Improved Management Practices for Culinary Bamboo Shoots

Local and Export Markets

A report for the Rural Industries Research and Development Corporation

by V Kleinhenz and D J Midmore

May 2002

RIRDC Publication No 02/035

RIRDC Project No UCQ-9A

© 2002 Rural Industries Research and Development Corporation. All rights reserved.

ISBN 0 0642 58434 6

ISSN 1440-6845

Improved Management Practices for Culinary Bamboo Shoots - Local and Export Markets

Publication No. 02/035

Project No UCQ-9A

The views expressed and the conclusions reached in this publication are those of the author and not necessarily those of persons consulted. RIRDC shall not be responsible in any way whatsoever to any person who relies in whole or in part on the contents of this report.

This publication is copyright. However, RIRDC encourages wide dissemination of its research, providing the Corporation is clearly acknowledged. For any other enquiries concerning reproduction, contact the Publications Manager on phone 02 6272 3186.

Researcher Contact Details

Prof. D.J. Midmore

Primary Industries Research Centre Central Queensland University Rockhampton Qld 4702

Phone: (07) 4930 9770

Fax:     (07) 4930 9255

email:    d.midmore@cqu.edu.au

Internet:  http://science.cqu.edu.au/psg/index.html

In submitting this report, the researcher has agreed to RIRDC publishing this material in its edited form.

RIRDC Contact Details

Rural Industries Research and Development Corporation

Level 1, AMA House

42 Macquarie Street

BARTON   ACT   2600

PO Box 4776

KINGSTON   ACT   2604

Phone:  02 6272 4539

Fax:       02 6272 5877

Email:   rirdc@rirdc.gov.au. Website:http://www.rirdc.gov.au

Published in May 2002

Printed on environmentally friendly paper by Canprint

Foreword

The bamboo industry in the mid-1900s consisted of a small number of individuals dedicated to the cultivation of a number of bamboo species with the view to satisfying the local market in Australia. It is the belief of the RIRDC that complementarity must exist between those involved in new rural industries, and that RIRDC research support should be sanctioned by informal or formal industry development groups.

As no such groups existed for bamboo, one of the first milestones for this project was to establish the Australian Commercial Bamboo Corporation (ACBC) to focus on the commercial aspects of bamboo, including production, marketing and research, and the Bamboo Society of Australia (BSA), to foster interaction between members of the public, not necessarily growers, interested in bamboo from a botanical and/or gardening perspective.

The research reported on in this report, addresses key issues identified by growers at the industry meetings held at the start of the project. These include production issues covering water usage, nutrient usage and culm management and post-harvest issues, focusing on the factors that can extend shelf life, including cooling, storage temperature and packaging. This report should go a long way to meeting industry demand for sound production and post- harvest management procedures.

This project was funded from RIRDC Core Funds which are provided by the Federal Government.

This report, a new addition to RIRDC’s diverse range of over 800 research publications, forms part of our Asian Foods R&D program, which aims to improve profitability and promote market growth by increasing the shelf-life of fresh Asian vegetable products.

Most of our publications are available for viewing, downloading or purchasing online through our website:

·         downloads at www.rirdc.gov.au/reports/Index.htm

·         purchases at www.rirdc.gov.au/eshop

Peter Core

Managing Director

Rural Industries Research and Development Corporation

Acknowledgments

The support and valuable contribution of the following is acknowledged:

·         Central Queensland University and Northern Territories Department of Primary Industry and Fisheries (NTDPIF) for in-part finance.

·         Durnford Dart, John O’Hare, Martin Hartmann and John White for allowing research plots to be superimposed at their bamboo plantations and providing plant material for research purposes.

·         Victor Cusack and Peter Topperwein for assistance in preparing the rules for the Bamboo Society of Australia and the memorandum and articles for the Australian Commercial Bamboo Corporation.

·         Kevin Blackburn, Mark Traynor and Melinda Gosbee of the NTDPIF for field and post- harvest trial cooperation.

·         Lynelle Vale for effective administration assistance.

Contents

Executive summary

From this research project a series of practical recommendations can be made with the proviso that they should be locally tested for their relevance and validity, given that management * environment interactions are known to exist for many plant species of economic importance.

The setting up of both the Bamboo Society of Australia and the Australian Commercial Bamboo Cooperation (ACBC) has been key to the interaction between growers and others interested in bamboo. The current www site for the BSA (http://www.bamboo.org.au/) contains information on both the BSA and ACBC, and provides the relevant contacts to which to direct oneself for membership details. Both the BSA and the ACBC are vibrant, with a healthy membership; both with a regular newsletter and the latter with greater than 70 registered commercial members. These will underpin the success of bamboo as a plant species of great interest and commercial relevance.

The collation of relevant literature on bamboo shoot and timber production has led to a publication in Volume 74 of Advances in Agronomy (2001), and strengthened the position of CQU as a repository for information on bamboo. A post-doctoral fellow (funded by ACIAR) and a PhD student both work full-time on bamboo at CQU, and on an ACIAR-funded collaborative project with growers in NSW, Queensland and the Northern Territory, and with the NTDPIF and various organizations in the Philippines coordinated through the Philippines Council for Agricultural Research and Development. This is currently providing further useful information on optimum field production practices.

Our data on water use by Bambusa oldhamii show that higher application rates can induce earlier shoot production, but yields even on clumps of 9 years of age only reached 2.5 t ha-1. These clumps had not received irrigation until the experiment began, when the clumps were 6 years old. In contrast, for Phyllostachys pubescens, without irrigation and only little rainfall, shoot numbers were greater with less drainage but better supply of underground water during the first 2 years of the study. With irrigation in the following years, bamboo shoot numbers and weight were much better under better-drained conditions with less supply of underground water. In those years, torrential rainfall and erratic irrigation negatively affected bamboo growth through overwet soil conditions. Shoot production was superior under better-drained soil conditions with little precipitation, and evenly distributed application of non-excessive amounts of irrigation water. Optimum soil water conditions were especially important during the shoot season and in the A soil horizon. Under those conditions, plants showed less water stress under high irrigation, and there was a significant response to application of higher rates of nitrogen.

Studies on usage by bamboo in the field focused on nitrogen (N), while a large scale pot trial allowed us to determine that bamboo supplied with N in the NH4-N form enhanced water uptake (and by implication, growth). The rates of N:P:K extracted from nutrients applied to young bamboo plants supplied with nutrient solution was 1:0.4:1.3, notably highlighting the importance of sufficient K reserve in the field for bamboo.

But over-wet and too dry conditions limited the ability of P. pubescens to respond to externally added N, but under ideal soil moisture applications of 130, 250 and 500 kg N ha-1 year-1 supplied yields of 6, 8-9 and 11-12 t ha-1 shoot yield. These figures will vary in other sites in accordance with the natural soil fertility and rate of soil mineralisation of N. In clumps not harvested for their shoots, higher N rates promoted the production of more, smaller diameter culms.

