1. Introduction
Grain sorghum (
Sorghum bicolor (L.) Moench) is the main summer grain crop in the northern grains region and is a significant part of the dryland cropping system of north-eastern Australia, which supplies feed grains to the animal industries [
1]. About 60% of the Australian sorghum crop is grown in Queensland, and the rest is grown in northern NSW. It is primarily a summer crop, with an extended season at higher latitudes including Central Queensland and farther north, and can be a good rotation crop, tolerating moisture and heat stress, as well as performing better than maize on soils with minimal potassium (K) levels [
1].
The most common and the most important root and stalk rot pathogen of sorghum is
Macrophomina phaseolina (
Mp), a soilborne fungus causing charcoal rot which, in most cases, can lead to lodging. It can infect via the roots of sorghum plants at almost any stage of plant growth but develops more rapidly after post-flowering stress [
2]. It is also a late-season disease that causes yield loss through poor grain fill, but more commonly through plant lodging, which impedes harvest and reduces grain quality. Unfavourable environmental conditions are key stressors known to promote disease development. The fungus is extensively distributed throughout Australia, infecting the root and stems of over 500 weed and crop species including winter cereals [
3]. Its microsclerotia can survive in the soil and on stubble for more than four years, and it is unknown what soil conditions are necessary to reduce the survival of microsclerotia in Australia, but overseas studies show that wet soil can significantly impede their survival [
4].
Numerous reports on the destruction of sorghum crops by charcoal rot are available; however, sound and reliable quantitative data on yield losses are not provided [
5]. A 35% reduction in the 1000 grain weight based on the combined effects of drought and charcoal rot that caused plants to lodge has been reported, with the loss in grain yield being due to a reduction in the size of the grain. Anahosur and Patil [
6] reported losses in the weight of sorghum seeds ranging from 15.18 to 54.59% in different genotypes depending on the lodging levels (21.99–78.37%), therefore concluding that there was a direct correlation between lodging and a loss in seed weight. Overseas yield losses due to charcoal rot have been estimated at more than 50% [
6]. Despite the lack of any formal quantification in Australia, significant yield losses have been associated with lodging, where prevailing hot and dry conditions have resulted in the widespread high incidence of charcoal rot and subsequent lodging [
4]. Lodged plants do not produce full heads of grain or fully formed heads and are difficult to harvest [
7]. The realised losses associated with lodging were varied and dependent on the capability of individual growers to recoup lodged heads with the harvesting equipment available. Yield loss quantification through in-field assessments is challenging, and varies depending on several factors including the weather, soil, time of infection, cultivar susceptibility, and the degree of lodging.
The early detection and quantification of stalk rot pathogens in sorghum paddocks is paramount to the development and implementation of disease management strategies. PREDICTA
®B, a quantitative polymerase chain reaction (qPCR) DNA test [
3,
8], developed by the Primary Industries and Regions, South Australia (PIRSA), through the South Australian Research Development Institute (SARDI), helps grain and pulse producers to identify which soilborne pathogens pose a significant risk to their crops before seeding, so that steps can be taken to minimize the threat of yield loss [
9]. PREDICTA
® B has been found to have the potential to provide an accurate and dependable system to estimate charcoal rot severity in sorghum, and it could be a useful tool to assess inoculum levels before sowing sorghum [
10]. PREDICTA
®B can also help pathologists to understand the relations between the pathogens within a paddock and to advise growers on management options to limit the impact of these disease complexes.
Macrophomina phaseolina is often detected using PREDICTA
®B in Australia’s northern region. These results are reported with categories based on population density so that growers and advisers can benchmark the levels of pathogen DNA detected in paddocks against the rest of the industry [
11]. Likewise, when the relationship between the initial pathogen level and disease has been defined, the level detected in the sample is reported with a disease risk rating. However, test results for the charcoal rot pathogen are still reported only as relative population densities rather than as a disease risk, since the level of yield loss associated with the pathogen DNA level is yet to be determined. Gray et al. [
12] have shown that the incidence of charcoal rot in sorghum was positively correlated with soil populations of
Mp. Still, the more recent investigations on this test, with only two seasons of data, were not sufficient to determine any correlation between the population density and disease severity, lodging, or effects on yield [
13]; therefore, further work is required in this area.
This study tried to determine if Mp inoculum loadings found in soil and stubble based on PREDICTA®B can be correlated with end-of-season disease levels, lodging, and associated yield loss. Accordingly, this can be developed into disease risk categories, to provide growers with a risk assessment for the likelihood of potential disease before planting, and to help inform crop and variety decisions and guide management to minimize losses.
4. Discussion
It is acknowledged that drought stress alone can cause lodging without aid from pathogens where an inoculum is absent [
23]. However, where pathogens are present, drought-stressed plants are invariably invaded by them, and this leads to increased damage to plants. Low or intermediate levels of drought stress may be tolerated by the plant except when combined with the pathogen. While drought alone must have contributed to some yield reduction, the compounded effects of charcoal rot and drought causing plants to lodge must have elevated the degree of yield loss [
5].
For this study, the basic hypothesis is that Mp causes symptoms in the plants and that those symptoms result in crop loss. It has been shown that the logarithm of the amount of disease present is a useful indicator of the potential harm caused. In some cases, there were zero diseases recorded, and in those cases, it is not possible to take a logarithm of the data. That problem is typically avoided by using the transformation y = log10(x + k), where k is taken as 1.0 and x is the untransformed value. The yield potential was estimated as Measured yield × 100/(100 − % lodging).
