Projected impacts of climate changes on agriculture
Climate change research
Horticulture in NSW is a very diverse and rapidly expanding industry. The range of industries that make up the horticulture sector include fruit, vegetables, nuts, turf, extractive crops (essential oils) and cut flowers. The production systems across each of these individual industries are extremely diverse, ranging from large-scale vegetable, viticulture and citrus operations in the Riverina, through ornamental horticulture and fresh fruit and vegetable production in the Sydney Basin, Hunter Valley vineyards, deciduous fruit on the Tablelands, to the North Coast macadamia nut and banana plantations.
The coping range of each of the horticultural crops is determined by the climate in which it has developed. Hazelnuts, for example, require 1200 hours of chilling at 5° to 7°C, and if they experience < -5°C at flowering, the crop will be damaged. The quality and weight of the macadamia harvest is ideal when daily maximum temperatures are between 30° and 35°C from December through to February; conditions outside this range will result in a loss of production. Similarly, citrus will suffer a production loss when temperatures over 37°C are experienced. There has been little work done on documenting the adaptation strategies required to extend the coping range of each of these industries in the face of climate change.
Determining the effect of temperatures outside those normally encountered, coupled with increased carbon dioxide, are important areas of future research, particularly in the key horticultural regions of NSW (Pittock 2003).
A study conducted by Webb (2006) on the impact of climate change on vineyards suggests that season duration in all grape growing areas will be compressed, with harvest usually occurring earlier. Webb (2006) also projected a negative impact on grape quality (using economic returns as a surrogate) if no adaptation strategies are implemented.
Projections for 2070 using the maximum warming scenario suggest that vineyards will become economically unviable in the Riverina. Photo: Michel Dignand
In the Riverina, a 16% decline in quality is projected for the minimum warming scenario, and a maximum decline in quality of 52% by 2030 for the maximum warming scenario. Projections for 2070, using the maximum warming scenario, suggest that vineyards will become economically unviable in the Riverina. Webb’s (2006) modelling also showed a shift towards the south and coastal areas of suitable conditions for many varieties. The adaptive capacity of the horticulture industries will be as varied as the industries themselves, and will depend on the investment cycle of the particular industry; for example, vineyards have a life of 30 years or more, posing a major challenge for adaptation (Pittock 2003).
Much of the beef, dairy and sheep industry in NSW comprises pasture-based production systems. While research has shown that a rise in carbon dioxide tends to promote pasture growth, this could be counteracted by reduced rainfall; a 10% reduction in average rainfall is predicted to counter the effect of a doubling of CO2 concentration in the atmosphere (Pittock 2003). If rainfall declines by more than 10%, the likely impact will be reduced pasture growth, which is not only important for animal production, but could also lead to potential environmental degradation of some grazing lands. In conjunction with the likelihood of reduced pasture growth, there is potential for increased variability of pasture production.
The nutritional quality of pastures is likely to decline, through a reduction in foliar nitrogen concentration due to elevated CO2, reflected in the impact on crude protein and water-soluble carbohydrates. However, the interaction between CO2 and the main drivers of plant growth (i.e. temperature, water and fertiliser) makes it difficult to determine the exact impact of climate change on nutritional quality.
Increased temperature and humidity will impact directly on the productive capacity of grazing animals, particulalry cattle. Photo: Alf Manciagli
Plants may differ in their ability to acclimatise to gradual increases in temperature, and the incidence of extreme temperatures outside the coping range may result in changes to the botanical composition of pastures. In the subtropics, C3 grasses (e.g. rye grass, used as winter forage for subtropical dairies) are vulnerable to both an increase in average and extreme temperatures in spring–early summer. Modelling has already indicated yield losses in spring (K Sinclair, pers. comm.). A shift in botanical composition towards C4 species (many of which have lower digestibility than C3 species) is likely, due to higher temperatures and the possible shift towards greater summer rainfall dominance. It is not, however, inevitable; a shift towards C3 species with increased CO2 may be equally likely, and has been suggested as an underlying mechanism of the worldwide encroachment of C3 ‘woody weeds’ in semi-arid rangelands.
In addition to these impacts on pasture yield and quality, increased temperature and humidity will impact directly on the productive capacity of grazing animals, particularly cattle. The temperature-humidity index (THI) is a measure of the heat stress on cattle, and hence a measure of their productive performance.
