Global Institute for Water Security

Jackfruit (Artocarpus heterophyllus Lam) is a vital tropical fruit species extensively cultivated across South and Southeast Asia, including Bangladesh and India. Renowned for its highly nutritious, large fruit, valuable timber, and medicinal properties, jackfruit demonstrates resilience across diverse climatic conditions. Its inherent highly heterozygous nature, resulting from cross-pollination, leads to extensive genetic diversity. While this makes jackfruit an excellent candidate for improvement through molecular breeding, comprehensive genomic studies to identify SNPs (single-nucleotide polymorphisms) within the jackfruit genome have been lacking until this study. This study is aimed to characterize the genetic variation and population structure of jackfruit using SNP markers. A total of 37 jackfruit accessions were collected from three regional stations of the Bangladesh Agricultural Research Institute and two local farmers' fields. Morphological traits, including fruit size, color, number of bulbs per fruit, seed weight, seed size, bulb weight, and brix (%), were recorded during the 2023-24 growing season. Genomic DNA was extracted, and Genotyping-by-Sequencing libraries were prepared using a two-enzyme protocol, followed by sequencing on the NovaSeq Illumina platform. After stringent quality control, trimming, and bioinformatics processing, 23,795 high-quality biallelic SNP markers were identified by filtering for minor allele frequency > 0.01 and missing data < 20%. These SNP markers were then utilized to comprehensively characterize genetic variation and population structure among the accessions. Population structure analysis of 37 jackfruit accessions identified five genetic clusters, reflecting both shared ancestry and distinct divergent lineages (e.g., GK9/GK9.1, AK005/AK008, BK1/PD19.1, GK2/GK4). The results demonstrate the efficacy of SNP markers for genetic diversity studies, providing crucial information for parental selection and supporting advanced breeding and conservation strategies. This study significantly contributes valuable insights into jackfruit's genetic diversity, establishing a foundation for genomic-assisted breeding and conservation initiatives for food security.

• Significant transparency and responsible modelling are vital for high-stakes AI forecasting. • The reported forecast accuracy of Google’s AI flood model is substantially overstated when evaluated against real observational data. • Google’s AI flood model relies on several subjective and questionable evaluation choices. The development of AI models is increasing at a rapid rate. However, when are they ready to be deployed in real-world operational settings? In this paper, we introduce a framework to support such assessments and apply it to Google’s recently released AI-based flood prediction system, which is claimed to achieve “reliability in predicting extreme riverine events” and provide “accurate and timely warnings” that are available “earlier and over larger and more impactful events in ungauged basins”. The system has been integrated into an operational early-warning platform producing open, real-time forecasts in more than 80 countries. While this development promises to usher in a new and exciting age in global flood forecasting, the supporting evidence relies heavily on several subjective choices, the implications of which have not been acknowledged or assessed. Here, we evaluate the consequences of these choices on claims of operational deployment readiness across four dimensions: predictive accuracy, forecast timeliness, the characterization of extreme events, and benchmarking against state-of-the-art models. Our assessment reveals that the system’s actual predictive accuracy is likely to be substantially lower than reported—particularly for extreme events—raising concerns about responsible practices across modelling and publicity in high-stakes applications. The deployment of the Google AI model therefore risks misinforming those who depend on its outputs for evacuation and preparedness decisions, particularly in less-developed countries such as those targeted by the enterprise, given its alarmingly high (>90%) rates of false positives and false negatives. Beyond the immediate operational consequences, if left unaddressed, these outcomes may erode public trust in AI within hydrological sciences. We conclude by calling for greater transparency, accountability, and methodological rigor in the integration of AI into flood forecasting.

The formation and stability of river ice covers in regulated waterways are critical for uninterrupted hydro-electric operations. This study investigates the modelling of ice cover development in the Beauharnois Canal along the St. Lawrence River with the presence and absence of ice booms. Ice booms are deployed in this canal to promote the rapid formation of a stable ice cover during freezing events, minimizing disruptions to dam operations. Remote sensing data were used to assess the spatial extent and temporal evolution of an ice cover and to calibrate the river ice model RIVICE. The model was applied to simulate ice formation for the 2019–2020 ice season, first for the canal with a series of three ice booms and then rerun under a scenario without booms. Comparative analysis reveals that the presence of ice booms facilitates the development of a relatively thinner and more uniform ice cover. In contrast, the absence of booms leads to thicker ice accumulations and increased risk of ice jamming, which could impact water management and hydroelectric generation operations. Computational efficiencies of the RIVICE model were also sought. RIVICE was originally compiled with a Fortran 77 compiler, which restricted modern optimization techniques. Recompiling with NVFortran significantly improved performance through advanced instruction scheduling, cache management, and automatic loop analysis, even without explicit optimization flags. Enabling optimization further accelerated execution, albeit marginally, reducing redundant operations and memory traffic while preserving numerical integrity. Tests across varying ice cross-sectional spacings confirmed that NVFortran reduced runtimes by roughly an order of magnitude compared to the original model. A test GPU (Graphics Processing Unit) version was able to run the data interpolation routines on the GPU, but frequent data transfers between the CPU (Central Processing Unit) and GPU caused by shared memory blocks and fixed-size arrays made it slower than the original CPU version. Achieving efficient GPU execution would require substantial code restructuring to eliminate global states, adopt persistent data regions, and parallelize at higher level loops, or alternatively, rewriting in a GPU-friendly language to fully exploit modern architectures.

