Case Study 12 Pollination Aware Cotton

This case study is the primary source of information on potential pollination services for the industry. It is based on data provided by industry, the ABS and other relevant sources. Therefore, information in this case study on potential hive requirements may differ to the tables in the Pollination Aware report (RIRDC Pub. No. 10/081) which are based on ABS (2008) Agricultural Commodities Small Area Data, 2005-06.

Introduction Cotton (Gossypium hirsutum) is grown worldwide for its soft, Cotton was first brought to Australia in the late-seventeenth staple fibre that develops around the seeds of the plant into century by the first fleet and since then has grown into a a form known as a boll. The shrub is native to tropical and significant industry. Today, not only is cotton an important part subtropical regions around the world, including the Americas, of many regional communities, it also contributes $1.7 billion to India and Africa. The fibre is most often spun into yarn or thread the national economy each year, despite competition from large and used to make a soft, breathable textile which gives it the subsidised cotton producers such as the USA and (CA status of the most widely used natural-fibre cloth in the world 2009). In Australia, most on-farm cotton activities occur from (Rhodes 2002). Additionally, cotton seeds can be crushed for oil September/October/November (planting) to March/April/ or animal feed and the leaves may be used as mulch (CA 2009). May (harvest).

Cotton production in Australia All Australian cotton is grown in and Queens- ble, high-quality, low-contaminant cottons that attract a premium land (see Table 1 and Figure 1) on about 800 farms, the majority on the world market. Up to 98% of cotton grown in Australia is of which are run by families (CA 2009). The major production exported, mainly to Asian spinning mill customers (ICAC 2007) areas in New South Wales stretch from the Macintyre River and in 2005/06 China, , Thailand, South Korea and on the boarder and cover the Gwydir, Namoi and Japan were the main markets for Australian cotton (ABS 2008). Macquarie valleys as well as along the Barwon and Darling Rivers In 2006/07 a total of 487,000 tonnes of raw cotton, valued at in the west and the Lachlan and Murrumbidgee rivers in the around $832 million, was exported; however, output in this south. In Queensland, cotton is grown mostly in the south, in year was severely drought affected and was only about 64% of the Darling Downs, St George, Dirranbandi and Macintyre Valley ‘normal’ production (ICAC 2007). The enormous growth in regions. The remainder is grown near Emerald, Theodore and Australia’s cotton production in recent years has coincided with Biloela in central Queensland. growth in the textiles markets of some of Australia’s nearest Although a relatively small producer on the world scale, Australia trading neighbours in Asia (CA 2009). is the world’s third-largest cotton exporter, producing sustaina- Table 1 Cotton production in Australia (ABS 2008)

NSW QLD Total Irrigated – area (ha) 168,863 100,844 269,707 Irrigated – cotton lint production (tonnes) 329,347 192,328 521,675 Irrigated – seed cotton production (total module weight) (tonnes) 871,467 490,335 1,361,801 Non-irrigated – area (ha) 28,602 28,885 57,487 Non-irrigated – cotton lint production (tonnes) 15,981 0 15,981 Non-irrigated – seed cotton production (total module weight) (tonnes) 44,712 0 44,712 Case Study 12 Total area (ha) 197,465 129,729 327,194 Total cotton lint production (tonnes) 345,328 0 345,328 Total seed cotton production (tonnes) 916,178 0 916,178 Pollination Aware

Figure 1 Production regions within Australia (ABS 2008)

F igure 1 Production regions within Australia (ABS 2008) Table 1 Pollination in cotton Cotton production in Australia (ABS 2008) Cotton is commonly regarded as being a partially cross-pollinated crop, and largely self-fertile and Irrigated – area (h) self-pollinating (McGregor 1976), although introducing insect pollinators intoNSW the crop during QLD Total flowering has resulted in increased quantity and quality of cotton lint and seed (McGregor 1976; Irrigated – area (h) Rhodes 2002; Moffett et al. 1975). Each cotton flower is only available for168,863 pollination on the100,844 one day 269,707 the flower opens, with the corolla and stiminal column falling off on the second day (Rhodes 2002). Irrigated – cotton lintNectar production is secreted (tonnes) by floral nectaries inside the flower and by extra floral329,347 nectaries on the outer192,328 or 521,675 sub-bracteal, foliar, and unipapillate (microscopic) areas on the flower peduncles and young leaf Irrigated – seed cottonpetioles production (McGregor (total 1976) ,module with all nectar weight) secreting (tonnes) sites shown to be attractive871,467 to foraging honey490,335 bees 1,361,802 (Rhodes 2002). Non-irrigated – area (ha) 28,602 28,885 57,487 Non-irrigated – cotton lint production (tonnes) 15,981 0 15,981 Non-irrigated – seed cotton production (total module weight)119 (tonnes) 44,712 0 44,712

