Overview Objectives

Therefore this project is examining conservation agriculture, using practises such as zero tillage and permanent raised beds (PRB), to reduce irrigation water use, maintain farm yields and improve farmer incomes.

Project Background and Objectives

Gansu is a north-western Chinese province in the Yellow River Upper Drainage Basin. Between Gansu and neighbouring Inner Mongolia lies a distinct valley, the Hexi Corridor. In the past, reliable snowmelt water from the adjacent Qianlian Mountains has sustained the irrigated agricultural areas along the length of the valley. In more recent times, reduced snowmelt water has led to significant reductions in available surface water, whilst over-extraction and decreased recharge has lowered water tables in groundwater driven systems. As a consequence farmers are facing severe water restrictions (up to 50% reduction in allocations). Better channel lining has reduced delivery losses, but farmers have received few practical solutions to cope with the policy-driven cutbacks in water allocations, water price increases and pumping costs. Other food production issues associated with water restrictions - for example small farms, low levels of mechanisation, high inputs, conventional tillage, low incomes and the loss of young men to the cities - are placing further pressure on farmer livelihoods. Therefore this project examined and tested conservation agriculture, using practises such as zero tillage and permanent raised beds (PRBs) to reduce irrigation water use, maintain farm yields and improve farmer incomes.

Progress Reports (Year 1, 2, 3 etc)

Project Objectives:

Develop and test conservation agriculture machinery designed around the 20hp tractors to mechanise PRB in wheat/maize systems.
Demonstrate the effectiveness of permanent raised beds in improving crop, nutrient, residue and water management in wheat / maize systems of the Hexi corridor.
Assess the cost benefit of PRB farming systems.
Extend conservation agriculture using PRB technology across the Hexi Corridor.

Develop and test conservation agriculture machinery designed around the 20hp tractors to mechanise PRB in wheat/maize systems.
The Zhangye Mechanisation Institute, under the supervision of the China Agriculture University designed and constructed a prototype bed-former and planter for the project this last winter. Subsequently new machines were tested under field conditions in March 06 with varying degrees of success. In general, planting wheat with a new machine into three different field configurations; permanent raised beds, freshly raised beds and zero till, was not achieved without considerable delays caused by frozen soil, breakages, adjustments, and modifications. A comprehensive assessment of the mechanisation development for the project was essential and key to the success of the new farming system. Dr Jeff Tullberg of CTF Solutions Australia was engaged to carry out the assessment in April 2006. His comprehensive report and significant recommendations are included with this report. In brief, he recommends a planter redesign, which incorporates bed forming and planting in one pass, bed surface depth control, advanced seed metering device, improved seed fertiliser separation at the delivery point, round shouldered beds and a capacity for simple adjustment to plant in differing field configurations. Other suggestions included the modification of an existing plot harvester in Zhangye, as opposed to the design and construction of a tractor mounted harvester.

Demonstrate the effectiveness of permanent raised beds in improving crop, nutrient, residue and water management in wheat / maize systems of the Hexi corridor.
Site establishment at the Zhangye research station commenced at the end of the 2005 cropping season. The focus was on residue management in preparation for the following in-project cropping regime and the design and local manufacture of irrigation water measuring devices. Calibration and testing of the new V-notch weirs were conducted following intensive training on their use and irrigation water management. Initial problems with the operation of data-logging instruments and bypass flow during the post harvest and pre-winter irrigations have been overcome by ongoing training and refurbishment of water supply channels. The 2006 cropping season commenced with the preparation of the 8 comparative fields, which included removing permanent raised beds in 6 of the 8 fields and establishing traditional, zero-till controlled traffic and fresh bed fields, 4 configurations in all in 2 replicates. The majority of the field conversion was done by hand to minimise compaction, retain residue, and reduce soil moisture loss.

Instrumentation of the research site was conducted in three phases. Phase one was intensive on-site training of GAAS personnel in the use and assembly of a weather station, enviroscans, salinity probes, tensiometers, soil temperature probes, soil moisture meters, and an array of 48 full-stop wetting front detectors. Phase two was installation of the instruments followed by familiarisation with data acquisition, data management and fault diagnosis. Phase 3 included an introduction to irrigation models with specific training in Hydrus 2D, Swagman destiny, Sirmod, Infilt and FAO56 evapotranspiration model. The team members were trained in data acquisition, instrument layout, and spreadsheet design and data management for the separate models. During the establishment phases, soil sampling was conducted for soil moisture characteristics, biotic activity, and instrument calibration. The team was trained in the use of new and locally manufactured sampling tools.