In order to circumvent the need to test for soil N, we developed a Diagnostic Recommendation Integrated System (DRIS) for N in bamboo. Our suggested optimum of 3.0% total leaf N represents a value above which plants only respond slightly to externally added N, while below which bamboo plants have a greater affinity for N. This value appears to be robust enough to be applicable across all the species studied, and leads to a balanced set of both number and diameter of new shoots.

Our studies on culm management focused on B. oldhamii, comparing constant numbers of culms of each age class (1-1-1, 2-2-2, 3-3-3 and 4-4-4 culms of 1-2-3 years of age). There was a trend that the yield of fresh edible shoots was promoted by lower standing-culm density, but yields overall were low in this experiment over the three years that it ran (it is still continuing). However, when optimising bamboo shoot and timber yield, the 3-3-3 ratio was most suitable. Older culms apparently contributed less to clump productivity (based upon data for photosynthesis), and a ratio of 4-3-2 of 1-2-3 years old culms may indeed be superior to the 3-3-3, but this remains to be proven.

The post-harvest studies concentrated on determining the relationships between shoot weight and external dimensions upon cooling (most effectively done by hydrocooling), on the contribution of cut versus remaining surface of shoots to weight loss and storage temperature (weight loss least at 1oC and greatest at 25oC) and on packaging materials (weight loss least by heat sealed PVC film or low density polyethylene (LDPE) bags). However, the latter two materials also induced most internal condensation, with deterioration in visual quality within 14 and 21 days, respectively. Shoot respiration was, as expected, greater at 20oC than at 2oC, accounting for 18% and 34% total weight loss for shoots stored in LDPE film at 2oC and 20oC respectively, but only 4-6% of total weight loss of open-stored shoots.

Visual quality of shoots stored at 8oC declined after 6 days of storage, while that of those stored at 1oC was not notable even for one month’s storage. Based upon these experiments, the optimum packaging was at 1oC with semi-permeable materials (micro-perforated LPDE bag or LPDE film) in which shelf life could be extended beyond 28 days.

Some of the discolouration was due to the presence and growth of bacteria and fungi, for, as with other fresh vegetables, we found contamination of shoots (leaf sheaths, cut ends and internal tissue) by several species of Bacillus, several genera of moulds including Fusarium, and the presence of some coliform bacteria was also indicated.

Overall, the studies reported in this publication add to the body of knowledge upon which a successful bamboo industry will be established. Reference should be made to the outputs of a parallel project on market assessment, Supply chain management, strategy and industry development for the commercial bamboo industry, (RIRDC project UQ-87A, forthcoming), before prospective growers invest in the industry, for the market in Australia for fresh shoots is currently of limited size, and very small compared to the overseas export potential.

Introduction and Objectives

There is much interest in Australia in the production of fresh bamboo shoots for the local and export markets, yet individuals’ efforts prior to 1997 to establish a viable industry in the past have been disjointed and ineffective.

The range of climate conditions across Australia ensure that shoot production, taking advantage of the seasonality differences between running and clumping species, can extend for close to eight months of each year.

Based upon predictions by Dooley (1992), a bamboo industry would be viable if yields of 15 t ha-1 or greater at farm gate prices of $4.5 kg-1 or higher are obtained. More recent data (Cusack, 1999) would suggest that 10-12 t/ha would realistically result in grower profit. Data from an earlier project (RIRDC-UCQ 4A) indicate that this is a reality, with P. pubescens reaching these milestones six years after planting. Prices varied between $5.50 and $7.50 kg-1 throughout the 1996/1997 season (they have since been halved due to expanded supply of shoots to the market) hence it has been shown from the production point of view that the industry is economically viable. Another grower has achieved a consistent >$10 kg-1 by caring for quality and sales to speciality users. All grades of fresh bamboo shoots were quickly purchased, for as with most vegetables the fresh product is preferred, even at a price differential, to the preserved product. The estimated planting of 50 ha would provide an annual gross farm-gate volume of produce between $3.4 and $5.2 m for the domestic market, in 1997. More recent data (Cusack, 1999) would suggest a gross income of $2 million from 50 ha of mature plantation. Imports of canned bamboo shoots to Japan and Taiwan have exploded over the first five years in the 1990’s (Vinning, 1995). Prices of fresh shoots peak (four-fold of normal price) in Taiwan in January just before Chinese New Year. Landed prices of $4.00 kg-1 in Taiwan (including the 40% tariff against imported bamboo shoots) and $4-6.00 kg-1 in Japan would be competitive with local production, and even desirable due to the clean/green image of produce from Australia. Additional annual income to producers is derived from the timber, which is always in demand but will require aggressive marketing to dissipate the volumes of timber anticipated in the future from current plantings.

Our earlier research has emphasised the need to manage irrigation when rainfall is insufficient to maintain optimal growth rates in bamboo. There is a paucity of information on water management for bamboo shoot production. Our own data suggests that doubling rates of irrigation prior to and during the shoot season for P. pubescens results in greater shoot diameter, whereas lack of irrigation in B. oldhamii led to fewer shoots being produced. Since shoots are initiated during the summer preceding spring shoot production, an understanding of the influence of objective temporal water supply on shoot yield (number and size – for these parameters affect quality and market price) at the time, when shoots develop, and at the time of shoot elongation, is clearly required. This should be gained in parallel with an understanding of plant water requirements to sustain total dry matter production.

Reference to culling rates of culms is more readily available in the literature (Chen, 1993; He and He, 1993) indicating that shoot number is inversely related to culm or pole number, i.e. if more shoots are left to grow on into poles fewer edible shoots can be harvested, or, expressed differently, if shoots are harvested, a greater number of shoots will be produced. An optimal culm number is probably achievable for optimal shoot production. A balance between culm or pole number (for dry matter production and timber) and shoot number (as fresh vegetable) will depend upon economic returns from both products. An understanding of the dynamics between growth of variously-aged rhizomes (Zhou et al., 1985) is important in optimising shoot yield. In monopodial bamboo, the most vigorous buds are those found in central nodes of rhizomes and those most likely to swell arise from 3- to 4 year old rhizome axes. Research on these topics is currently underway in the People’s Republic of China.

Methods to extend storage of bamboo shoots have been trialled in Australia but so far unsuccessfully (Earthcare Enterprises trialled the vacuum packing of non-sterile and chlorox- treated shoots). However, literature exists on a plethora of topics of storage concerning bamboo shoots. Boiled bamboo shoots show microbial contamination (e.g., Nakanishi, et al., 1989). Retention of B. oldhamii shoots in high humidity and low temperature reduces the build-up of fibre in storage (Chen et al., 1989). Peroxidase activity rapidly increased in harvested shoots (Chen et al., 1989), but no trend was noted for polyphenol oxidase. Snap freezing and preservation in oil are known to be used for bamboo shoots, and temporary storage is practiced by immersion in water. Further investigation into the literature, and discussions with colleagues at IHD (Knoxfield/CFT (Hamilton) will ensure the most suitable experimental approach to post-harvest procedures.

