Background and aimsBananas are demanding in nitrogen (N) input; therefore, there is a temptation for organic farmers for using synthetic N fertilisers, which are not allowed under organic standards. The aim of our study was to develop a tool that identifies high banana yields obtained with suspiciously low organic N input.MethodsWe systematically reviewed literature from experimental studies on N fertilisation in bananas from all over the world. We also developed a simplified N balance model for organic bananas. Furthermore, N fertilisation and banana yield data from organic and conventional farmers in different countries were collected. From these, a subset of trustworthy organic farms was identified, as a reference concerning plausible ratios of yield versus fertilisation. A model was developed to estimate the deviation from the regression of trustworthy farms and thus identify suspicious cases.ResultsNeither literature nor the N balance led to a meaningful benchmark for differentiating plausible from non-plausible yields. The regression of yield on N fertiliser rate from the trustworthy organic farmers, however, turned out to be a helpful reference, and the deviation from this regression helps to achieve our aim. Depending on the alert limit, that is, the probability of obtaining false positive results, 4, 6, or 9 out of 157 data-pairs from organic farmers turned out to be suspicious.ConclusionMeasuring deviation from the regression of the trustworthy farms is a useful tool for identifying organic banana farmers suspected to be using synthetic N fertilisers but is not in itself a proof of fraud. The model will improve as more data becomes available.

Mango (Mangifera indica L.), a leading tropical fruit crop, is a prime candidate for intensification through modern orchard-management techniques, including canopy manipulation to improve light interception. This study investigated how leaf-level acclimation to light gradients within the canopy of a high-yield, dwarfing mango cultivar (Calypso™) could be used to examine integrated canopy-scale responses. We quantified foliar morphological, biochemical and physiological traits across a range of canopy positions using this information to model canopy-scale productivity within digital-twin representations of mango under both conventional (i.e., open-vase) and espalier-trellis training canopy systems. Key findings demonstrated that leaves exposed to higher light exhibited increased leaf mass per unit area, nitrogen content and photosynthetic capacity (Asat), but decreased chlorophyll-to-nitrogen ratios and photochemical reflectance indices, reflecting trade-offs between light capture and photoprotection. Phenolic content increased under high irradiance, indicating investment in photoprotective compounds at the expense of net carbon gain. Modelled leaf-level productivity increased with light availability, following a Michaelis-Menten saturating response, with diminishing returns under high light. Digital modelling of canopy light interception revealed that espalier-trellis training enhanced light distribution efficiency per unit leaf area but resulted in a 6.5% reduction in total canopy productivity due to a smaller total canopy leaf area. However, when normalized by total canopy leaf area, the espalier-trellis system showed a 3.6% productivity advantage over conventional canopies at the time of year modelled. These results highlight the role of canopy structure and light-use efficiency in determining orchard productivity. Integrating spatially explicit mechanistic models with LiDAR-derived canopy data offers a promising pathway for designing high-density, resource-efficient mango orchards. Future work should expand modelling to account for dynamic canopy shape throughout the growing season and evaluate the interaction of modified canopy structures with environmental stressors, particularly under climate variability.

AbstractMolecular detection of environmental DNA (eDNA) and RNA (eRNA) allows highly sensitive qualitative (i.e., presence or absence) and quantitative (i.e., abundance) monitoring of aquatic bacteria. However, bacterial molecular diagnostics are limited by low positive predictive values. Protocols for bacterial eDNA and eRNA molecular monitoring have primarily focused on optimizing specimen collection, and the optimal method to purify bacterial nucleic material from postcollection aquatic specimens to maximize the analytical sensitivity of molecular diagnostics remains poorly defined. Accordingly, strategies to isolate bacterial eDNA and eRNA from fresh and saltwater were investigated. We evaluated two filtration and four nucleic acid purification systems as representative of current generation bacterial eDNA and eRNA isolation strategies for capacity to isolate bacterial eDNA and eRNA from prelysed (i.e., free‐nucleic acids) and viable (i.e., colony forming units, CFU) bacterial cells. We also compared the sensitivities of reverse transcription quantitative PCR (RT‐qPCR) and metagenomic shotgun microbiome sequencing. The optimal protocol used 0.7 μm borosilicate glass filters (Whatman plc) followed by extraction with the RNeasy PowerWater kit (Qiagen). The protocol had a very high analytical sensitivity (10−3–100 ng and 102–101 CFU detected in 500 mL) across multiple species of bacteria, when tested with either RTqPCR or metagenomic sequencing. Importantly, this study highlighted several limitations which are restrictive to both qualitative and quantitative bacterial eDNA and eRNA studies. First, a 12‐h time course between sampling and extraction revealed significant species‐specific changes in cell number and free‐nucleic acid concentrations can occur postspecimen collection. Second, we found Gram‐positive bacteria yielded less nucleic material compared to Gram‐negative bacteria suggesting bacterial eDNA and eRNA studies could be biased by microorganism genome stability and extraction efficiency. This study highlights the need to define the species‐specific diagnostic sensitivity of a protocol when monitoring aquatic bacterial eDNA and eRNA with molecular diagnostics.