The impacts of increased heat stress in cattle include reduced grazing time (partly as a result of animals seeking shade), reduced feed intake, increased body temperature, increased respiration rate, and weight loss. In dairy cows, heat stress reduces milk yield, reduces milk fat and protein content, and decreases reproduction rates (Jones & Hennessy 2000). High-producing dairy cows are the most susceptible to increases in the THI. Heat stress days with THI > 80 lead to a substantial effect on reproduction of dairy cows, particularly for Holstein-Friesian. When assessing the impact of climate change on THI, it is important to assess, not just the change in the mean, but also the change in the number of extreme days (Howden et al. 1999b).
The response of beef cattle to THI is similar to the response of dairy cattle, although Bos taurus indicus cattle seem to be about 10% more tolerant than Bos taurus. All cattle require significantly more water when under stress. Significant stress is experienced at a THI of 80, and a recovery period is important in minimising production losses (Davison et al. 1996).
A cooling strategy, such as the provision of shade and sprinklers, is a key factor in minimising the impacts of increased THI for both dairy and beef cattle. The THI threshold at which a cow will generally start to be impacted by heat when no shade is provided is ~72. This can be increased to 76 by providing shade in feeding areas, and to 78 through the provision of shade and sprinklers (Jones & Hennessy 2000).
Implementing strategies to reduce heat stress is more practical for intensive livestock systems; however, shade infrastructure can be expensive for very large operations. Using sprinkler systems to reduce heat stress in dairy cows can also increase the risk of mastitis, because udders can become wet and dirty, creating ideal conditions for the growth of bacteria (Dairy Australia 2007).
For intensively fed (feedlot) beef cattle, heat stress is already a monitored health risk in mid to late summer. The risk is a function of the duration and intensity of heat load, together with the capacity for heat dissipation. A risk analysis program available to feedlot operators links with internet-supplied regional heat load index forecasts (www.katestone.com.au/mla). Climatic change leading to prolonged periods of sustained hot weather and greater peak temperatures can be expected to extend this risk period and increase feedlot operators’ reliance on such a service, which will allow them to invoke heat protection procedures (MLA 2007).
An additional risk from climate change to livestock industries, both intensive and extensive, is the potential for changing patterns of parasite risk to animals; for example, the potential for the ‘tick line’ (the cattle tick boundary) to move further south. Managing the impact of this increased parasitic risk to animals will require changes to operational practices, such as dipping and drenching. Adaptation to climate change is likely to require more flexibility and improved management of seasonal risk. An example of a risk management strategy for the extensive livestock industries is the maintenance of a higher proportion of ‘disposable’ animals in the flock. Adaptation to increased heat stress could involve cross-breeding.
Other intensive animal industries, such as poultry and pigs, are also vulnerable to increases in temperature and the resultant heat stress on animals. Structures may have to be redesigned to accommodate the conditions likely to be encountered in a changed climate. An alternative strategy is to relocate to a more favourable climatic region. Either option will be expensive, with the latter having flow-on effects to local communities, such as changed employment options.
Wheat is the predominant crop grown in NSW, with ~3.9 Mha of wheat for grain being planted in 2004, which was 55% of the total area prepared for crop production (ABS 2005). The area of wheat production peaked in 1969 at just over 4 Mha, when 5.8 Mt were harvested. The peak harvest, however, was in 2000, with 8.6 Mt harvested from just 3.4 Mha. These variations are a result of continued improvement in varieties, cropping practices and technology. However, there has been considerable year-to-year variation in the yields and the area grown. This variability highlights the uncertainty already being managed in this important cropping system.
Climate change will bring mixed results for wheat: increased yields for most regions, but a decrease in quality. Photo: Alf Manciagli
There have been only a few studies to date of the specific impacts of climate change on Australian wheat cropping systems. Most of these studies used the Agricultural Production System sIMulator (APSIM) wheat model, in conjunction with GCM outputs. The studies include an assessment of the conditional probability of not meeting the critical wheat yield threshold in South Australia (Luo et al. 2005a), the effects of a changing climate on wheat cropping systems in northern NSW (Power et al. 2004), the impact on grain protein levels of doubled CO2 in Qld (Reyenga et al. 2001), changes in wheat yields, grain quality, and gross economic margins at 10 sites across the Australian wheat belt (Howden et al. 1999a), as well as the increased likelihood of heat shock and the projected boundary changes of Australia’s viable wheat cropping areas (Reyenga et al. 2001). None of these studies considered the possible improvements that may be made through the adoption of adaptation measures.