The current approach to spring wheat production in Hexi is very concerning as it would appear to over use nitrogenous fertiliser. Dr Tullberg’s report also asks why the use of N fertilisers is 5-6 times higher than Western Nations. Preliminary measurement of applied irrigation water and soil moisture deficits suggests over-watering in traditional farming is considerable and could be leaching nutrients well below the root zone, hence the need for more fertiliser.

Assess the cost benefit of PRB farming systems.
Dr Tim Sun of the University of Queensland and Professor Tian of CAU conducted an economic bench marking survey in 2005. The survey was conducted in Shandan county to the east of Zhangye city and will form the basis of the cost benefit analysis of PRB

Extend conservation agriculture using PRB technology across the Hexi Corridor.
GAMB has established the first of 3 demonstration sites for the Hexi corridor in Huan Zhuang Village, Shandan County. Three adjacent 5800 m2 fields will compare permanent raised beds, zero-till and conventional farming practices. The management panel from the Shandan Agricultural Mechanisation Station and GAMB have installed V notch weirs to measure irrigation water and will assess crop performance by intensive plant mapping. All field inputs, are also being compared. To assess changed farming practices on a commercial farm scale, site management and decisions on inputs are based on those used at the Zhangye research station.

During November and January (05/06), 3 Chinese project team members attended the 45 day capacity building tour in Australia. They travelled from Central Queensland to Southern NSW visiting a wide variety of farming enterprises and institutions. However, the tour coincided with the summer holiday period and Christmas festivities and although this was a rewarding cultural experience for them it restricted the visits available to them. The next tour will be scheduled for September/October 2006 and will coincide with the Controlled Traffic Conference in Ballarat.

Develop and test conservation agriculture machinery designed around the 20hp tractors to mechanise permanent raised bed (PRB) farming in wheat/maize systems.
The planter and bed former developed prior to the previous season was modified slightly to cope with frozen soil, high moisture content, and to improve seed/fertiliser separation in the seed bed. Although not perfect, planting for the second season at the Zhangye Research Station and the demonstration site at Shandan County was more successful than that of the previous season, achieving plant emergence in the order of 90% compared to 65% for the permanent raised bed sites in 2006. The bed former and planting implement were duplicated for the demonstration site in Jui Quan. During the planting operation in March 2007 similar problems were experienced to those in Zhangye and Shandan. Consequently the Jui Quan team have suggested a considerable number of improvements for the planter. Using these recommendations and along with those made previously by Dr Jeff Tullberg, Dr Jack McHugh, Dr HeJin and the designers at the Zhangye Mechanisation Institute (ZMI) have designed a new implement to meet the needs of the project. The new implement will improve trash flow, incorporate depth control and have the flexibility to plant in flat systems as well as raised beds. ZMI have successfully designed, built and tested a tractor mounted grain harvester. This modular designed machine can operate in small fields and on raised beds, offering the farmers an inexpensive method of using their own tractor for multiple tasks. Although the machine requires some modifications in the area of manoeuvrability, driver comfort, and visibility, it appears robust, effective and of commercial value. Some considerable thought has been given to planting wheat after maize and a new permanent raised beds (PRB) planter is being designed based on those developed by the China Agriculture University (CAU) under the auspices of other dryland CT projects.

Demonstrate the effectiveness of permanent raised beds in improving crop, nutrient, residue and water management in wheat / maize systems of the Hexi corridor.
There was no significant difference between wheat yield of conventional practice (CT) and PRB even though plant emergence was 20% lower under PRB in the first season. WUE efficiency for PRB was 11.2mm/kg/ha, whereas CT was 9.2mm/kg /ha. The other treatment’s WUE were less than CT at ~8.7mm/kg/ha. The 90% emergence observed in the 2007 season at the Shandan demonstration site and Zhangye station should see a considerable improvement in yield overall. Despite that emergence at the JuiQuan demonstration site was down to 83% on average, yield estimates for 2007 are equivalent to regional averages. Levels of total nitrogen, phosphorus, potassium and available potassium in PRB were the highest amongst all the treatments. PRB and ZT also increased the population of bacterium, actinomyces, and anaerobic cellulose-decomposing bacterium. Based on this information fertiliser quantity was reduced during planting for the 2007 season. Water balance data have demonstrated a 40% irrigation water saving using PRB in the 2007 season, which followed a 22% in saving during first season. The first season at Shandan reported an 11% saving of irrigation water followed by a 20% saving on 2007. PRB consistently displayed increased soil moisture storage after the first month of plant growth, requiring 32mm less irrigation water on average per irrigation event. Data from salinity probes suggests that there is no build up of salinity in the PRB site at 60cm. Salinity levels CT are twice that of PRB, but well below the threshold salinity tolerance for wheat. Percentage of total salts in the profile in CT was 2.5 times higher than PRB, but similar to Zero till (ZT) sites. Standing stubble of 20cm height has been retained on ZT and PRB fields for two seasons at Zhangye research station and a single season at Shandan County demonstration site. The stubble cover at both sites maintained high soil moisture and reduced soil temperatures, hence planting was delayed at both sites by up to a week compared to CT in the first season. The second season saw similar delays, but most of this was caused by the poor performance of the planter in moist and frozen soil conditions with heavy trash. Standing stubble will be reduced to 10cm in 2008 in an attempt to reduce blockages during planting and raise soil temperatures faster after winter.

