP, Rijkse W, Wilde H, van der Weerden TJ, Beare MH, Francis GS (2002) Implementing soil quality indicators for land. Landcare Research Contract Report: LC0102/015, Ministry for the Environment, New Zealand.
Wilson BR (2002) The influence of scattered paddock trees on surface soil properties: A study on the Northern Tablelands of NSW. Ecological Management & Restoration 3, 213-221.
Wilson BR, Growns I, Lemon J (2007) Scattered native trees and soil patterns in grazing lands of the Northern Tablelands of N SW, Australia. Australian Journal of Soil Research 45, 199 205. doi: 10.1071/SR07019
Young RR, McLeod MK (2001) Estimation of historic deep drainage under grassed woodland, lucerne pastures and cropped paddocks using the SODICS model. In "Wanted: Sustainable Futures for Saline Lands. Program, Papers & Proceedings of the 7th National PURSL Conference'. Launceston, Tas., 20-23 March 2001. (Eds C Grose, L Bond, T Pinkard) pp. 115-119. (PURSL)
Young R, Wilson BR, McLeod M, Alston C (2005) Carbon storage m soils and vegetation under contrasting land uses in northern New South Wales, Australia. Australian Journal of Soil Research 43, 21-31. doi: 10.1071/ SR04032
Zhang GS, Chan KY, Oates A, Heenan DP, Huang GB (2007) Relationship between soil structure and runoff/soil loss after 24 years of conservation tillage. Soil & Tillage Research 92, 122-128. doi: 10.1016/ j.still.2006.01.006
B. R. Wilson (A,B,E), Ivor Growns (C), and J. Lemon (D)
(A) NSW Department of Environment and Climate Change, PO Box U221, University of New England, Armidale, NSW 2351, Australia.
(B) School of Environmental and Rural Sciences, University of New England, Armidale, NSW 2351, Australia.
(C) NSW Department of Water and Energy, PO Box U245, University of New England, Armidale, NSW 2351, Australia.
(D) NSW Department of Environment and Climate Change, Gunnedah Resource Centre, PO Box 462, NSW 2380, Australia.
(E) Corresponding author. Email: [email protected]
Table 1. Site details for each property including lithology (Department of Mines 1971) and soil type (Isbell 1996) Property Grid reference l. Braemar 29°50'13.27"S, 150°28'46.60"E 2. Dingwall 29°49'22.81"S, 150°27'00.59"E 3. Yeral 29°48'30.74"S, 150°40'25.36"E 4. Pinetrees 29°48'09.32"S, 150°40'38.19"E 5. Myall Creek 29°45'38.33"S, 150°32'09.16"E 6. The Valley 29°52'24.24"S, 150°29'46.27"E Property Altitude Lithology l. Braemar 360m Parry Formation; Carboniferous sediments 2. Dingwall 351 m Parry Formation; Carboniferous sediments 3. Yeral 408 m Whitlow Fomiation; Permian/ Carboniferous Meta-sediments 4. Pinetrees 420 m Whitlow Fomiation; Permian/ Carboniferous Meta-sediments 5. Myall Creek 363 m Parry Formation; Carboniferous sediments 6. The Valley 384 m Parry Formation; Carboniferous sediments and Tertiary Basalt colluvium Property Soil type Fertiliser application l. Braemar Red Chromosol None 2. Dingwall Red Chromosol N+P to cultivation only 3. Yeral Red Chromosol N+P to cultivation only 4. Pinetrees Red Chromosol N+P to cultivation only 5. Myall Creek Red Chromosol None 6. The Valley Red Chromosol/ None Dermosol Table 