Total area (ha) 197,465 129,729 327,194 Total cotton lint production (tonnes) 345,328 0 345,328 Total seed cotton production (tonnes) 916,178 0 916,178

Pollination in cotton Cotton is commonly regarded as being a partially cross-pollinat- nectaries on the outer or sub-bracteal, foliar, and unipapillate ed crop, and largely self-fertile and self-pollinating (McGregor (microscopic) areas on the flower peduncles and young leaf 1976), although introducing insect pollinators into the crop petioles (McGregor 1976), with all nectar secreting sites shown during flowering has resulted in increased quantity and quality to be attractive to foraging honey bees (Rhodes 2002). of cotton lint and seed (McGregor 1976; Rhodes 2002; Moffett et al. 1975). Each cotton flower is only available for pollination About 50 ovules must be fertilised if a full complement of seeds on the one day the flower opens, with the corolla and stiminal is to be produced; therefore, at least 50 viable pollen grains column falling off on the second day (Rhodes 2002). Nectar is must contact the stigma (McGregor 1976). Insect pollination, secreted by floral nectaries inside the flower and by extra floral in particular honey bee pollination, can aid in this process and re- sult in higher yields and better quality lint (Rhodes 2002; Kaziev

Page 2 Cotton

1960). Several authors have shown that honey bees are effective and total number of seeds per sample (16.5%) for plots receiv- pollinators of cotton (Rhodes 2002; Moffett et al. 1975; Kaziev ing the highest number of bee visits compared to plots receiving 1960). Benefits of honey bee pollination include increases in the lowest number of bee visits (Table 2). the percentage of bolls per 100 flowers, more seeds per boll, more seed cotton Table 2 Values for variable measure for the three per boll and more seed cotton per flower cage treatments (Rhodes 2002) (Rhodes 2002). Improvements in lint quality characteristics such as fibre strength Variable Control Cage with Cage with- and length have also been demonstrated (open plots) bees out bees (McGregor 1976). Total boll mass (g) 29.5 19.4 9.8 Rhodes (2002) found significant increases Total seed mass (g) 16.3 10.0 4.9 in the total number of bolls harvested Total lint mass (g) 14.2 7.8 4.1 (11.1%), total mass of bolls (16.5%), total lint mass (15.8%), total seed mass (19.7%) Total number of 246 117 76 seeds per sample

Pollination management for cotton in Australia There are a number of factors within the field which have a Density of bees direct bearing on the pollination efficiency of honey bees: There is little in the way of recommendations as to honey bee management on cotton crops apart from recommendations of one hive per hectare by Ward and Ward (2001). Rhodes (2002) Crop layout and Moffet et al. (1975) both recommend the use of honey bees Row spacing and blossom density • : Cotton seeds are most on cotton crops but give no indication as to the density of hives commonly planted in spring, in long rows using specialised required for adequate pollination. Rhodes (2002) used an apiary farm machinery. Seed capsules (bolls) can grow as big as of 30 hives on a 47ha plot of irrigated cotton (0.6 hives/ha) an egg whereby after a further 35–55 days, the bolls burst which showed significant improvements in quality and quantity open and expose the cotton. The point at which most of the of cotton lint and seeds. bolls are open indicates that the crop is ready to pick (Ward and Ward 2001). • Access: From a beekeeper’s point of view, all-weather truck Arrangement of hives access is highly desirable. Limited access may lead to an Most seed is set within a 100m radius of a colony. Research has increased workload for the beekeeper, uneven placement of suggested colonies should be deposited in groups of 12–18 per hives and thus inefficient pollination. location, with about 150m between locations(Somerville 2005).