Reproductive development of fruiting trees, including mango (Mangifera indica L.), is limited by non-structural carbohydrates. Competition for sugars increases with cropping, and consequently, vegetative growth and replenishment of starch reserves may reduce with high yields, resulting in interannual production variability. While the effect of crop load on photosynthesis and the distribution of starch within the mango tree has been studied, the contribution of starch and sugars to different phases of reproductive development requires attention. This review focuses on mango and examines the roles of non-structural carbohydrates in fruiting trees to clarify the repercussions of crop load on reproductive development. Starch buffers the plant’s carbon availability to regulate supply with demand, while sugars provide a direct resource for carbon translocation. Sugar signalling and interactions with phytohormones play a crucial role in flowering, fruit set, growth, ripening and retention, as well as regulating starch, sugar and secondary metabolites in fruit. The balance between the leaf and fruit biomass affects the availability and contributions of starch and sugars to fruiting. Crop load impacts photosynthesis and interactions between sources and sinks. As a result, the onset and rate of reproductive processes are affected, with repercussions for fruit size, composition, and the inter-annual bearing pattern.

Context Banana production in Australia is in three primary sub-regions within tropical North Queensland and the industry faces a variety of challenges including costs of production, disease and pests, and environmental impacts. The range of soil characteristics and banana leaf nutrient status on banana farms has not previously been systematically described. This knowledge gap makes it difficult to adapt research, management recommendations, and regulations to the needs of the three primary growing sub-regions. Aims In this work, we aimed to identify key soil factors that differentiate growing sub-regions, and provide context for future research and industry regulation. Methods We characterised soil and banana leaf samples from 28 banana farms on soil types accounting for >85% of Australia’s banana production. Key results and conclusions Variation in soil properties and leaf nutrient concentrations were driven largely by site- (principal component 1 in both cases) and management-related variables (principal component 2 in both cases). Management-related foliar nutrient concentrations did not differ between regions despite differences in the associated soil variables. The most important site characteristics appeared to be soil parent material and climate. The Mareeba sub-region has basaltic soils, low rainfall and temperature, whereas the other two sub-regions are hotter, wetter and have a variety of soil parent materials. Leaf nitrogen concentrations were mostly below the regulated limit for additional nitrogen fertiliser application. Implications Our findings can facilitate sub-region-specific site selection for research, extension, and monitoring and more targeted regulation of banana production- and environment-related issues.

Nitrogen (N) fertilizers are routinely applied to bananas (Musa spp.) to increase production but may exacerbate plant diseases like Fusarium wilt of banana (FWB), which is the most economically important disease. Here, we characterized the effects of N rate and form on banana plant growth, root proteome, bacterial and fungal diversity in the rhizosphere, the concentration of Fusarium oxysporum f.sp. cubense (Foc) in the soil, and the FWB severity. Banana plants (Musa subgroup ABB) were grown under greenhouse conditions in soil with ammonium or nitrate supplemented at five N rates, and with or without inoculation with Foc. The growth of non-inoculated plants was positively correlated with the N rate. In bananas inoculated with Foc, disease severity increased with the N rate, resulting in the Foc-inoculated plant growth being greatest at intermediate N rates. The abundance of Foc in the soil was weakly related to the treatment conditions and was a poor predictor of disease severity. Fungal diversity was consistently affected by Foc inoculation, while bacterial diversity was associated with changes in soil pH resulting from N addition, in particular ammonium. N rate altered the expression of host metabolic pathways associated with carbon fixation, energy usage, amino acid metabolism, and importantly stress response signaling, irrespective of inoculation or N form. Furthermore, in diseased plants, Pathogenesis-related protein 1, a key endpoint for biotic stress response and the salicylic acid defense response to biotrophic pathogens, was negatively correlated with the rate of ammonium fertilizer but not nitrate. As expected, inoculation with Foc altered the expression of a wide range of processes in the banana plant including those of defense and growth. In summary, our results indicate that the severity of FWB was negatively associated with host defenses, which was influenced by N application (particularly ammonium), and shifts in microbial communities associated with ammonium-induced acidification.