2. F values and significance values from mixed effects ANOVA * P<0.005; *** P<0.0001; n.s., not significant Land-use Variable Land-use Depth x depth Bulk density 42.5 *** 242.3 *** 2.2 n.s. Carbon 30.9 *** 1899.2 *** 17.9 *** Nitrogen 9.4 *** 462.0 *** 24.1 *** Phosphorus 35.6 *** 675.3 *** 12.1 *** pH 6.9 * 697.8 *** 89.5 *** Chloride 3.4 n.s. 2.9 n.s. 15.8 *** EC 13.4 *** 53.4 *** 6.6 * Calcium 1.6 n.s. 317.4 *** 12.1 *** Magnesium 2.0 n.s. 782.3 *** 3.5 n.s. Sodium 23.7 *** 2277.8 *** 27.1 *** Potassium 1.7 n.s. 1417.2 *** 4.1 n.s. Table 3. Mean values for soil parameters and significance of differences between land-uses using Tukey's post-hoc test Within a row, means followed by the same letter are not significantly different at P=0.05 Soildepth Land-use (m) Cultivation Grass Woodland Bulk density (A) 0-0.05 1.27a 1.21a 0.98b 0.05-0.10 1.33a 1.28a 1.23a 0.10-0.20 1.44a 1.256 1.266 0.20-0.30 1.50a 1.35a 1.43a 0.30-0.40 1.59a 1.43a 1.45a 0.40-0.50 1.63a 1.53a 1.41a Carbon (%) 0-0.05 1.73a 2.506 4.51c 0.05-0.10 1.58a 1.66a 2.316 0.10-0.20 0.93a 1.07a 1.436 0.20-0.30 0.67a 0.75a 1.00a 0.30-0.40 0.62a 0.64a 0.71a 0.40-0.50 0.58a 0.58a 0.61a Nitrogen (%) 0-0.05 0.14a 0.18a 0.266 0.05-0.10 0.13a 0.13a 0.15a 0.10-0.20 0.1Oa 0.09a 0.11a 0.20-0.30 0.08a 0.07a 0.08a 0.30-0.40 0.06a 0.06a 0.06a 0.40-0.50 0.06a 0.05a 0.05a Phosphorus (mg/kg) 0-0.05 43.11a 47.59a 65.76a 0.05-0.10 32.40a 38.94a 48.20a 0.10-0.20 12.55a 25.04a 37.04a 0.20-0.30 8.72a 19.546 33.46c 0.30-0.40 6.72a 14.036 29.79c 0.40-0.50 6.66a 11.94a 27.12c pH (Ca[Cl.sub.2]) 0-0.05 5.90a 6.OOa 6.23a 0.05-0.10 5.75a .95a 5.92a 0.10-0.20 6.28a 6.lba 5.83b 0.20-0.30 6.79a 6.37a 6.01b 0.30-0.40 7.08a 6.63a 6.356 0.40-0.50 7.29a 6.84a 6.576 EC (dS/m) 0-0.05 0.13a 0.14a 0.13a 0.05-0.10 0.12a 0.08a 0.08a 0.10-0.20 0.09a 0.07a 0.06a 0.20-0.30 0.09a 0.06a 0.06a 0.30-0.40 O.l0a 0.08a 0.06a 0.40-0.50 0.12a 0.08a 0.076 Sodium (cmol/kg) 0-0.05 0.14a 0.22a 0.04a 0.05-0.10 0.16a 0.15a 0.05a 0.10-0.20 0.25a 0.26a 0.09a 0.20-0.30 0.39a 0.39a 0.18a 0.30-0.40 0.64a 0.65a 0.336 0.40-0.50 0.84a 0.81b 0.43c (A) Bulk density based on 40°C dry soil weight. Table 4. Correlation matrix Significant correlations in bold (P<0.01) Bulk Carbon Total density (%) nitrogen Bulk density 1.00 Carbon (%) -0.46 1.00 Total N -0.45 0.84 1.00 Extract. P -0.13 0.48 0.43 pH 0.17 -0.16 -0.18 EC -0.18 0.38 0.33 Na 0.26 -0.20 -0.31 Extractable pH phosphorus (Ca[Cl.sub.2]) EC Bulk density Carbon (%) Total N Extract. P 1.00 pH -0.39 1.00 EC -0.04 n.s. 0.49 1.00 Na -0.33 0.39 0.31
 Printer friendly
 Cite/link
 Email
 Feedback
| |
| Author: | Wilson, B.R.; Growns, Ivor; Lemon, J. |
|---|---|
| Publication: | Australian Journal of Soil Research |
| Article Type: | Report |
| Geographic Code: | 8AUST |
| Date: | Jun 1, 2008 |
| Words: | 6326 |
| Previous Article: | Comparing the effects of continuous and time-controlled grazing systems on soil characteristics in Southeast Queensland. |
| Next Article: | Evaluation of long-term soil management practices using key indicators and soil quality indices in a semi-arid tropical Alfisol. |
| Topics: | |