In essence, this research project had four separate objectives:

1.   Plan and manage an industry-wide workshop with the express purpose of bringing all interested in the bamboo industry together, and to discuss and vote upon the setting up of a Bamboo Industry Council to service the needs of the industry in terms of coordination for market identification and supply of product and agreement on industry standards for fresh shoots such as size, shape, freshness and post-harvest management. Collating and translating (where necessary) information from relevant overseas reports on bamboo production and post-harvest management, and publishing in a readily accessible format.

2.   Understanding of water management for bamboo.

(a)  In the field at Northern Territory and Eumundi with irrigation treatments varying in intensity and periodicity.

(b) In the glasshouse at Rockhampton to study the influence on water requirements and contributions of variously-aged culms on shoot production.

(c)  In waste water reuse trials in comparison with other perennial species.

3.   Clarification of culm-thinning and nutrient management practices suitable for clumping bamboo species.

(a)  In the field in Eumundi and in the glasshouse at Rockhampton to understand the contribution of variously-aged culms to shoot yield.

(b) In the field at Northern Territory, Eumundi and Dayboro for data on critical nutrient concentrations.

4.   Testing on-farm and under controlled conditions simple post-harvest practices following, or not, peeling of shoots. The treatments include forced air chilling or hydro cooling and assessment of relative water loss through cut or undamaged surfaces of the shoot.

Outcomes

1. Industry Cooperation

Industry cooperation began with a meeting organised by CQU. The number of participants (80) who converged on Brisbane in October in 1997 exceeded the expectations of RIRDC who part-funded and Central Queensland University (CQU) who organised the workshop. Originally planned to review the focus of a currently funded project researching production, post-harvest and marketing issues for the bamboo shoot industry, the agenda was opened up to cover basic production techniques, irrigation and fertilisation, and other issues such as propagation protocols, physiology and marketing.

Dr Jeff Davis of RIRDC opened the meeting, outlining the rationale for the use of Commonwealth funding for encouragement of new industries, especially those with recognisable financial returns for the benefit of the people of Australia. This was followed by two presentations, one each by Victor Cusack and Durnford Dart of Bamboo World and Bamboo Australia, respectively. Their talks concentrated on the pros and cons of various bamboo species suitable for shoot production, including clumping (in particular Dendrocalamus asper) and running (in particular Phyllostachys pubescens) types. While Mr Cusack explained the nature of gregarious flowering, and gave out information on establishment costs and cash flows, Mr Dart spoke of value-adding by removal of leaf sheaths from shoots and storing and transporting at 2-4°C, and marketing fresh produce from Queensland to the southern states.

With relatively high current prices for planting material, interest is keen as to the possibility of in situ propagation from mother clumps of bamboo, and Mr Jeff Barnes, of QDPI, spoke of his experience in this area. Using "canesett" application (propiconazole application at x 3 concentration recommended for sugar cane setts) to single and double node culm cuttings, and branch cuttings, the rate of success in achieving new plants was reduced compared to simply soaking in water culm. Of the six species trialled, Gigantochloa atter was most prolific in producing plants from culm cuttings, with 50-90% success.

Summarising three years of research by CQU on irrigation and nutrition, Dr Volker Kleinhenz highlighted the importance of access to water just prior and during the shoot season. Without sufficient water, estimated at >2000 mm per annum and concentrated just prior to and during the shooting season, bamboo cannot respond to high fertilizer rates necessary for large yields. Yields reaching 15t/ha in the sixth year after planting P. pubescens from seedlings and presenting a complete canopy and 1t/ha in the sixth year for Bambusa oldhamii with a 30% canopy, attest to the growth potential of these species in southern Queensland. Northern Territory plantations of nearly three years age are ready for experimentation on nutrition and water requirements.

Post-harvest management of bamboo can be as crucial as pre-harvest management, and Dr Peter Hoffman of QDPI drew participants’ attention to this. Through an understanding of the physiology of bamboo shoots, using the analogy of asparagus shoots, he highlighted the causes of product deterioration, the importance of pre-harvest management, and opportunities with modified atmosphere package and semi permeability of plastic wraps to CO2, O2, H2O and C2H2. Participants were left in no doubt as to the importance of correct handling of the living bamboo shoots.

Another section of the program dealt with generic marketing, product promotion and benefits of associations amongst producers. Dr Peter Twyford-Jones of QDPI spoke of the need to latch onto a champion for the industry, a leader to observe, coordinate and lead, and of the commitment to adhere to a common vision within whatever association is to be formed for the bamboo industry. Opportunities for self-improvement, such as with the QDPI DOOR (Do Our Own Research) approach through Participative Action Management can impact upon group synergy, and examples of other successful new industries, e.g., the lychee industry, offer models upon which to mould a bamboo association. The Northern Rivers Agricultural Development Association's President Mr Peter McLaughlin championed the need to invest time to cultivate export associations with Asians, an approach that has set his association in good stead. John Wilkie of John Wilkie and Associates emphasised the difference between marketing and selling, with the customer in the fore for successful marketing. Success depends on attention to detail in all transactions, from product volume, timing, pricing, currency of transaction, delivery system to the nicety of keeping promises along the marketing trail. The late Mr Matthew Johnson of CQU presented the concept for development of A business and marketing plan for a new product such as bamboo shoots, comparing standard, controlled and simulated test market scenarios. Such a plan was to have been developed by CQU for the bamboo industry, but funding was not forthcoming.

Before the open forum to gain consensus on programmed research, communication and the setting up of an association, or cooperative, amongst industry representatives Professor David Midmore of CQU presented a overview of the analysis of survey results received from 55 of 360 recipients across Australia. Already more than 25 ha of bamboo have been planted in Australia, with approximately one third as P. pubescens and one third as B. oldhamii.  Only three producers currently market shoots and timber, and the harvest period across species spans September to May. Prices for shoots ranged from $4.00/kg for small diameter shoots to $18.00/kg for premier shoots sold directly to restaurants. The survey highlighted the lack of objective information with respect to irrigation and nutrition requirements, and the wish by respondents to access information on pricing, marketing, and export opportunities.

During the open forum the proposed continuation of research by CQU into irrigation and nutrition requirements for bamboo shoots, studies on post-harvest storage and transport protocols, and into the balance between and shoot and timber production was approved, as was the development of an industry business plan. The latter tied in with a motion from the floor to nominate a steering committee, charged with developing a draft constitution for an inclusive Australian Bamboo Association.  Following synthesis of a constitution drawing on examples from other industry associations, the draft was forwarded to participants from the workshop (and others interested persons, who may care to contact the author) for feedback and voting during an inaugural meeting of the association planned for June 1998 in Brisbane. Also planned for that meeting was a session within which some of the basics of bamboo structure and physiology would be explained.

The planned meeting workshop for June 1998 was postponed to August 1998 in order that the proposed rules for the proposed Bamboo Society of Australia and the proposed memorandum and articles for the proposed Australian Commercial Bamboo Corporation be circulated as hard copy to all interested parties before the meeting.