Assess the cost benefit of PRB farming systems.
Benchmarking was completed at project commencement and final cost benefit analysis will be conducted toward the project completion date in 2009. The GAMB extension staff are recording farming inputs and based on preliminary data gathered during the establishment phases of the demonstration sites, PRB incurred savings of ~415Yuan/ha, which includes extensive first year bed forming. In subsequent years this figure should improve markedly as evidenced by the 744Yuan/ha saving in ZT fields.

Extend conservation agriculture using PRB technology across the Hexi Corridor.
The second demonstration site has been established in Jui Quan City. Planning and discussions are well under way toward the establishment of the third site in March 2008 in Wuwei City. At this stage there is no documented evidence of farmer uptake of PRB. However, the Guangzhou District Government Agricultural Department is setting up a PRB demonstration site near Zhangye City, independent of this project.

The aim of this work is to introduce and extend conservation agriculture (CA) to improve water use efficiency and the prosperity of small farmers in the water-stressed area known as the “Hexi Corridor” of north-west China. Key constraints to implementation of CA in China are the lack of appropriate machinery and competition for crop residues. Comparative performance of irrigated spring wheat in permanent raised beds (PRB), fresh raised beds (FRB), zero till-control traffic (ZT) and conventional tillage (CT) was assessed on the Zhangye Research Station near Zhangye City, complemented by three demonstration sites which compared PRB, ZT and CT located in Shandan County, Wuwei and Jiuquan Cities, in Gansu Province. Irrigation volume was based on soil moisture deficit for PRB, FRB and ZT, while CT received 1.5ML/ha per irrigation (up to 6 irrigation events), which was normal farming practice. In 2007 yield of PRB was 7.1t/ha, 9% better than CT and significantly improved on all other treatments, despite a 10% poorer establishment than CT. Irrigation water savings from PRB was 43% in 2007, compared with that of CT. Soil profile salinity (EC) varied across the beds in both PRB and FRB. The centre of the bed at 0.20-0.30m soil depth had the greatest salt accumulation at 2.79dS/m. At 0.50-0.60m EC was also considerably higher than the other profiles at 2.47dS/m, but not enough to affect wheat yield.
Crop emergence, yield and applied irrigation water at the demonstration sites followed similar patterns to those found at the Zhangye Research Station. This indicates that conservation agriculture can be implemented in this region without loss of yield, and with considerable gains in natural resource conservation, provided that the operational capabilities of the prototype machinery continue to improve.
A second generation no-till planter was built in the winter of 2007 to enhance trash flow, improve depth control, reduce soil disturbance and to provide flexibility for use in flat and PRB farming systems. The machine incorporated 3 tool bars to maximize tine spacing and an optional powered cutter to clear >6 tonne/ha of residue. Depth control was achieved with press-wheels on floating seed tube knives. Furthermore, narrow vertical openers were used to minimise soil disturbance. The maize planting operations fell below expectations due to short pre season testing, heavy trash conditions, very moist soil, poor soil flow and the lack of appropriate row preparation and slot closing devices. The operational conditions and the necessity to minimise planter weight to operate on <20hp tractors also hampered maize, wheat and barley seeding operations at the demonstration sites.
A powered disc no-till planter was constructed and tested in Jiuquan City specifically for wheat after the 2008 maize crop. It too suffered from residue blockages, but was suitable behind a 20hp tractor. The planter will be retested in September 2008 after minor modifications to the maize stool cutting disc, bed renovation tools and structural strengthening. A weeder was also designed and developed in Jiuquan during the 2007-08 winter and is currently in the final stages of manufacture and initial testing.
Following minor modifications to blower speeds and seed separators, a tractor mounted wheat harvester was successfully tested in July 2007 on the research station site and local Zhangye farm. However, there were still minor issues with machine stability on uneven terrain and driver comfort that should be addressed prior to commercialisation.
A third demonstration site has been established in Wuwei City by the Gansu Agricultural Mechanisation Bureau. This site is the final of three sites established by the project along a 500km section of the Hexi Corridor. The site managers are fully committed and are very pleased with the results from all three sites. Local farmers invited to field days recognise the merit in the PRB farming system through reduced inputs such as labour, fuel and water, but still find it hard to believe yield can increase when less land is planted to cropping.
At this stage there is no documented evidence of farmer uptake of PRB, but there are other demonstration sites for conservation tillage and conservation agriculture in close proximity to each ACIAR demonstration site and the Zhangye Research Station.
In October 2007 training in Conservation Agriculture concepts and extension was delivered in Beijing to 25 people associated with the project. This is being followed up by ‘Train the Trainer’ training for village technicians and extension staff in Zhangye and by Farmer Farm Schools in CA concepts at each of the demonstration sites in June 2008. The training is a joint endeavour between ACIAR and the China-Canada Sustainable Agriculture Development Project, based in Gansu and Sichuan Provinces.