<h3>Abstract</h3> As the threat of flooding intensifies, the need for policies that align with household preferences for addressing flood risk becomes increasingly important. We conducted a discrete choice experiment to elicit homeowner preferences for purchasing overland flood insurance and/or making risk-reducing home improvements. Results from a latent class logit model indicate the presence of four preference classes, characterized by their preferred response to flood risk: insurance, home improvements, insurance and home improvements, and neither insurance nor home improvements. We subsequently evaluate household responses to insurance and home-improvement subsidies, offering insights into the effectiveness of incentives for encouraging protective actions against flooding.

Winter camelina (Camelina sativa) is a climate-resilient oilseed crop that has received attention as a feedstock crop for advanced, low-carbon-intensity biofuels. Breeding programs working on winter camelina improvement have to contend with heterogeneous germplasm, oftentimes erroneously identified as winter biotypes, and a gene pool that is much smaller than that of spring-type camelina, the latter having motivated crosses between winter and spring biotypes. For the unequivocal differentiation of winter from spring types at an early stage, breeders require a tool to track the vernalization requirement trait in segregating breeding populations as well as in putative winter cultivars, breeding lines, and accessions to be used as parental lines. Linkage mapping in a winter ('Joelle') × spring ('SES0787LS') C. sativa biparental F2 population identified two major quantitative trait loci (QTLs) for vernalization requirement on chromosomes 8 and 13. Both regions contained orthologs of Flowering Locus C (FLC), a gene known to have a significant effect on flowering time and vernalization requirement in plants. Based on the FLC gene sequences, allele-specific PCR-based markers were developed, suitable for the routine screening of C. sativa germplasm for the presence of the winter and spring alleles of all three C. sativa FLC orthologs, including a chromosome 20 locus. The analysis of the winter cultivar 'Joelle' and a diverse C. sativa germplasm panel uncovered greater than expected variability for the FLC alleles, with most lines possessing several different allele combinations and still undergoing genetic segregation. Contrary to previous reports, spring camelina lines can carry the spring and/or winter alleles of Csa.FLC.C20, indicating that this gene by itself only plays a subordinate role in the regulation of flowering and vernalization requirement. In winter C. sativa germplasm, combinations of Csa.FLC.C08 winter alleles with the winter alleles of one or both of Csa.FLC.C13 and Csa.FLC.C20 result in vernalization requirement, while winter Csa.FLC.C08 by itself leads to a semi-winter type. The results of this study and the tools developed herein are a first step to orchestrating the genes underlying vernalization requirement in C. sativa and developing winter camelina cultivars optimized for different winter environments.

Groundbreaking advances in gene editing technologies are transforming modern plant breeding by enabling precise genetic modifications that dramatically accelerate crop improvement. Haploid and diploid induction systems have emerged as particularly powerful tools in this landscape, offering both efficient gene editing capabilities and rapid production of homozygous lines while seamlessly integrating with the advanced genome-editing platforms such as CRISPR-Cas systems. This review synthesizes the current state of knowledge regarding the mechanisms, applications, and recent progress in haploid and diploid induction systems for gene editing. We examine their transformative potential for enhancing genetic gains and compressing breeding timelines, with significant implications for global food security. Additionally, we provide a critical analysis of emerging challenges of genome editing in crops and outline promising future directions for research and development.

Nontargeted high-resolution mass spectrometry (HRMS) allows for the characterization of a large fraction of the exposome, i.e., the entirety of chemicals an organism is exposed to, and helps detect important exogenous chemical compounds that could be key drivers of toxicological impact. Along with these chemical compounds occur endogenous metabolites that are essential for the health of the host organism. Chemical compounds derived from the biotransformation of xenobiotics present in the exposome are referred to as the xenometabolome, while endogenous metabolites derived from the host organism are referred to as the endometabolome. Recent advancements in HRMS technology allow for the detection of chemical features of biological and ecological importance in the context of chemical safety assessments with unprecedented sensitivity and resolution. In this perspective, we highlight the application of HRMS-based metabolomics of organisms in the context of ecotoxicology, the complexity of comprehensively characterizing the endometabolome, and distinguishing chemical compounds of the xenometabolome.