Over one hundred enthusiasts and visionaries converged on Brisbane at the end of August 1998 to have their say in the setting up of both the Bamboo Society of Australia (BSA) and the Australian Commercial Bamboo Corporation (ACBC). The BSA was set up to promote all non-commercial aspects of bamboo interests in Australia, by way of a newsletter and annual meetings, and the ACBC to focus on the commercial aspects of bamboo including shoot and timber production, supporting research, development and marketing. Voting in of each at the inaugural meetings was unanimous.

Organised by Central Queensland University (CQU), with Rural Industries Research and Development Corporation (RIRDC) support, the one day meeting and one-day field trip provided a fine venue for formal and informal linkages between bamboo growers from Queensland, Northern Territory, New South Wales, Victoria and Western Australia. As an emerging industry, bamboo (for culinary shoots, and for timber) offers attractive financial returns to growers after the three to five year investment period. RIRDC encourages the establishment of industry corporations, preferring where possible to fund research through them than with individual growers. Following the formal voting in of the BSA and ACBC, and their respective Boards and Committees, Dr Volker Kleinhenz of CQU presented recent findings on the microbial loads and storage characteristics of bamboo shoots. Dr Jeff Barnes of QDPI (Bundaberg) followed this by providing data on success rates for vegetative propagation of bamboo while Mr Victor Cusack of Bamboo World related Thai experiences with the cultivation of Dendrocalamus asper, a popular culinary and timber species grown in Thailand. The first day concluded with a presentation by Mr Kevin Blackburn (NTDPIF) on the varietal comparisons and nutrition/irrigation trial set up at the Coastal Plains Research Station in the Northern Territory.

The second day attracted participants to two commercial bamboo properties north of Brisbane.  At Crystal Waters, Mr Hans Erkin of Earth Care Enterprises showed his highly productive and well managed Phyllostachys pubescens running bamboo patch, his Bambusa oldhamii clumps, and his range of other special-purpose bamboo species, while at Belli Park, Mr Durnford Dart of Bamboo Australia guided visitors through plantations of the same two species plus younger groves of Dendrocalamus latiflorus.

Staff from CQU also organised (CQU Rockhampton, 12 December 1998), or participated in (Tatura, Victoria, 23 February 1999; Eumundi, Queensland, 14 December 1998, 24 April 1999, 30 September 1999, 7 October 2000), various Bamboo Workshops during the course of the project to provide to growers and others updated information from on-going research.

2. Information gathering

We have collected c. 200 abstracts of scientific articles on bamboo agronomy overseas. Most of these were not available in Australia. As an alternative to ordering those articles through Australian channels which would have cost thousands of dollars, we used our overseas contacts to collect them. This is the most significant collection of scientific literature on the subject in Australia of which a great percentage is not written in English. After sighting our collection of 'growers booklets' with help from our own Research Officers from overseas and overseas CQU students we conclude that the information contained therein is only of limited use but will nevertheless be considered for inclusion in the manuscript. This manuscript is currently with Advances in Agronomy with anticipated publication in June 2001. The publications will also be used extensively for publication of our research results in refereed journals.

A full list of articles published in magazines, conferences proceedings and referred journals is presented in the appendix.

3. Water usage

Most, or more likely all, of the results of experimentation with bamboo published overseas, simply estimate qualitative figures for water usage by bamboo based upon many years of experience. However, there is a more pressing requirement for quantitative information about water usage of bamboo in dry continental Australia. Commercial producers of bamboo, whether for vegetable shoots and/or timber, are seeking figures to estimate demand for irrigation but there is also interest amongst governmental agencies, environmentalists and private households to review such data for the use of bamboo in wastewater dissipation.

We, therefore, quantitatively measured water usage by young bamboo (Bambusa oldhamii) cultivated soilless in nutrient solution in Rockhampton. This isolation of plants into hermetically confined structures guaranteed that processes such as leaching and precipitation under field cultivation did not add errors to measurements. Due to their size, mature plants are difficult to isolate and, therefore, water demand was extrapolated using data from our own measurements and data published in overseas literature. Based upon this, optimal amounts and timing of irrigation were determined under field conditions for two bamboo species, the monopodial (‘running’) bamboo Phyllostachys pubescens and the sympodial (‘clumping’) B. oldhamii at Bamboo Australia in Belli Park near Eumundi. We also tested performance of bamboo (B. oldhamii) under continuous, year-round supply of wastewater in Rockhampton.

3.1 Water usage by young plants

There is no quantitative information on water usage of bamboo available in overseas literature. We, therefore, grew young (2-year-old) bamboo plants of B. oldhamii in 0.16 m3 containers with sand and nutrient solution during 1998 and 1999. These isolated systems were constantly supplied with water using an automatic drip irrigation system.

While transpiration of plants generally increased during the study period studied (Figure 1) there were two peaks during the first year of investigation. Transpiration increased from September to November, decreased from November to January, increased from January to May and decreased from May to August. After August 1999, a similar pattern seemed to have commenced for the ongoing second season. It is apparent that the first peak coincided with the development of new leaves in spring. The subsequent drop in transpiration might have been due to rainfall and associated cloudiness during the mid-summer season.

3.2 Water usage by mature plants

Measuring water usage of mature bamboo quantitatively would require transplanting young plants in a big enclosure (lysimeter) and waiting several years until they reach maturity or, alternatively, uprooting mature plants and transplanting them into such an enclosure. Since both methods are beyond the means of this project, we calculated a figure using reasonable estimations from the overseas literature and from our own measurements. Water usage of plants essentially depends on their total leaf area and some authors have estimated total leaf biomass and ‘specific leaf area’ (ie leaf area per unit leaf biomass) for bamboo which makes it possible to calculate total leaf area. If those figures are known, water usage of bamboo can be calculated as the product of total leaf area and transpiration rate. This assumes that all green leaf is transpiring at the average measured rate. This may lead to an over-estimate of actual transpiration. We were unable to find estimates for transpiration of bamboo leaves in the literature and, therefore, measured this for a number of bamboo species currently grown in Australia. Table 1 summarises estimates and measurements on which our calculations of bamboo water usage were based and the source of these estimates/measurements.

The estimates for leaf area were made based on a measurement for a running bamboo species (P. pubescens) which had 3,900 culms ha-1. Clumping bamboos may have 15 culms clump-1 planted at a density of 250 clumps ha-1 (= 40 m2 clump-1 @ 6 m × 7 m planting distance). It follows that one mature clump may transpire (3,300 l m-2 × 40 m2 =) 132,000 l water year-1.

Water usage of 3,300 mm per year (last parameter in Table 1) should only be regarded as an estimate for the capacity of mature bamboo for transpiration under well-watered conditions. This figure does not imply that for optimal shoot/timber yields such quantity should be available; rather it indicates the probable maximal water dissipation rate of bamboo.

Due to water shortages under field conditions in semi-arid parts of Australia, irrigation may be feasible only prior to and during the shooting season which barely extends to a period of 6 months. Figure 2 presents the response of 4-year old (top) B. oldhamii and (bottom) Dendrocalamus latiflorus shoot yield as affected by the natural rainfall pattern during the 1998/99 season at DPIF-NT. It is clear that shoot emergence in both species coincided with peaks of rainfall. Shoot emergence and peaks of rainfall in B. oldhamii nearly overlapped whereas shooting in D. latiflorus occurred about 10 days after major rainfall events. This might be related to the greater demand for time to accumulate biomass in the relatively bigger shoots of D. latiflorus (Table 2).