Page 3 Case Study 12 Pollination Aware

The distribution of honey bees within the crop suggest that bees content of one sample of nectar collected from the sub-bracteal tend to forage around the edges of the crop. Use of a greater nectar produced a reading of 17% sugar content, which is an number of bee hives may improve the result by increasing honey acceptable level for foraging bees (Robertson and Rhodes 1992). bee activity in the centre of the crop. Cotton honey is light in colour and flavour and like all flower Whatever the distribution, pattern hives should be placed in honeys, has a tendency to crystallise fairly rapidly. This is shady areas to avoid extreme temperatures. Honey bees collect because tree honeys generally granulate more slowly and some significant amounts of water for use in the hive and as tempera- of the sugars are stable as a crystal, not as a liquid (Rhodes 2002; tures rise, the need for water increases, diverting many field bees Robertson and Rhodes 1992). into water gathering duties. Ensuring hives are located relatively close to water and in shady areas will significantly reduce stress levels of colonies, aiding in optimal pollination of the target Availability of honey bees for pollination crops (Somerville 2005). While cotton may be a significant crop for the grower, it is not seen as a primary resource by apiarists. The nutritional value of cotton has been said to be acceptable for honey bee foraging, Timing having a sugar content of 17% (Robertson and Rhodes 1992), Australia’s cotton growing season lasts approximately six although research describing pollen and nectar characteristics is months, starting in September/October/November (planting) lacking. Use of honey bees in cotton is not common in Australia and ending in March/April/May (picking) – depending on thus little information exists on pollination contracts and the region. Each cotton flower is only available for pollination associated costs. on the one day the flower opens, with the corolla and stiminal column falling off on the second day (Rhodes 2002). Thus to ensure sufficient pollination, bee hives must be available after Feral bees around 5% of the cotton crop has begun flowering and stay until The majority of recent research in Australia identifies significant flowering has ceased (Robertson and Rhodes 1992). production benefits when introducing honey bees into cotton crops. Cotton growers relying on feral bees for part or all of their pollination services should be similarly aware first, that Bee husbandry in the paddock feral colonies are unlikely to be at full strength at the time that Moving hives into a crop during the night is less stressful on cotton flowers bloom and, second, that even if they were, forag- the bees, because they are not flying and the representatives are ing by these bees is unlikely to be sufficiently intense to achieve generally cooler. the level of pollination required for optimal seed production.

Attractiveness, nutritional value of Risks pollen and nectar Pesticides: Placing hives well back from the crop also may help In cotton, honey bees collect nectar from extra-floral nectaries the grower. If a crop needs spraying with pesticide the location as well as floral nectaries and may be foraging in a cotton crop of the hives is crucial. The further the beehives are placed away before, during and for a short time after the crop has flowers from the crop the better. If spraying is necessary, then this present. Honey bees require nectar and pollen to satisfy their should be conducted in late afternoon or evening when foraging nutritional requirements. The quality of pollen collected by bees have ceased their foraging activities. One of the biggest honey bees directly affects the welfare of a colony. Poor quality dangers of placing bees near any agricultural crop is the possibil- or low volumes of pollen can equate to shorter lived bees and ity of colonies or field bees being sprayed by pesticides. a colony that has a greater susceptibility to disease. The sugar

Page 4 Cotton

It is strongly recommended that growers take the following steps activity tends to remain at a relatively high level. Decreases in to prevent or reduce bee losses: both numbers of bees visiting blossoms and the distance from the hive at which bees forage occur with a decrease in temperature. • Apiarists should always inform cotton growers where hives Under rainy conditions bees fly between showers but only are located usually for very short distances. Wind, particularly strong wind, • Cotton growers should always inform beekeepers when they tends to reduce the ground speed of bees and hence reduces the are likely to spray and with what insecticides, so that the number of flights per day. beekeepers may take precautionary measures. Due to the logistics involved in moving an apiary, beekeepers require Colony strength will also have a direct bearing on the tem- as much warning time as possible and, where practical, 24 perature at which honey bees will leave the hive. Only strong hours warning is required. colonies will fly at lower temperatures. Bees need to keep their brood nests within their hives at a constant temperature of 37oC. • Apiarists should position their hives deep within tree stands The cooler the external temperature, the more the bees are where practicable to minimise possible spray drift impacts. required within the hive to maintain that temperature. Hence if Apiarists should avoid locating hives on the prevailing the colony is strong in numbers the surplus bees not required for downwind side of cotton fields. maintaining hive temperature are available for foraging duties. • Where possible and practical, cotton growers should restrict spray activities (within flight range of apiaries) until after Environmental factors have a direct bearing on the amount 4.00pm by which time the numbers of honey bees present of nectar secreted. It has also been found that nectar is the in a cotton crop would be at low levels. Growers and most concentrated in old flowers about to wither, but nectar applicators need to be aware, though, that it is in the early concentration fluctuates widely in accordance with the relative mornings and late afternoons that inversion layers are either humidity throughout the day. The number of honey bees that present or forming and this can increase the likelihood of visit the blossom has been directly correlated with the amount spray drift. and concentration of nectar produced.