The findings of these studies are diverse, with the impacts of temperature increases, rainfall changes and increases in CO2 concentration varying markedly across different regions (Howden & Jones 2001). The authors projected that Western Australia has a high likelihood of significant yield reductions; conversely, north-eastern Australia has a high likelihood of moderate yield increases, although there is also a small probability of substantial yield reductions in this region. The likely impact of climate change on wheat and sorghum production in central Queensland was studied by Potgieter et al. (2004). They concluded that large declines in yield (especially for wheat) were likely to occur by 2030, and recommended regional adaptation strategies (e.g. management practices, such as water conservation measures and adjusting planting dates and crop choice), as well as the development of more drought-resistant and more water-efficient crop cultivars, to mitigate the likely impact of climate change.
In analysing the impact of climate change on yields, it is also important to include a particular rate of improvement in variety adaptation. Breeding is a dynamic activity, and selections are going to be made as the climate changes are occurring. It is likely that some considerable compensation will occur, so that the selected varieties will be better adapted anyway (J Oliver, pers. comm.). Luo et al. (2003) projected that the conditional probability of not exceeding the critical yield at Roseworthy in South Australia increased from 27% for current climate conditions to 35%–50% under a mid-range climate change scenario.
Howden and Jones (2001) included three NSW sites in their assessment of the combined effects of possible atmospheric CO2 increases and the associated temperature increases and rainfall changes on the Australian wheat industry for the year 2070. The three sites spanned the southern (Wagga Wagga), central (Dubbo) and northern (Moree) inland regions. The results of this work indicate that the northern and central areas can generally expect beneficial impacts, but with a small risk of negative impacts. Combined, these two areas average 19% of the national yield. For the southern region of NSW, Howden and Jones (2001) predict a likelihood of largely beneficial impacts. Though it is likely that climate change conditions will favour increased wheat yields across NSW, climate change is likely to reduce wheat quality as a result of increased carbon dioxide concentrations. Elevated carbon dioxide reduces the protein content of wheat grain, which can reduce feed value, particularly when used as a protein supplement. However, the amino acid imbalance in wheat (low lysine) generally means it is purchased as an energy source for the starch. A reduction in protein due to starch dilution could therefore enhance the feed quality (J Oliver, pers comm.).
Free air carbon dioxide enrichment (FACE) experiments are currently being implemented in Australia, to better determine the effect of elevated CO2 on wheat growth (AGO 2007b). The outputs of climate change modelling exercises depend on the input variables; it is, therefore, unsurprising that different scenarios are predicted. However, there does not appear to be any dispute over the prediction that much of NSW will experience hotter and drier conditions in the future. Therefore, a wider range of crop species and planting windows is likely to be needed to help spread the risk.
Sorghum is the most widely grown summer crop in NSW, accounting for 63% of the average area sown during the period 1992/93–2001/02 (F Scott, pers. comm.). Cotton, maize, mungbean, sunflower and soybean make up the other major dryland summer crop options. Although sorghum has many advantages as a summer crop (it is more drought-tolerant than maize), it prefers a soil temperature of at least 17°C during sowing, which means that sorghum is not normally sown in northern NSW before early October. In contrast to sorghum, maize can be planted when soil temperatures reach 12–14°C, so maize can be planted 4–6 weeks earlier than sorghum. In northern NSW, maize can be sown as early as late August or early September.
Sunflower is an alternative summer crop that can be sown at similar soil temperatures to maize, and so has a similarly wide planting window. However, risk of bird damage and uncertain prices make it less attractive.
Apart from using tactical sowing opportunities to take advantage of prevailing soil moisture options, summer crops can be sown early or late to avoid the stress of flowering and grain fill during peak summer temperatures. The wide range in the maturation rate of maize varieties allows growers to take advantage of early or late sowing opportunities under dryland conditions. Quicker maturing varieties can:
- take advantage of a full profile of soil water, minimising the risk of running out of water before the crop matures in low rainfall seasons
- beat the summer heat before tasselling, for spring plantings
- avoid frost damage to late plantings.
Dryland cotton has been produced in Australia in seasons in which the starting soil moisture is reasonably high and the outlook for rainfall is promising. Dryland cotton production has been extremely limited for the past few years, as these conditions have not been met. Reductions in rainfall due to climate change are likely to limit future dryland cotton production. Dryland cotton is grown in regions that experience moderate to high rainfall variability during the key January to March period (Ford & Forrester 2002), and increases in rainfall variability and extreme events due to climate change (CSIRO 2006) will further affect this production.