3.3 Field irrigation of bamboo

3.3.1 Phyllostachys pubescens

From 1994 through 1998, an existing stand of P. pubescens (planted in 1990) was split into two experimental areas to which different rates of irrigation (‘low irrigation’ and ‘high irrigation’) were applied. Both areas were fed by a main irrigation line, but the high-irrigation area had additional rotary sprinklers. Water meters indicated that approximately 50 percent more water was applied to this area. Total amounts of rainfall as measured at DPI, Nambour were almost 1,500 mm during 1994, 1995 and 1998 but less than 1,400 and 1,300 mm in 1996 and 1997 (Table 3). Figure 3 shows the distribution of rainfall and irrigation at the two different rates during 1996, 1997 and 1998.

The graphs show that rainfall patterns were similar in 1996 and 1997 but in 1998 heavy rainfall occurred before and during the usual shooting season of P. pubescens which is from the end of September until the middle or end of October.

During 1997, tensiometers were installed in both irrigation areas and at two depths (30 and 60 cm) with four replications and during 1998 only at 30 cm. Soil moisture tension was usually lower in the low irrigation area since this area was lower lying than the high irrigation area. In the low irrigation area (Figure 4, left) soil moisture tension at 30-cm soil depth usually exceeded soil moisture tension at 60-cm depth when soils were relatively dry (ie at high soil moisture tensions). The opposite was true in the high irrigation area: tension at 60-cm depth exceeded tension at 30-cm depth during dry periods (Figure 4, right). This shows that more underground water was present in the low irrigation area. Irrigation in the high irrigation area was effective in wetting the topmost soil layer with little effect on soil water status at 60-cm depth (eg August-September).

Figure 4 Soil moisture tension at 30 and 60 cm soil depth at (left) low and (right) high irrigation in P. pubescens at Belli Park 1997

In 1998, heavy flooding occurred in mid-September. The high irrigation area provided better drainage than the low irrigation area as indicated by lower averages for soil moisture tension from August through October (Figure 5).

Data for leaf water potential were collected after a week with only 5 mm of precipitation, but 141 and 94 mm of irrigation applied in the ‘high-irrigation’ and ‘low-irrigation’ areas. Although significant only in the afternoon (1400 hours), leaf water potential was more positive in the ‘high-irrigation’ area (Figure 6). This corresponded to soil moisture tension at 30-cm depth, which was significantly more positive in that area (-21 ± 5.4 kPa) than in the ‘low-irrigation’ area (-32 ± 4.0 kPa). Soil moisture tension at 60-cm soil depth was not significantly different (‘high-irrigation’ area: -38 ± 3.3 kPa; ‘low-irrigation’ area: -34 ± 4.7 kPa), indicating that plant water status was affected by soil water status at the shallow soil depth.

Irrigation affected the number, the diameter and the weight of annually produced fresh bamboo shoots. In both irrigation treatments, numbers of shoots increased from 1994 until 1996 and decreased until 1998. During the first two years of experimentation, more new shoots were produced when less irrigation was applied (Table 2). This effect was statistically not significant in 1994 but was in 1995. After 1995, significantly more shoots emerged in the high irrigation area.

Diameter of new shoots was only measured in 1995 and 1996 and weight per shoot only measured during 1996-1998. During these periods, shoots from the high irrigation area were significantly bigger (Table 3) and heavier (Table 4). Cumulative yields under high irrigation, therefore, significantly exceeded those under low irrigation (Figure 7).

Water supply and its effects on soil water conditions primarily affected bamboo growth and yield. Without irrigation and with only 76 and 137 mm precipitation during the shoot seasons (September to October) in 1994 and 1995, bamboo growth was better in the lower-lying low irrigation area as indicated by number of new shoots (Table 2). The low annual average of soil moisture tension at 60-cm soil depth showed that there was better supply of underground water in that area. This is in contrast to the heavy-rainfall conditions in May 1996 (Figure 3) when bamboo growth and yield (Table 2) were better in the higher-lying high irrigation area with less underground water and, thus, better drainage. Well-drained soil conditions in combination with high irrigation provided optimum soil water conditions for bamboo growth. This was supported by data for leaf water potential (Figure 6).

Water supply was especially important during the shoot season. Although precipitation accumulated to similar amounts (800-900 mm) before the onset of the shoot season in September of each year between 1996 and 1998 (Figure 3), different irrigation rates exhibited tremendous differences in bamboo yield in those years (Figure 7). There was also indication that steady supply of non-excessive water volumes during the shoot season benefited bamboo production. Irrigation was more evenly distributed in 1996 than in 1997, and heavy rainfall occurred during the shoot season in 1998 (Figure 3). Bamboo yields were much better in 1996 than in 1997 and 1998, indicating that erratic irrigation and sudden rainfall created short events of overwet soil conditions which negatively affected bamboo growth and yield. These data emphasise the need to ideally plant bamboo on a friable soil that is not liable to water- logging. Well balanced water supply is of foremost importance during the shooting season. It can only be speculated to what degree water supply can be reduced to an amount that guarantees transpiration and photosynthesis during the non-shooting season. However, particularly in 1996 when almost 1,000 mm of irrigation was applied, the harvest season could be extended until the end of October.

Qiu et al. (1992) and Biswas (1988) showed how dramatically bamboo responds to increasing availability of water. Our results from 1994-1995 confirm this and suggest cultivating bamboo in a location where more water is available (eg near rivers) when rainfall is limited and/or seasonal, and no irrigation is available. This may be advantageous despite the danger of occurrence of temporary overwet soil conditions. Farrelly (1984) pointed to the detrimental effects of the latter conditions on bamboo growth and our data from 1996 to 1998 shows that a location with less underground water, but good drainage (ie the high irrigation area) is superior when combined with irrigation.

In 1994 and 1995 bamboo roots probably elongated to greater depth to take advantage of the available soil moisture in the low irrigation area. However in 1998, differences in leaf water potential of bamboo (Figure 6) were only related to differences in soil water potential at 30- cm soil depth. This indicates that plant water status depended on water status of the soil A horizon. Most roots were, consequently, located in this soil layer. This is in agreement with many studies showing that bamboo root systems are usually confined to the topmost soil layer with only a few roots extending below 40-cm depth (eg Wu, 1984 and Li et al., 1999).

3.3.2 Bambusa oldhamii

From 1997 through 2000, three experimental areas were superimposed onto two existing stands of B. oldhamii to which different rates of irrigation (‘low irrigation’, ‘medium irrigation’ and ‘high irrigation’) were applied. The ‘low irrigation’ and ‘medium irrigation’ treatments were applied onto the same bamboo stand whereas the ‘high irrigation’ treatment was applied to a separate stand. Table 7 presents average amounts of precipitation and irrigation during the shoot season of the clumping bamboo species which was between January and March of each year.