Fusarium wilt of banana, caused by Fusarium oxysporum f.sp. cubense, is a serious threat to banana production, with no effective treatment. Mulches and oils from Eucalyptus have been shown to inhibit the growth of F. oxysporum, and reduce the severity of Fusarium wilt. Eucalyptus cloeziana leaves contain many compounds commonly found in Eucalyptus as well as a very high concentration of the uncommon compound tasmanone. We applied E. cloeziana mulch, fresh or partially steam distilled, or no mulch as a control, to potted soil, inoculated the soil with F. oxysporum f.sp. cubense and grew susceptible banana plants to test the capacity of the mulch to reduce the severity of Fusarium wilt of banana. Disease severity, analysed by internal corm discoloration, was reduced from 90% in the control to 37% in the treatment with steam distilled mulch.In another experiment, comparison of steam distilled with fresh mulch showed no difference between the two, with a disease severity of 13%, compared to 49% in the control. Mulch (fresh or steam distilled) had no detrimental effects on plants grown without F. oxysporum inoculation. We propose that disease suppression was likely due to tasmanone, as it was the primary oil constituent of the mulches.

Previous research has shown that application of iron chelates to soil reduces Fusarium wilt in several crop species. The aim of this work was to test the effect for bananas grown in tropical soils. Disease severity and plant characteristics were measured in banana plants (cv. Ducasse, Musa ABB) grown in pots inoculated with Fusarium oxysporum f. sp. cubense , Race 1 in two experiments. Experiment 1 compared amendment with iron chelates (with ligands of differing iron binding stability) with water, plus an uninoculated unamended control, in two tropical Australian soils. Experiment 2 examined the effect of Fe-HBED application rate with high or low calcium addition. In Experiment 1 , iron-saturated chelate application did not significantly affect disease severity or plant tissue iron concentration, irrespective of the iron binding stability of the chelate. In Experiment 2 , disease severity was not affected by Fe-HBED or calcium application rate. The concentrations of iron and aluminium in plants were both somewhat affected by the addition of Fe-HBED, calcium and their interaction. Fusarium wilt of bananas was not affected by addition of iron chelates to these soils. The lack of effect was likely due to high iron availability in the soils overwhelming the capacity of the treatments to alter iron availability to the host plant and pathogen. Application of strong chelating ligands increased the concentration of aluminium and decreased the concentration of manganese in plant tissue, with possible detrimental effects.

The sorption behaviour of three perfluoroalkyl substances (PFASs), namely perfluorooctane sulfonic acid (PFOS), perfluorooctanoic acid (PFOA) and perfluorohexane sulfonic acid (PFHxS), was determined on 28 tropical soils. Tropical soils are often highly weathered, richer in sesquioxides than temperate soils and may contain variable charge minerals. There are little data on sorption of PFASs in tropical soils. The highest Kd values were found for PFOS with mean values ranging from 0 to 31.6 L/kg. The Kd values for PFOA and PFHxS ranged from 0 to 4.9 L/kg and from 0 to 5.6 L/kg, respectively. While these values are in the range of literature sorption data, the average Kd values for PFOS and PFOA from the literature were 3.7 times and 3.6 times higher, respectively, than those measured in this study. Stepwise regression analysis did explain some of the variance, but with different explanatory variables for the different PFASs. The main soil properties explaining sorption for PFOS and PFOA were oxalate-extractable Al and pH, and for PFHxS was pH.

Naturally occurring and synthetic chelating ligands can act as suppressants for fungal pathogens, nematodes and weeds, based on their ability to alter micronutrient bioavailability in soil, particularly iron. Chelators are also used as detergents, for remediation of heavy metal contamination and for supplying metals as fertiliser. The aim of this work was to test the ability of chelators to solubilise metals, in particular iron, in tropical soils over an environmentally relevant pH range. Six topsoils from farms in North Queensland, Australia were adjusted to pH 5, 6 and 7 and then extracted with CaCl2, EDTA, DTPA, EDDHA and mimosine. The extracts were analysed for concentrations of aluminium, copper, iron, magnesium, manganese, potassium, strontium and zinc. EDDHA solubilised iron effectively under all of the conditions tested, indicating its likely suitability for pest suppression. The concentration of aluminium in EDDHA extracts was positively correlated with pH, and at pH 7 the concentration of aluminium was far greater than that of iron. An increase in the mobility of aluminium from EDDHA application to soil may lead to aluminium toxicity in plants, which should be considered further in any practical application of EDDHA. Mimosine, which is also a strong chelator, was a poor extractor of all metals, possibly due to adsorption to the soil.

