New Zealand Journal of Agricultural Research

ISSN: 0028-8233 (Print) 1175-8775 (Online) Journal homepage: http://www.tandfonline.com/loi/tnza20

Irrigation and soil physical quality: An investigation at a long‐term irrigation site M. S. Srinivasan & R. W. McDowell To cite this article: M. S. Srinivasan & R. W. McDowell (2009) Irrigation and soil physical quality: An investigation at a long‐term irrigation site, New Zealand Journal of Agricultural Research, 52:2, 113-121, DOI: 10.1080/00288230909510495 To link to this article: http://dx.doi.org/10.1080/00288230909510495

Published online: 22 Feb 2010.

Submit your article to this journal

Article views: 202

View related articles

Citing articles: 4 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tnza20 Download by: [37.44.207.50]

Date: 15 January 2017, At: 21:17

New Zealand Journal of Agricultural Research, 2009, Vol. 52: 113-121 0028-8233/09/5202-0113 © The Royal Society of New Zealand 2009

113

Irrigation and soil physical quality: an investigation at a long-term irrigation site

M. S. SRINIVASAN* R. W. McDOWELL† AgResearch Limited Invermay Agricultural Centre Private Bag 50034 Mosgiel 9053, New Zealand *Current address: National Institute of Water and Atmospheric Research Limited, PO Box 8602, Christchurch 8840, New Zealand. † Corresponding author: richar d.mcdowell@agrese arch, co.nz Abstract The hypothesis that irrigation improves soil quality by improving its moisture holding capacity compared to dryland (non-irrigated) soils was tested using data on soil physical properties and irrigation frequency and timing (dryland, and irrigated at 10, 15, and 20% v/v and 3-weekly) from a longterm irrigation trial at Winchmore, South Island, New Zealand (soil type: Lismore silt loam). Despite previous work showing that macroporosity decreases and bulk density increases with irrigation and grazing, the most frequently irrigated plots showed no significant difference in these measurements, or in saturated and unsaturated hydraulic conductivity, compared to dryland plots. This implies that the soil had an inherent resilience to soil physical degradation. With time (years), the number of irrigations had declined and date of the first irrigation needed during the irrigation season had become later, leading to the perception of improved moisture holding capacity. This was supported by an increase in soil carbon concentrations in all treatments with time since development. Irrigations were scheduled, on average, at 10.9, 15.3 and 20.6% v/v for 10, 15, and 20% treatments, respectively, until 1997, but at 9.9, 14.1 and 19.2% for 10, 15, and 20% treatments, respectively, afterwards. However, several reasons explain this and refute the hypothesis of improved A08041; Online publication date 23 March 2009 Received 31 July 2008; accepted 25 January 2009

moisture holding capacity. The most pertinent was that irrigation timing was probably affected by the use of a gravimetric method prior to 1997, and a TDR-based technique since. In addition, no difference was seen in pasture production relative to irrigation with time. Thus, apart from soil C concentrations there was little conclusive evidence to support the hypothesis that irrigation improves soil moisture holding capacity at this site. Keywords border-dyke irrigation; irrigation frequency; soil physical properties; soil quality; soil resiliency

INTRODUCTION Irrigation is central to production agriculture. Farmers perceive irrigation as an input that maximises the return from other inputs such as fertiliser, labour and fuel (New Zealand Ministry of Agriculture and Forestry 2000). Published evidence on the impact of irrigation on soil properties is mixed, with Metherell et al. (2002) noting that although there is some anecdotal evidence suggesting irrigation improves soil quality, there is limited data available to support this conclusion. They defined soil quality as its ability to store moisture, needing fewer irrigations. Rickard & Cossens (1968) and Houlbrooke et al. (2006) reported that irrigated soils tend to have greater bulk density and lower macroporosity than non-irrigated (dryland) soils. Hussein et al. (1992) compared irrigated and dryland soils in the semi-arid south-east of Zimbabwe, and found no significant differences in soil compaction and erodibility under both natural and simulated rainfall conditions. Anderson et al. (1990) found an increase in hydraulic conductivity due to irrigation, while Yen & Akan (1983) reported a decline in soil hydraulic conductivity due to irrigation, implying an increased overland flow of irrigation water. We studied the influence of border-dyke (flood) irrigation on soil quality in a Lismore silt loam soil at Winchmore Irrigation Research Station near

New Zealand Journal of agricultural Research, 2009, vol. 52

114

ashburton, New Zealand. The long-term irrigation trial enables the study of dryland pasture compared to pastures irrigated at different soil moisture contents since 1958. our study compared the differences in soil physical properties under irrigated and dryland conditions, as well as frequency and timing of irrigation events among the irrigated soils. Specific soil properties studied were saturated and unsaturated hydraulic conductivities, micro- and macroporosities, particle size distribution and bulk density of surface soil layer (top 5 cm). We hypothesised that long-term irrigation influences surface soil (physical) properties, and hence the soil's ability to hold water and support pasture growth, compared to dryland soil. Given that any difference would likely decrease with depth, we concentrated measurements on the top 5 cm of soil.

MAterIAl And

Methods

Winchmore irrigation trial The long-term irrigation trial at Winchmore (171°48 E and 43°47 S) began in 1958. The trial is on a lismore stony silt loam (Pallic Firm Brown) derived from moderately weathered greywacke loess over gravel (Fraser et al. 1994). The a horizon is approximately 180 mm deep with an average stone concentration of 5 g kg–1 (Francis & Kemp 1990). The average particle size distribution in the 2 mm was washed until only stones remained. The stones were ovendried and weighed to calculate the percentage of stones in the soil sample. The washed soil was added to the sieved sample (