Although irrigation was applied at different rates and to bamboo stands at different locations, averages for soil moisture tension during 1998 were not significantly different between treatments (Table 8).

In contrast to the cumulative effect of higher irrigation on yields in P. pubescens, total cumulative yields in B. oldhamii were only significantly improved by greater volumes of irrigation water in 1998 (Figure 8). However, higher irrigation promoted later yields in 1998 and highest irrigation promoted earlier yields in 1999 and 2000. This could be attributed to distribution of rainfall which was higher in the early shoot season in 1998 but higher in the late shoot season in 1999 and 2000 (data not shown). Shoot weight was not significantly affected by irrigation within years but greater increases in weights between years were observed with higher irrigation (Table 9).

Yields in B. oldhamii increased from circa 1 t/ha in 1998 to almost 2.5 t/ha in 2000, indicating that plants had not reached maturity and, consequently, their full yield potential. In contrast to plants of P. pubescens which had attained their yield potential, extra supply of irrigation water to B. oldhamii was apparently not as effective. The overall difference between irrigation treatments for B. oldhamii was only 120 mm (504-622 mm), due to drought-induced lack of irrigation water. Greater differences between treatments, and greater overall yields would likely have been recorded if irrigation had supplied 500, 1000 and 1500 mm in the low, medium and high treatments.

Trials on a stand of B. oldhamii, which had not received supplementary irrigation until they were 6 years of age, only showed a response to supplementary irrigation in one of the three years of experimentation. On average, only between 60 and 180 mm/yr was applied as supplementary irrigation (due to the prevailing drought conditions over those years, and most stored water having been earlier used for P. pubescens) therefore, it was not surprising that yield benefit were minimal.

The data from Northern Territory show the close association between timing of precipitation and shoot production in B. oldhamii and D. latiflorus and the data from Eumundi suggest that water just prior and in the shoot season should reach 1000 mm for high shoot yields to be obtained.

3.4 Usage of wastewater

In 1998, bamboo (B. oldhamii) was introduced to an experimental wastewater treatment site at SW Kele & Co. Pty Ltd in Rockhampton (Figure 9). Wastewater is pumped into a holding tank and consists of two fractions: (1) treated workshop wastewater (treated graywater) and (2) household wastewater (blackwater).

Figure 9 Wastewater treatment system installed in 1998 at S W Kele & Co. Pty Ltd, Rockhampton

Graywater is untreated wastewater from the wash up area for the workers at the company. It contains high amounts of oil, grease, soaps, and human sweat and dead skin. This water is directed to a grease trap. The water is allowed to pass through a vertical grease trap until the majority of the oil and grease has been removed. This treated graywater still smells of soap; the grease trap’s effectiveness in removing soaps is unknown. Blackwater is the sewage water of a standard household after it has passed through, and has been treated in, a septic tank. This water is anaerobic and high in nutrients, such as nitrate and ammonia. It is also high in faecal coliform bacteria with measurements normally in the range of 105 – 107 organisms per 100 ml. At present the treated graywater has approximately four times the volume of the blackwater and about 400 litres of this mixture is produced daily and pumped into the system.

A mixture of treated graywater and blackwater is pumped out to the underground garden troughs. The wastewater flows into the bottom of garden troughs into a layer of gravel on top of a punched plastic film. The plastic film allows water to penetrate but essentially prevents movement of the garden soil in the upper part of the trough into the gravel. Plants root in the garden soil which is covered by another plastic film. This film is cut lengthwise to allow new bamboo shoots to emerge, but prevents rainwater entry to the system. After passing through the system of garden troughs, excessive water is collected in an outflow holding tank and pumped back to the inflow.

During the first season of B. oldhamii grown with wastewater, there were two periods of shoot growth, the first one from September 1998 to January 1999 and the second from January until June 1999 (Figure 10).

Data from the second season (1999/2000, not shown) confirmed that when the growth factor water is available in virtually unlimited amounts such as in this system, only low temperatures limit year-round growth and shoot production. It would be interesting to determine whether under conditions of permanent high temperatures such as in the tropical parts of Australia (eg the Northern Territory), effective irrigation systems could provide year-round production of edible bamboo shoots.

4. Nutrient usage

Nutrient availability is usually the most important soil chemical property governing bamboo growth and yield. The contents of available nutrients in the soil are positively related to yield and explain much of the variation in yield across bamboo-growing sites and regions (eg in China; He and Ye, 1987; Hong, 1994). Nutrient availability can be managed comparably easily through fertilisation.

4.1 Nutrient usage by young plants

As outlined in section 3.1, we grew young bamboo plants (B. oldhamii) in containers during 1998 and 1999. Field-based fertilisation experiments may not reflect actual nutrient absorption rates due to the complicated interactions of nutrients after their application in soil but in isolated systems which are constantly supplied with nutrients, these interactions can be eliminated. During 1998, different forms of nitrogen (ammonium, nitrate and ammonium nitrate) were applied to study preference of bamboo for nitrogen forms and effect of nitrogen form on water usage and absorption of other major nutrients (ie phosphorus and potassium) by bamboo.

The form of nitrogen applied had a significant effect on water usage of bamboo. Understandably, application of any form of nitrogen accelerated water usage of bamboo (Figure 11). However, plants absorbed more water when only NH4-N was applied. This might be explained by the preference of plants to the molecular state of nitrogen: in contrast to dicotyledonous plants, graminaceous plants have comparably low values of ion-exchange capacity and are, therefore, more effective in absorbing monovalent cations (NH4+) than divalent cations (NO32+). This is known for grasses such as rice but has never been shown for bamboo.

In many overseas field-based experiments, the optimum ratio between N, P and K for fertilisation purposes was determined. Results from such trials depend on soil parameters and particularly the nutrient-supplying capacity of soils. A much more limited number of studies has covered results of nutrient analysis in different plant parts of bamboo and these studies generally agree in that K is the nutrient most significant to bamboo growth, followed by N and to a much lesser degree P. In our direct measurements of absorption of these major nutrients by bamboo, this was confirmed (Figure 12). The N:P:K ratio between absorbed ions was 1.0:0.4:1.3 after 16 months of cultivation in nutrient solution.

4.2 Nutrient usage of Phyllostachys pubescens

From 1994 through 1998, three fertiliser treatments: (1) a ‘standard’ rate, (2) 1.5 times the ‘standard’ rate and (3) 2.0 times the ‘standard’ rate were applied to an existing stand of P. pubescens which was irrigated with two different rates (section 1.3.1). Individual irrigation areas were split into four replicate rows, with three plots along each row randomly assigned one of the fertiliser treatments. From 1994 through 1996, inorganic fertiliser was applied at a (‘standard’) rate of 250:50:141 kg ha-1 year-1 N:P:K and during 1997 and 1998 composted chicken manure (approximately 130:90:50 kg/ha/year N:P:K) was used. Fertiliser was applied in July before the shoot season.