Soils are known to differ in suppressiveness to soil-borne diseases, but the suppressiveness or otherwise to Fusarium wilt of Australian soils used to grow bananas is unknown. In this work we tested the relative suppressiveness of six key soil types. Banana (Musa (AAB group) ‘Pome’, cultivar ‘Lady Finger’) was grown in pots of the soils inoculated or not with Fusarium oxysporum f.sp. cubense (Foc) ‘Race 1’. Sixteen weeks after inoculation the plants were harvested and disease severity was assessed by measuring discoloration within the rhizome. In the inoculated pots, disease severity was greatest in the alluvial Liverpool and Virgil soils and least in the basaltic origin Tolga soil. No disease was detected in the non-inoculated pots. Soils with the lowest disease severity had the highest root mass, irrespective of inoculation, and the largest (negative) effect of inoculation on root dry mass. Disease severity in inoculated pots was negatively correlated with soil clay content and β-glucosidase activity. The results indicate that the risk of Fusarium wilt negatively impacting banana growth differs between soils of the main Australian banana-growing region.

The severity of Fusarium wilt of bananas has long been classified based on visual assessment of necrosis in rhizome or pseudostem cross-sections. The improved method proposed here uses digital image analysis to quantify the proportion of rhizome tissue that is necrotic. It agrees well with visual classification, but provides greater reproducibility, precision and statistical power.

Fusarium wilt, a plant disease caused by pathogenic strains of the soil fungus Fusarium oxysporum, affects over 100 crops including banana, tomato, ginger, cucumber, oil palm, watermelon, wheat, and flax. Fusarium wilt of bananas, also known as Panama disease, is rapidly spreading around the globe, threatening banana and plantain production.

Plant-associated microorganisms are known to strongly influence host fitness but are poorly characterised for bananas and the soils used for their production. We have characterised the microbiomes of: 1) 17 soil types representing c. 70% of the main banana production area of Australia; 2) 55 banana genotypes; and 3) multiple plant compartments using high throughput phylogenetic marker gene sequencing. In addition, we have ranked the suppressiveness of multiple soil types to Fusarium wilt (Fusarium oxysporum f. sp. cubense) and used a novel bait trapping technique to identify microorganisms that are attracted to compounds released by the pathogen. Consequently, we have identified a range of microorganisms that are associated with susceptibility and resistance to the disease. These findings will assist other projects that aim to identify on-farm management practices that encourage the establishment of Fusarium wilt suppressive banana plant and soil microbiomes under field conditions.

Panama disease, caused by the fungal pathogen Fusarium oxysporum f.sp. cubense (Foc), has had devastating effects on banana production globally and is now threatening Australia's largest banana production region, Far North Queensland (O'Neill et al., 2016). Severity of the disease, and growth of the causal organism have been linked to soil properties in other parts of the world, but not in this key agricultural region of Australia (Deltour et al., 2017; Peng et al., 1999). Disease severity is determined by the host plant, the pathogen and the environment. For soil-borne diseases such as Panama disease the soil microbial community is a particularly important aspect of the environment. The host plant, pathogen and soil microbial community are all influenced by abiotic characteristics of the soil. The aim of this study was to determine a) if Panama disease severity differs between soils of Australia's main banana growing area, b) which abiotic soil characteristics are influential, and c) if a rapid assay can be used to rank soils for their effect on disease severity.

Production of many crops, including bananas, is threatened worldwide by the spread of pathogenic strains of Fusarium oxysporum, the causal agent of Fusarium wilt. Severity of the disease is related to soil biotic and abiotic attributes, which influence the plant, the pathogen and the other soil organisms. Across a variety of crops, soil temperature, redox potential, and extractable iron and manganese contents are generally positively correlated with disease severity, whereas pH, nitrate:ammonium ratio, organic matter content and extractable calcium, zinc, silicon, potassium, phosphorus and boron contents are negatively correlated, but less consensus exists for bananas. There are numerous incompletely understood interactions between soil abiotic attributes and disease severity, including those between pH- and redox-controlled micronutrient availability, buffering by organic matter and clay, and effects of nutrients on plant defence mechanisms. Though not all soil attributes can be managed, pH, organic matter content and availability of nutrients show promise for manipulation to reduce disease severity and mitigate risk.

Quantification of soil organic carbon (SOC) content is important for sustainable agricultural management and accurate carbon accounting. Infrared (IR) absorbance can be used to estimate SOC content, but the relationship differs between regions due to matrix effects. We developed an IR-based model specific for SOC in Papua New Guinean soils. A total of 437 samples from 0.0–0.3m depth were analysed for SOC using Dumas combustion. IR absorption spectra were collected from the same samples, and a predictive regression model was developed using the 6000–1030cm–1 spectral range. Using a validation set, predicted SOC values resulting from the IR-based model compared well with values from Dumas combustion (R2=0.905; ratio of performance-to-deviation=5.64). Constraining wavelengths to positively correlated regions of the spectra was also explored and showed improved model performance (R2=0.932). Overall, IR analysis provides a robust method for estimating SOC content for a range of Papua New Guinean soils.