The objective of the project is to develop and test machinery suitable for raised bed farming, to compare crop performance, residue and water management of current practices, modified current practices and permanent raised beds (PRB) in wheat /maize systems of the Hexi Corridor, to assess the cost benefit of PRB farming systems and to extend PRB technology.
Development of a zero-till planter that can be operated by <20hp tractors on both flat and raised bed farming systems has gone through a number of iterations. The final version is a powered flail, zero-till, 5 row planter, weighing less than 350kg developed by China Agricultural University and Shandong Wannongda agricultural machinery factory in 2009. It was tested at Zhangye by planting wheat after maize on PRB in March 2009, but failed to meet expectations due to high soil moisture, blockages, and insufficient adjustment to fine tune the implement to the conditions. The previous 2008 version, with chopper removed and modified to a three bar configuration for enhanced soil and trash flow, was more successful than its contemporary, but soil disturbance was still too high and had limited scope for adjustment to suit soil conditions. Fresh raised bed (FRB) system utilising the Shandong planter has been more successful, even though this method uses considerably more inputs (energy, labour) than other tillage systems. However it conserves water by saving 25% of applied irrigation. Water savings and ACIAR training have prompted the development of double bed planter and considerable extension of FRB along the Hexi corridor.
Integrated mechanical weed control in maize was achieved with a combined sprayer and inter row cultivator, designed and developed by CAU. The cultivator removed weeds in the furrow and inter-row space in maize and in the furrow for wheat and barley. Alternatively, shielded sprayers can be used in the inter-row for maize. Weed suppression was also achieved by planting the furrows with wheat during the wheat rotation or legumes during the maize crop. However normal practice for weed control is to apply a pre-emergent herbicide during tillage operations. Notwithstanding the short comings of the current zero-till planter mentioned previously, adoption of zero-till farming has also been hampered because a pre emergent applicator/process remains to be developed that can be used at planting without contaminating the seed bed.
Increased water productivity has been the standout success of the project in that PRB has achieved 40% water savings under controlled conditions (Deficit replacement with instruments) and 25% on farm (no instrumentation). Anecdotally, FRB adopted in the counties of Shandan, Minle, Ganzhou and others are conserving similar amounts of water. Average root zone salinity of 2.2dS/m in PRB, FRB and ZT treatments validate previous solute modelling predictions of salt accumulation and moderate leaching requirement. Extension officers have been advised of the increased risk of salt accumulation in raised bed farming when using groundwater and how to combat it.
Lower soil temperatures at planting retarded emergence in PRB and ZT by ~2 weeks compared with conventional and FRB treatments. The maize was a long season variety and unsuited to the cooling conditions at season’s end, therefore grain potential was not achieved. Increased root zone salinity may have contributed to the 20% yield loss by about 3% in PRB and ZT compared with CT. Although the number of kernels per ear was greater in PRB than conventional, the weight per thousand kernels was significantly higher in conventional treatments, due to the earlier maturation of that treatment.
A 4% yield reduction in maize at Jiuquan PRB demonstration sites still produced a gross profit increase of 3.95%. Barley production in Shandan in both ZT and PRB continues to outperform conventional farming by 6%, producing increased gross profits of 16%.
Crop residue is continuing to maintain high soil moisture and low soil temperatures, which hampered planting and seedling emergence. The simple solution of planting later for wheat and barley has cultural and water availability implications and will continue to stifle adoption for some time. Although farmers and extension staff have an appreciation of the benefits of residue retention it flies in the face of convention, culture and current weed control methods.
Throughout 2008 training of extension staff, researchers and farmers in conservation agriculture (CA), extension methodology, and equipment was conducted in north western Gansu. The program included Train the Trainer and field day in Zhangye, and 5 Farmer Field Schools in Zhangye, Jiuquan, Shandan, Wuwei and Minchin. Resulting in 42 trained technicians in CA delivery and 290 township level officials and farmers receipting CA training, whilst handing out over 1000 booklets for CA. CA appears on the Jiuquan Government web site and in agricultural mechanisation information in Gansu web pages.

