Context. Tea leaf blight (TLB) stands as one of the most destructive diseases affecting tea plants, posing a significant threat to both the yield and quality of tea crops.

Aims. Our aim is to employ efficient deep learning techniques to achieve precise remote sensing monitoring of TLB in natural environments.

Methods. We present an innovative methodology that leverages the combined power of ECDet and MobileNetv3 for the detection and severity assessment of TLB from unmanned aerial vehicle (UAV) remote sensing images. ECDet is constructed with a lightweight backbone to reduce the complexity of the model, and a MicroEA-FPN feature pyramid structure and a decoupled spatial attention-weighted head to achieve balance between focusing on the detailed information of tea leaves and extracting semantic information from small targets. In addition, transfer learning has been implemented to address the performance degradation owing to low UAV image resolution, and the MobileNetv3 is used to improve the accuracy of severity assessment.

Key results. The accuracy of our method was 78.46% in detecting TLB and 83.57% in assessing the severity levels of TLB leaves.

Conclusions. Compared with other object detection and assessing methods, this proposed method achieved a good balance by maintaining high accuracy while requiring fewer parameters and computational resources.

Implications. The proposed method will aid farmers, policymakers, and researchers in better understanding the impact of the TLB disease on tea yield and in taking timely and effective measures.

Context. Kangaroo grass (Themeda triandra Forssk.) is a native perennial C₄ species significant to Dja Dja Wurrung people who seek to restore its presence across Country (Djandak) through broadacre seed crop production. To achieve this, agronomic challenges to establishment must be overcome.

Aims. To understand the effects of harvest time on seed viability and sowing time on crop establishment.

Methods. In Experiment 1, seed viability was assessed in a remnant Djandak stand in three seasons and seed colour assessed and cumulative seed shed measured in two of these seasons. In Experiment 2, seed from two Djandak ecotypes was sown at two sites at eight sowing dates over two seasons and plant emergence, culm number and canopy cover were recorded.

Key results. In Experiment 1, seed was shed from mid-December to late-January and seed viability varied intra- and inter-seasonally. Viability of early shed seed was low (0–24%) but increased with time to a peak of 68–69% in the first two seasons and 28–37% in the final season. Most seed had shed when peak viability was reached. Dark-coloured seeds with a caryopsis exhibited both high viability and high dormancy. In Experiment 2, sowing in September–October resulted in the optimal combination of highest mean establishment, lowest variability and no establishment failures.

Conclusions. To maximise crop establishment, seed should be sown in September–October on Djandak and be harvested when 30–50% of seed has shed.

Implications. These guidelines inform T. triandra establishment supportive of its development as a broadacre seed crop.

Context. In plant breeding, rapid development of homozygous lines is vital for accelerating varietal improvement. Doubled haploid (DH) breeding enables the production of fully homozygous lines in a single generation, whereas the effect of uniparental elimination following inter-generic hybridisation on seed fertility remains unclear.

Aims. To determine reliable and efficient approach of DH breeding by using a diverse panel of wheat types with Imperata cylindrica and maize as pollen sources and to estimate the relative fertility of the first-generation DHs.

Methods. Inter-generic hybridisation using pollens of composite variety of maize and I. cylindrica with 16 hexaploids, seven tetraploids and three wheat × rye derivatives (BC₁F₃) was undertaken and various haploid and DH induction parameters were evaluated. General combining ability (GCA) analysis was conducted to identify superior lines and the DHs developed were assessed for seed setting frequency.

Key results. Significant differences were found for most haploid induction parameters. GCA tests identified hexaploid genotypes (DH 86, HS 295, HPW 155) and tetraploid genotypes (A-9-30-1, PDW 314, PDW 191) as superior combiners, and I. cylindrica emerged as the most effective pollen source, especially in hexaploid wheat genotypes. Nineteen new first-generation DHs were developed with seed setting at par to their parental variety.

Conclusions. I. cylindrica was shown to be a more effective pollen source than was maize for DH production in wheat and the new DHs developed were true to type.

Implications. DH breeding can accelerate wheat breeding programs by producing homozygous lines efficiently, while retaining fertility levels similar to those of their parent lines.