HempCyberFarm South Australia

Background
Until the late nineteenth century industrial hemp (Cannabis sativa L.) was the world's largest agricultural crop and the most traded commodity. It was grown for the production of fiber and seed. Because of its fiber strength, it was the preferred source of fiber for the manufacture of rope, fabric and paper, while oil was extracted from the seed.

With the mechanization of cotton processing, cotton became the main fiber source. At the same time the decline of sailing ships decreased the market for rope and canvas sails. This decline continued in the 1930s with the introduction of synthetic fibers such as nylon.

During the past decade there has been a resurgence of interest in hemp, particularly in Europe for the production of paper, particleboard, fiberglass replacement products, geo-textiles, environmental matting and textiles.

Both industrial hemp and marijuana are forms of Cannabis sativa L. The important difference between them is that Industrial hemp has a low narcotic level, too low to produce any psychotropic effect.

Industrial hemp is a tall, annual herbaceous plant with a deep taproot, growing to a height of 1 to 4 m high depending on variety and growing conditions. It has a single slender main stem and when grown at commercial crop densities the stems are generally un-branched. Stem thickness varies from 4 to 20mm in diameter.

Hemp is normally dioecious, meaning it has separate male and female plants, but monoecious varieties have been bred.

The seed is small, brown to grey in color and has an oil content of about 35% and a protein content of 25%. The stem fibers can be divided into two very different types of fibers, the bark or bast comprising the outer stem tissues and the core or hurd consisting of the inside stem tissues.

The bark or bast contains primary and secondary bast fibers, which are tick walled, have a high cellulose and a low hemi-cellulose and lignin content Primary bast fibers are 5-50mm long, while secondary bast fibers are shorter, being about 2mm long and are more lignified, Primary bast fibers are about 34um and secondary bast fibers about l7um wide. The bast composition is approximately 67% cellulose, 13% hemi-cellulose and 4% lignin. Bast comprises 30-35% of the stem dry matter. As the bast is the most valuable component for paper or textile production bast fiber quality is of prime importance.

The woody core fibers or hard are thin walled (average 2um) and short (0.55mm) and have a chemical composition which resembles hard wood being approximately 40% cellulose, 20% hemi-cellulose and 20% lignin.

Paper production requires high quality bast and hurd fiber containing a long fiber length, a high cellulose content and a low lignin content.

Hemp is used to produce specialty printing and writing papers or as an additive to strengthen and improve quality in wood and straw based paper manufacture.

In 1991 the Tasmanian Government was the first Australian State to grant a research license enabling evaluation of the agronomic and market potential of a viable industrial hemp industry. Industrial hemp research is currently being conducted in South Australia, Tasmania, Victoria and New South Wales. In SA research commenced in 1995 to determine the agronomic potential as well as the processing and market potential of industrial hemp. To enable the potential of industrial hemp under SA conditions to be evaluated a research license has been issued under Section 56 of the Act to the Yorke Regional Development board with the research being conducted by South Australian Research and Development Institute and IAMA Technical Services. The trials were conducted under stringent Security conditions and in accordance with an agreed trial protocol.

Three trials were to be sown, at Turretfield Research Center in the lower North, Kybybolite in the South East and Arthurton on Yorke Peninsula. Five French hemp varieties were to be sown at six weekly intervals between May and October. All trials were grown wider dryland conditions except for Kybybolite where trials were irrigated The trials at Arthurton and Turretfield were unsuccessful as June and July sowings flowered too early and produced negligible plant growth. Sowings in October did not emerge due to dry soil conditions.

At Kybybolite five varieties of French hemp were sown over a range of sowing dates. The site had been prepared before winter and 2.5 t/ha of lime had been incorporated to raise the soil PH to about 6. Plot Size was 8 rows at 18cm row spacing by 10 m long and 4 replicates were used. buffer plots were sown on either end of each time of sowing subplot. The experimental design was a split plot with time of sowing as main plots and varieties as subplots. Sowing rate was adjusted for seed size and germination percent and seed was sown about 2 cm deep. At sowing DAP at 144 kg/ha was applied and after emergence additional nitrogen at 50 kg/ha was top- dressed on all plots. Irrigation at about 25 mm per week was applied by a travelling irrigator, higher water application rates could not be used because the different times of sowing meant that small seedlings were often present. Plant emergence was similar for all varieties with about 200 plants per square meter being achieved.