Climate change may exacerbate the impacts of weeds, pests and diseases, through increased prevalence and changes in geographic distribution. There is potential for increased rust incidence in crops and pasture species; however, a drier climate may reduce the impact of cereal diseases.
Though only ~1.5% of agricultural land is irrigated in NSW annually, it accounts for an average of ~30% of the total agricultural production value (NSW Irrigators 2002). Nationally, irrigated farm profit contributes over 50% of total agricultural profit (CRC IF 2006).
In NSW, the majority of irrigation is carried out in irrigation schemes. These operate as companies, under licences issued by the NSW Office of Water. The major crops are cotton and rice.
Climate change may have some benefits for the irrigated rice industry. Photo: Alf Manciagli
Generally, irrigated cotton and rice together contribute over $1 billion to the NSW economy (CRC IF 2006). However, between 2000/01 and 2004/05 there were significant reductions in the value of irrigated cotton (from $930 million to ~$500 million) and rice (from $350 million to $178 million), as a result of reduced water availability.
The main genetic limitation to rice-growing is the cold sensitivity of the temperate varieties used in Australia. Deep watering is used as a strategy to ameliorate cool temperatures at the sensitive times around panicle initiation. Increased temperature under climate change will mitigate cold sensitivity, so the need for deep watering should be dramatically reduced. Hence, climate change may have some benefits for the irrigated rice industry. It may also permit alternative rice species to be considered.
Climate change is likely to have a number of key impacts on cotton production in NSW. Cotton is suited to warm climates; it is affected by temperatures below 12°C (cold shock) and above 36°C. Decreases in the number of cold days due to climate change, as predicted by the CSIRO (2006), may prove beneficial to cotton production; however, the associated increase in days above 35°C may be detrimental. Fibre quality of both irrigated and dryland cotton is significantly affected by both temperature and water availability. These impacts vary with the time of season and interactions with other variables. Because international markets are increasingly focused on optimal fibre quality, managing this important characteristic is likely to become a greater challenge to Australian growers as the climate changes.
Most of the Australian cotton crop and the entire rice crop are irrigated; however, production in recent years has been significantly diminished, due to greatly reduced water availability. Most cotton growing regions in NSW started the 2006/07 season with announced allocations of around 20% or less, and some regions had 0%. Reduced water supply reliability limits the options for alternative cropping systems, some of which may require large capital investments. Compared with alternative crops, cotton has one of the highest returns per ML of water used; therefore, it is unlikely that growers will move away from irrigated cotton production in the short term.
Both rice and cotton are irrigated predominantly by gravity-fed surface irrigation systems. These systems are less energy-intensive than pressurised systems, and have sufficient flexibility to allow production to be reduced in years of lower water availability.
The majority of vegetables grown in NSW are irrigated (83%), as are 75% of fruit and 83% of all grapes grown in NSW. The irrigated sector represents an important source of fresh food production, especially in times of drought, and also provides some stability for regional livelihoods.
In recent years, there has been a move to introduce more efficient irrigation systems for most of the irrigated commodities in NSW. When irrigation becomes more efficient, through either technology or improved practice, salts are concentrated in the rootzone. This can have a negative impact on production, and therefore needs to be managed. Related to this is the increasing reuse of water, which grew from 1% of all water supplied (134 424 ML) in 1996/97 to 4% (516 563 ML) in 2000/01 (CRC IF 2006). There is generally a higher salt load in reuse water and, when coupled with more efficient irrigation, the understanding of the movement of salt and nutrients in the rootzone and beyond becomes even more critical. This is a high priority research area for the CRC for Irrigation Futures.
Before use, most irrigation water is stored in dams, ranging from very large, government-managed storages (24 814 GL total capacity in NSW) to small, on-farm dams of only a few megalitres. In the current climate, up to 50% of stored water can be lost by evaporation before use. These losses are projected to increase over much of NSW, due to climate change; therefore, mitigating these losses is another priority area for research. The CRC for Irrigation Futures, in conjunction with the CRC for Polymers, is researching new polymer formulations, in an effort to improve their ability to mitigate evaporation.
Climate change is likely to increase pressure for irrigation to become more efficient. To be able to effectively meet this demand, there is an increasing need for improved technology that will enable systems to be managed and operated in a more responsive mode. The areas of focus range from the plant (plant-based sensors of water stress) to catchment-wide responses (remote sensing of crop evapotranspiration).