Project Outcomes

This 4-year project was based on the premise that conservation agriculture (CA) using permanent raised bed (PRB) technology could save water without yield penalties. The project team worked with farmers to successfully test a bed former, a 20 hp tractor-mounted wheat harvester and an integrated mechanical and chemical weed control implement, and later identified factories for commercialisation. The team also developed a tactile implement guidance system to facilitate precision planting.
Prototypic lightweight zero till (ZT) planters based largely on local capacity were built and assessed. However, testing showed that soil disturbance was too high, slot closure inadequate and residue handling poor. A final version built in eastern China showed great promise, but still required minor modifications to cope with heavy residue, poor soil flow and adjustability to suit various soil and planting conditions.
Under research conditions water savings were up to 43%, with minimal soil salinity accumulation. Average PRB wheat yield (6.4 t/ha) increased marginally. In contrast, due to low soil temperatures at planting, average PRB maize yield (11.3 t/ha) was 20% lower than conventional farming.
Average water saving for PRB in three on-farm demonstration sites established in the northern, central and southern Hexi Corridor was1.5 ML/ha (28.13%). Advances in on-farm comparative yield were not significant. However, PRB had significant economic, social and environmental benefits, which can be extended across the Hexi Corridor. On-farm data suggest a 2% annual increase of PRB adoption over 5 years would produce accumulated water savings of 2,520 GL.
A cost-benefit analysis revealed that PRB recorded reductions in water use, diesel use, labour input and machinery costs, totalling $312/ha less than conventional farming and producing a $318/ha higher net profit. However, PRB maize recorded an average loss of $92/ha. The increases in net profit were not considerable, but in relation to input costs and the low cost of living in the region, they could provide a substantial benefit to farmers in the longer term.
Capacity building included extension, monitoring and measurement, CA farming, mechanisation and trainer courses for 160 technicians and 45 researchers. Elements were conducted in Australia, Canada, various Chinese provinces, major cities in Gansu and at the demonstration sites. Project members and 2,000 farmers learnt a new and easier method of farming, which facilitated advances in water saving technology and machinery development many years ahead of current practices. With a greater appreciation of sustainable farming techniques they have extended fresh raised-bed farming to about 33,000 ha. Though not yet agronomically or economically sustainable, it was considered an interim step toward conservation agriculture until suitable PRB zero-till (ZT) machinery and weed control technology for ZT farming become available. These were the only issues limiting the broad adoption of CA in Gansu.
PRB as part of CA technology is on the cusp of rapid advancement in northern China. The Provincial Government in Gansu is promoting PRB as one of three key technologies for water saving; also the Ministry of Agriculture and Department of Agricultural Mechanisation have expressed keen interest in promoting PRB in northern China. So the resolution of the performance issues of the lightweight planter is of the highest priority.

Project ID
Project Country
Inactive project countries
Commissioned Organisation
University of Queensland, Australia
Project Leader
Dr A.D.Jack McHugh
Allen (Jack) McHugh a.mchugh@cgiar.org
 Cell: +86 13995419104
+61 (0) 7 4631 1870 
Collaborating Institutions
China Agricultural University, China
Gansu Academy of Agricultural Sciences, China
Gansu Agricultural Mechanisation Bureau, China
Project Budget
Start Date
Finish Date
Extension Start Date
Extension Finish Date
ACIAR Research Program Manager
Dr Mirko Stauffacher