At harvest, the outside 2 rows on either side of the plot were discarded and the inside 4 rows were cut at ground level and weighed. A sub-sample of 1 kg was kept, and oven dried to determine dry weight. Results are detailed in Table 1.

Table 1. Total production (t/ha) of dry matter by 5 hemp varieties sown at 5 dates at Kybybolite 1995

Sowing date	Futura 77	Ferimon	Fedrina 74	Fedora 19	Felina 34
11 September	5.182	3.605	4.195	3.477	3.955
5 October	9.981	5.229	8.156	6.282	6.530
15 November	8.875	4.983	6.098	6.797	6.699
4 December	4.795	3.691	4.672	4.299	3.617
11 December	5.540	3.406	4.175	3.485	4.114
Mean	6.874	4.182	5.459	4.868	4.983

LSD (0.05) for comparisons of variety means within a time of sowing, 1.300
LSD (0.05) for comparisons of variety means between times of sowing, 2.030

The trials have confirmed that industrial hemp is a short day plant suited to spring sowing. Attempts to sow the crop in winter were unsuccessful as plants grew to only about 30 cm high before flowering. These sowings occurred in 2 June and 10 July and conditions during winter were very wet. Plants flowered during September and early October and these sowings were destroyed.

The crop appears to be best adapted to either irrigated areas or areas where summer crops are traditionally grown. Under irrigation, mid to late spring plantings gave the best growth. Results from Kybybolite showed that total dry matter yields of October and November sown hemp ranged from 9-10 t/ha for the variety Futura 77 down to 5 t/ha for the variety Ferimon. The other varieties produced intermediate yields. Futura 77 produced the highest yields at all harvests and appears the best-suited variety of the five under test. Hemp sown after November and before October grew poorly so that the optimum sowing date appears to be during October and November. The 10 t/ha produced in this trial compares quite favorably with yields obtained overseas. The only areas in SA where industrial hemp may warrant farther evaluation as a dryland crop are in the mid and lower South East.

Attempting to sow hemp before October was unsuccessful because the plants flowered early and produced little dry matter. Sowings during October and November produced substantially more growth because plains had a longer period from emergence until the start of flowering. December sowings again resulted in more rapid flowering and reduced plant growth. The variety Futura 77 produced the most dry matter at all sowing dates, while all other varieties produced less. In order to produce higher levels of plant growth it may be necessary to grow varieties that flower even later than Futura 77. Similar results have been found in Tasmania (S. Lisson- pers.comm.).

Weeds presented few problems in hemp production when sowings occurred in October and November. Plants grew rapidly and smothered weeds to produce a weed free stand. However, in earlier sowings. weeds competed with hemp. Grasses were killed with normal grass herbicides with no evidence of damage to the hemp, however broad leaf weeds could not be controlled.

Insect pests had little effect on hemp growth. Lucerne flea required control in early sowings at all sites but later sowings at Kybybolite were unaffected.

Soil conditions appeared to have little effect on emergence at Kybybolite. Likewise overhead watering also seemed to be adequate for plant growth. However, in Victoria, plant growth was severely reduced by waterlogging in irrigation bays (D. Pye, pers-comm.) and obviously waterlogging could be a severe impediment to hemp production. In the trial at Kybybolite the plants were grown with similar nutrient inputs to a well grown canola crop, however no nutrition trials were conducted so perhaps yields could have been further increased with higher inputs.

Plants were not to be allowed to set seed in the South Australian trials so that seed yield can not be estimated. However studies in Tasmania have suggested that gross returns of $100-1200 per hectare are possible from producing oil from hemp seed.

The Yorke Regional Development Board feasibility study on end-use and market potential identified the high cost of processing equipment as the main limiting factor. They conclude an industrial hemp industry in the medium term would not be viable, as hemp products would not be competitive against products made from other fibers e.g. wood or cotton.

There is strong community interest in industrial hemp from environmental groups. A hemp conference was held in Melbourne in December 1995 to identify the issues that need to be addressed before a hemp industry is established. Post farm gate processing needs further investigation and development. Fuelled by this enormous community interest further industry research and development may be possible despite the lack of apparent viability of a hemp industry.