After three years of N application, soil N was not significantly different between N application rates (Table 10). In contrast, leaf N increased significantly with higher N rates. All leaf N concentrations were much greater compared with a nearby non-fertilised bamboo stand (1.70 ± 0.054% N).

N application treatments had no significant effect on shoot numbers in the ‘low irrigation’ area throughout the study period. This was so for the ‘high irrigation’ treatment except in 1996 when the number of new shoots in the “2.0 × ‘standard’ rate” treatment significantly exceeded that in the 1.5 x standard treatment rate (Table 11). There was no significant effect of N application rates on weight of shoots (Table 12).

Figure 13 presents cumulative shoot yields in the two irrigation areas from 1996 to 1998. These data reflected data for number of shoots (Table 12): higher fertiliser application improved shoot yields only in 1996 in the ‘high irrigation’ area. Since the latter was due to a significantly greater number of shoots (Table 11) and not to significantly greater shoot weight (Table 12), it could be concluded that nitrogen improved yield due to the effect on number of shoots rather than their weight. When considering all data, there was only a trend that greatest yields in the ‘high irrigation’ area were achieved under the highest N application rates.

Figure 13 Effect of irrigation and fertiliser rate on cumulative shoot yield of P. pubescens at Belli Park 1996-1998. Vertical bars represent LSD.

Plant growth generally follows a ‘minimum factor’ concept, ie there will be no response to growth factors other than the growth factor which restricts growth most significantly. Bamboo did not respond to greater N rates when soil water conditions were not optimal. Therefore, bamboo did not respond to more N in the better-drained ‘high irrigation’ area when there was only little precipitation and no irrigation in 1994 and 1995. Underground water was beneficial for shoot production in the ‘low irrigation’ area in those years. However, water supply was probably insufficient for optimal bamboo growth and, therefore, response to N. During the following years, occasional overwet soil conditions and suboptimal soil moisture due to less irrigation restricted N uptake of bamboo in the ‘low irrigation’ area. Productivity and, consequently, N uptake was much greater in the ‘high-irrigation’ area. Nutrient availability, however, limited bamboo growth and yield only in 1996 when better soil water conditions provided greater potential for biomass production. Leaf N concentrations reflected greater N application rates (Table 8), which were associated with greater shoot production. Nitrogen, which was not absorbed by bamboo, was probably leached; hence, there were no differences in soil N between fertilization treatments after 3 years of imposing treatments. It can be concluded that fertilisation can only improve bamboo growth if soil water conditions are well managed. Application of 130, 250, and 500 kg ha-1 year-1 N supported annual shoot yields of 6, 8-9, and 11-12 t ha-1 in P. pubescens, revealing the great nutrient demand of bamboo but also a ‘diminishing-return’ relationship between nutrient application and shoot yield.

4.3 Diagnosis Recommendation Integrated System (DRIS) for N

Effects of nutrient application on response in bamboo depend on a large number of factors including soil-related factors (eg adsorption and absorption capacities, nutrient-supplying power and soil water relations), plant-related factors (eg species) and numerous other environmental factors (eg temperature). There can be little doubt that analyses of plant tissues are a much better indicator of the nutrient status of crops than are soil analyses. Response of crops to available nutrients in the growth media usually follows a hyperbolic relationship with great yield response to nutrient application at low plant nutrient levels and a decreasing rate of response with increasing plant nutrient contents (the ‘Michaelis-Menten’ relationship). This theory has been applied to nutrient relationships in many crops and is valid for most plant nutrients. Therefore, tissue analyses have been adopted to guide fertilisation in many important annual and perennial crops. The last fully expanded leaf is often used as an index tissue since this plant part is usually the best indicator of the current nutrient status of the plant during its active growth phase. This method can be used across bamboo species and cultivation sites free of extreme conditions (eg drought or waterlogging). If fertilisation is to be based on plant analysis, there must be calibration of application rates to plant response. To calibrate this curve for specific site conditions, response of bamboo to varying nutrient application rates should be measured on individual sites to define response curves locally. This approach has been named ‘Diagnosis Recommendation Integrated System’ (DRIS) and is the basis for fertilisation in many important plantation crops such as oil palm, coffee and cocoa but also annual crops such as vegetables. For those crops, relationships and cross- relationships between macronutrients and even micronutrients have been studied to guide fertilisation. As a first attempt to develop a DRIS for bamboo, we developed a system (a response curve between leaf nitrogen and application rate of fertiliser N) for the element ‘nitrogen’ which is the second-most significant nutrient for growth of bamboo.

In the first experiment, four rates of inorganic nitrogen (250, 500, 750 and 1000 kg N ha-1 as urea) were applied to single clumps of B. oldhamii in February of 1998. Soil nitrogen analyses proved to be ineffective in showing differences between treatments, eg one month after application of fertiliser, only 0, 0, 35 and 7 ppm of available N could be found in the soil for application rates of 250, 500, 750 and 1000 kg N ha-1, respectively. This indicates that either the root system of bamboo is very effective in absorbing available soil nitrogen, or less likely that leaching and/or immobilisation removed N not used by bamboo. Analyses of total N in the last fully expanded leaves revealed differences between fertiliser treatments and are, therefore, effective measures of the nutritional status of plants. Concentrations of leaf N increased with increasing N application rates (Figure 14). Differences between 500, 750 and 1000 kg N ha-1 were small, but leaf N was lower when only 250 kg N ha-1 were applied. Leaf N decreased with time after application of fertilisers.

Before transplanting B. oldhamii to the wastewater treatment facility (section 3.4), leaf N concentration was below 1.5 percent but increased to about 3.5 percent within three months. Concentrations of nutrients in the growth medium were excessive since not all nutrients were absorbed. Therefore, N concentrations around 3.5 percent would indicate the likely ‘maximum’ concentration of leaf N in B.oldhamii. In fact, the highest N concentration that we have ever measured in this species was 3.617 percent, after application of 1,000 kg ha-1 N to a clump of B. oldhamii (Figure 14).

In a similar experiment to the abovementioned with B. oldhamii, we applied 300 kg ha-1 N to a stand of P. pubescens at Dayboro. There was a quick and significant response compared to the control one month after N application (Figure 15). Leaf N reached a maximum after two months and was lower thereafter. Rapid absorption of N was associated with increased photosynthesis and, therefore, biomass accumulation as indicated by leaf C concentrations (Figure 16). Leaf C increased in the fertilised plots whereas C in the unfertilised plots decreased.

Figure 17 presents a summary of leaf-N data of different bamboo species at different locations in Australia. We roughly estimated a ‘lower limit’ for leaf N at about 2.0% N since the colour of leaves of several bamboo species at or below this level appeared yellow to light green. N concentrations at or above the ‘upper limit’ of 3.5% N were only measured when nutrient supply was excessive, eg after application of ≥ 500 kg ha-1 N and in the wastewater trial. The ‘optimum’ at 3.0% N may be suggested as an orientation point for growers.