Weather Temperature and rainfall have a marked effect on honey bee activity. Bee activity is very limited below temperatures of 13oC with activity increasing up to around 19oC, above which

Page 5 Case Study 12 Pollination Aware

Potential pollination service requirement for cotton in Australia Optimal use of managed pollination services in all cotton planta- tions in Australia would require a service capacity as indicated in Table 3 below.

Table 3 Potential pollination service requirement for cotton in Australia State Month Area (ha) Average hive Estimated Number of (flowering) density (h/ha)* hives required NSW September 197,465 0.6 118,479 QLD October 129,729 0.6 77,837 Total 327,194 196,316 Notes: * Area sourced from ABS (2008) Agricultural Commodities Small Area Data, Australia 2005-06, flowering times and average hive density estimated from McGregor (1976)

References AUSTRALIAN BUREAU OF STATISTICS (ABS) 2008. Agri- RHODES, J. 2002. ‘Cotton pollination by honey bees’. Austra- cultural Commodities: Small Area Data, Australia, 2005–06 lian Journal of Experimental Agriculture, 42: 513–518. (Reissue), ABS No 7125.0. ROBERTSON, L.N. & RHODES, J.W. 1992. ‘Honey bee COTTON AUSTRALIA (CA) 2009. 2008/2009 Annual Report (Apis mellifera L.) deaths near sprayed cotton and – Advancing Australian Cotton. Cotton Australia. observations on bee foraging behaviour in flowering INTERNATIONAL COTTON ADVISORY COMMITTEE cotton (Hymenoptera: Apidae)’. Australian Journal of (ICAC) 2007. The International Cotton Advisory Com- Entomology, 31: 243–246. mittee. The International Cotton Advisory Committee. SOMERVILLE, D. 2005. ‘Best practise in a honeybee pollina- KAZIEV, T. 1960. ‘The influence of bee pollination on cotton tion service’. Agnote. NSW Department of Primary quality’. Pchelovodstvo, 37: 33–35. Industry, Goulburn. MCGREGOR, S.E. 1976. Insect pollination of cultivated crop WARD, R.N. & WARD, K.E. 2001. ‘Impact of honeybee pol- plants. USDA, Tucson, Arizona. lination activities on Bt cotton production in northern Alabama’. In: Annual meeting of the Entomological MOFFETT, J.O., STITH, L.S., BURKHART, C.C. & SHIP- Society of America, South Eastern Branch, 2001, MAN, C.W. 1975. ‘Honey bee visits to cotton flowers’. Augusta, GA. Environmental Entomology, 4: 203–206.

Page 6 Cotton

Notes

This case study was prepared as part of Pollination Aware – The Real Value of Pollination in Australia, by RC Keogh, APW Robinson and IJ Mullins, which consolidates the available information on pollination in Australia at a number of differ- ent levels: commodity/industry; regional/state; and national. Pollination Aware and the accompanying case studies provide a base for more detailed decision making on the management of pollination across a broad range of commodities. The full report and 35 individual case studies are available at www.rirdc.gov.au.

Page 7 This project is part of the Pollination Program – a jointly RIRDC funds for the program are provided by the funded partnership with the Rural Industries Research Honeybee Research and Development Program, with and Development Corporation (RIRDC), Horticulture industry levies matched by funds provided by the Aus- Australia Limited (HAL) and Government tralian Government. Funding from HAL for the program Department of Agriculture, Fisheries and Forestry. The is from the apple and pear, almond, avocado, cherry, Pollination Program is managed by RIRDC and aims to vegetable and summerfruit levies and voluntary contri- secure the pollination of Australia’s horticultural and ag- butions from the dried prune and melon industries, with ricultural crops into the future on a sustainable and prof- matched funds from the . itable basis. Research and development in this program is conducted to raise awareness that will help protect pollination in Australia. RIRDC Publication No 10/119 ISBN 978-1-74254-089-4