The analyses presented here show that the cause of the lodging of sorghum depends on numerous factors including rainfall, region, and variety, as well as the number of pathogens in the soil. No doubt there are other factors, including soil properties, cultural practices, and other meteorological factors that were not assessed in this report. The data analysed included the amount of Mp in the soil. No direct method of assessing the sampling variation was available for those data, but it was inferred from the correlation between the sampling times that at least half of the information from one sampling was contained in the second sampling. That variation would have included the sampling variation from both times and the variation in the actual changes between the two times. From that, it was concluded that the sampling method used was satisfactory.
The current study only analysed site variation, so no data on the amount of variation within a paddock were provided. Such a study would be of aid in designing future sampling schemes. Some exact duplicates would be useful in assessing field sampling variation.
The results showed that there is a strong relationship between the incidence of charcoal rot and lodging. It was also noted that there is a quadratic relationship between rainfall and lodging, but with an adjusted R
2 of 0.326 compared to the value of 0.899 based on the incidence of charcoal rot. This implies that charcoal rot is a causative factor in lodging. Furthermore, the fitted relationship showed that there is minimal lodging when the incidence of charcoal rot is less than 20%. There seems to be a relationship between the pathogen inoculum density in the soil and disease intensity, and between disease intensity and yield loss [
6,
24,
25]. It has been proven previously that there is a definite relation between the sink source and the charcoal rot severity, where plants become susceptible to diseases at the post-flowering to grain filling stage, when the food reserves are translocated from the stem to the ears and the food supply to the roots is reduced [
25]. Thus, some agricultural practices have intended to reduce the inoculum density, such as the use of bio solarization [
26], a high soil moisture above 60% [
27,
28], and rotation with non-host crops [
29,
30]. Moreover, due to
Mp having heterokaryotic mycelium, there are varied asexual sub-phases, along with phenotypic, geographical, and genetic variations, making it a challenging pathogen for designing effective and long-lasting disease management [
31]. Therefore, charcoal rot is frequently controlled by applying cultural measures that reduce plant stress. Utilizing resistant cultivars and adhering to crop- and time-specific cultural practices that preserve soil moisture are two ways to manage charcoal rot.
As set in the PREDICTA
®B disease risk thresholds, a low or non-detectable level of inoculum means that the risk of disease is low, while at high levels of inoculum, the occurrence and severity of disease depend on the susceptibility of the variety and the conduciveness of the environment [
20]. There is evidence from this study that at an
Mp load of ≥2.5 k copies per gram of soil on a log
10 scale could result in a high incidence of charcoal rot, and its associated risk of serious lodging can be used as a management decision tool. However, that cut-off is set using only the results from twenty-nine sites, and further data across other sites and seasons are needed to refine that figure.
There is no evidence of crop loss associated with the amount of Mp present. The relationship between disease and crop loss may be seasonally dependent. There may be some compensation for healthy plants using the available water more rapidly and thus restricting later plant growth. Further information concerning the effect at different growth stages may be obtained in future studies by including data on the 1000 grain weight, panicle size, and the number of plants (panicles) per square meter.
There is a strong suggestion of differences in the susceptibility of varieties to charcoal rot. For example, as shown in
Figure 13, the single sample of Taurus was very resistant. The varietal effects may be particularly important and further data should be sought to establish the relative vulnerability to disease. Conversely, varietal effects may be masking other effects in this study and these should be considered in future studies.
The two sampling times provided some interesting information, especially on the sampling precision. The similarity between the two data sets suggests that sampling before sowing would also give useful information that could be used to aid in management decision making. The current data do not give useful information on the persistence of Mp in soil, or when soil sampling can be undertaken to aid with future management decisions. It may be possible to use DNA samples, taken months before sowing, to aid with management practices.
The disease index of Das et al. [
30] combines the incidence (which is a ratio) with the amount of disease present (measured as the length of lesions in mm). The mixture of units could be removed by considering the fraction of a given length of stem that was affected by lesions. The current study found that disease incidence is a better predictor of damage than the disease index of [
30], and that it avoids the problem of a mix of units. If future studies confirm that incidence alone is an effective predictor, it could make disease assessment simpler.
To our knowledge, this is the first paper on assessing PREDICTA®B as a potential disease risk management decision tool for sorghum charcoal rot. From the obtained results, it should be noted that high levels of inoculum do not mean that disease will occur, but that there is a high risk if the conditions are favourable for disease development. If available, disease risk categories should be used as a general guide only. Other factors such as the climate, management practices, soil type, crop type, variety, and seasonal conditions should be considered in interpreting PREDICTA® results and assessing disease risk. Moreover, the repeated use of PREDICTA® tests within a cropping system for patterns observed in pathogen levels and disease occurrence can be used to refine interpretation.
While this study has provided much useful information, it has highlighted the need for further data collection. Further data should be collected to confirm the relationship between the incidence of charcoal rot and the amount of lodging. This should include data across the northern cropping region, hopefully with rainfall levels above the bottom quartile. A study should be undertaken to collaborate the suggested log of 2.5 copies per gram as an indicator of the high incidence of charcoal rot and the associated risk of serious lodging to use this as a management tool. Such a study should be designed so that it isolates background effects and is sufficiently precise to give reliable estimates of the effects. The persistence of Mp in soil should be established by taking a time series of samples over a year. Those data could then be used in the development of a sampling scheme for a predictive model. Other agronomic measures including the 1000 grain weight, panicle size, and plant density should be collected to enable the growth stages to be quantified. Basic field sampling variation across a large paddock should be quantified—this would include a spatial part within a paddock, as well as sampling variation from adjacent points, to prove the sampling precision of various strategies. Information should be obtained on the susceptibility of varieties to charcoal rot. Such a study should include information on the disease load in the soil and be replicated across sorghum growing regions over several years. Where possible, studies should be limited to a few varieties to simplify the experimentation.