Figure 17 Leaf N concentrations in different bamboo species at various locations in Australia

In our studies with B. oldhamii and P. pubescens, leaf N did not decrease for about four months after application of nutrients. Therefore, 4-6 months could be a suitable time interval for analysing leaf N and accordingly applying nutrients. These events can be scheduled to coincide with the onset of distinct growth phases in bamboo, (1) the ‘shoot season’ and (2) the subsequent phase of rhizome growth. An additional measurement could be accommodated in (3) early spring for the clumping bamboos.

The response of fertilisation in leaf N 3-4 months after application was similar across the two bamboo species (Figure 18).

The similarity in response to nutrient application in bamboo across species is not surprising. In comparison with plenty of other plant genera, it appears that the root system of bamboo is to an overwhelming percentage confined to a very shallow soil depth of about 40 centimetres in which it develops a root density that is much greater than in many other plants (Kleinhenz and Midmore, 2001).

This growth habit makes fertilisation in bamboo less affected by soil conditions and can explain our findings that under field conditions, bamboo is able to absorb nutrients up to very high rates.

The calculated hyperbola may be suggested as a guide to how much N fertiliser needs to be applied to raise N concentrations in bamboo leaves. The data points at 0 kg ha-1 N application represent the natural ‘fertility’, ie N mineralisation rate of the particular soil. The leaf-N value for P. pubescens was derived from the plantation in Dayboro and the value for B. oldhamii in Belli Park. The small differences in leaf-N values at 0 kg ha-1 N can be explained by soil ‘fertility’ which was lower in Dayboro but higher at Belli Park. The regression line indicates that relatively little N is required to lift leaf-N from low levels but more from higher levels. The equation estimates the theoretical maximum leaf-N concentration at 4.13 percent. Our suggested ‘optimum’ of 3.0 percent leaf N is at a level below which bamboo has a great affinity for N and above which plants respond only slightly to additional units of N application.

The following example resulting from our research with B. oldhamii may demonstrate a practical application of the suggested fertilisation strategy for farmers. When leaf N in unfertilised plants was 1.8 percent, (360 — 120 =) 240 kg ha-1 N was needed to raise leaf N to 3.0 percent. If leaf N would have dropped to 2.4 percent after 4-6 months, (360 — 180 =) 180 kg ha-1 N would have been required to raise leaf N back to 3.0 percent. From this, the total annual N application rate for 2-3 applications per year after the first year would sum up to about (2-3 × 180 =) 360-540 kg ha-1year-1N. This range confirms our first estimates for optimum fertiliser application in 1997.

We tested this DRIS application for different leaf-N levels (Control, 2.5, 3.0 and 3.5% N) in B. oldhamii in three sites, Belli Park, Dayboro and Iveragh. Size of clumps decreased in this direction as indicated by number of culms clump-1 * culm diameter (Table 13).

N fertiliser rates were calculated based upon a first measurement of leaf N in November and were applied in early December. Across locations, average N application rates were 0, 64, 158 and 307 kg N ha-1 for the control treatment and target leaf N levels of 2.5, 3.0 and 3.5%. Figure 19 presents the course of leaf N of B. oldhamii at the three sites from November 1999 until March 2000. Leaf N principally increased with greater N application rates and target leaf N levels were nearly reached in December 1999, with the exception of the target level of 3.5% leaf N in Dayboro (Figure 18).

Figure 19 Course of leaf N concentrations at three sites for different leaf N target levels of B. oldhamii

Figure 20 shows the inverse relationship between number of culms and their diameter, in clumps not harvested for their shoots, as governed by the target N level. Higher N levels promote a greater number of new shoots whereas less N promotes their diameter. A target N level of just below 3.0% produced a balanced set of both number of shoots and their diameter.

Yield and shoot weight were only measured in the more mature bamboo stand at Belli Park. Differences between treatment means were not statistically different (Table 14).

4.4 Usage of ions from wastewater

Details of bamboo cultivated in the wastewater facility in Rockhampton have been mentioned in sections 1.4 and 2.3. Table 15 presents data for element concentrations in leaves of B. oldhamii several months after having been transplanted to the sewage system. Many element concentrations increased significantly (exceptions were P, K, S, and Sr which significantly decreased), in spite of the greater size of plants over the same time frame.

5. Culm management

From 1997 to 2000, we conducted culm management experiments in an existing stand of B. oldhamii at Bamboo Australia, Belli Park. The treatment was ‘standing-culm density’ with four levels, 1,200, 2,400, 3,600 and 4,800 standing culms ha-1. There were 400 clumps ha-1, although Chinese recommendations for B oldhamii are 625 clumps/ha. Each plot comprised three clumps arranged in a one-factorial design with four replications. Only culms of up to 3 years of age were used and the number of culms for each age class was kept constant (ie 1-1-1, 2-2-2, 3-3-3 and 4-4-4 culms of 1-2-3 years of age).

Each year, the number of felled standing culms (excess new culms and all 4-year old culms) was recorded. Since it was too laborious to determine directly, their biomass was estimated by sampling in 2000 (Table 16). The greater biomass of the 4-years old culms was to the greater part due to the higher specific gravity which was caused by their lower water content (data not shown). Those estimates did not account for increases in culm length and diameter with stand age.

During the shoot season of each year, total yield of fresh edible shoots and their weight were recorded. Maximum yields increased from 0.9 t ha-1 in 1998 to 1.4 t ha-1 in 1999 to 2.5 t ha-1 in 2000 (Figure 21). The lowest standing-culm density (1,200 culms ha-1) performed least in 1998, best in 1999 and least in 2000 again. In contrast, the highest standing-culm density (4,800 culms ha-1) was the best treatment in 1998 but the worst in 1999. The medium high standing-culm densities (2,400 and 3,600 culms ha-1) performed more stable between years.

In addition to marketable yields, Table 17 presents data for number of harvested bamboo shoots and for shoot weight. Although differences were usually not significant, there was a trend that greater shoot numbers resulted in less shoot weight. It appears that management techniques that promote greater shoot numbers result in higher yields but reduced individual shoot yield (see sections 4.2 and 4.3). The yearly increases in yield were due to a greater number of shoots harvested rather than increases in shoot weight. Differences in total yield, number of shoots and average shoot weight were statistically not significant but there was a trend that yield of fresh, edible bamboo shoots was promoted by lower standing-culm density.

Table 18 presents the effect of standing-culm density on timber yields. Understandably, yield of 4-years old culms increases significantly with standing-culm density since more culms were allowed to mature at greater density. However, the opposite was true for culms <1 year old: yield was higher at lower standing-culm density. This reflected the farmer’s practice of ensuring survival of required new culms by allowing culms in excess to elongate. Four-year old culms of sympodial species such as the B. oldhamii of this experiment have reached maturity for multipurpose timber use. However, the use of <1-year old culms is more restricted and they are particularly suited to pulping and papermaking, for culms from one to six years old may be used as raw material for pulping (Shanmughavel and Francis, 2001).

The diameter of harvested culms was only determined for the <1-year old culms. Table 19 shows that except in the first year after imposing treatments, there was a trend that higher standing-culm density favoured greater culm diameter.