Global Dataset Research Articles

Reducing the uncertainty in estimating soil microbial-derived carbon storage

Soil organic carbon (SOC) is the largest carbon pool in terrestrial ecosystems and plays a crucial role in mitigating climate change and enhancing soil productivity. Microbial-derived carbon (MDC) is the main component of the persistent SOC pool. However, current formulas used to estimate the proportional contribution of MDC are plagued by uncertainties due to limited sample sizes and the neglect of bacterial group composition effects. Here, we compiled the comprehensive global dataset and employed machine learning approaches to refine our quantitative understanding of MDC contributions to total carbon storage. Our efforts resulted in a reduction in the relative standard errors in prevailing estimations by an average of 71% and minimized the effect of global variations in bacterial group compositions on estimating MDC. Our estimation indicates that MDC contributes approximately 758 Pg, representing approximately 40% of the global soil carbon stock. Our study updated the formulas of MDC estimation with improving the accuracy and preserving simplicity and practicality. Given the unique biochemistry and functioning of the MDC pool, our study has direct implications for modeling efforts and predicting the land-atmosphere carbon balance under current and future climate scenarios.

Spatially Heterogeneous Responses of Planktonic Foraminiferal Assemblages Over 700,000 Years of Climate Change

ABSTRACTAimTo determine the degree to which assemblages of planktonic foraminifera track thermal conditions.LocationThe world's oceans.Time PeriodThe last 700,000 years of glacial–interglacial cycles.Major Taxa StudiedPlanktonic foraminifera.MethodsWe investigate assemblage dynamics in planktonic foraminifera in response to temperature changes using a global dataset of Quaternary planktonic foraminifera, together with a coupled Atmosphere–Ocean General Circulation Model (AOGCM) at 8000‐year resolution. We use ‘thermal deviance’ to assess assemblage responses to climate change, defined as the difference between the temperature at a given location and the bio‐indicated temperature (i.e., the abundance‐weighted average of estimated temperature optima for the species present).ResultsAssemblages generally tracked annual mean temperature changes through compositional turnover, but thermal deviances are evident under certain conditions. The coldest‐adapted species persisted in polar regions during warming but were not joined by additional immigrants, resulting in minimal assemblage turnover. The warmest‐adapted species persisted in equatorial regions during cooling with similarly minimal assemblage change. Assemblages at mid‐latitudes mostly tracked temperature cooling and warming.Main ConclusionsPlanktonic foraminiferal assemblages were generally able to track or endure temperature changes: as climate warmed or cooled, bio‐indicated temperature also became warmer or cooler, although to a variable degree. At polar sites under warming and at equatorial sites under cooling, the change in bio‐indicated temperature was less than, or even opposite to, what would be expected from estimated environmental change. Nevertheless, all studied species persisted across the study interval, regardless of thermal deviance—a result that highlights the resilience and inertia of planktonic foraminifera on an assemblage level to the last 700,000 years of climate change.

Enhancing the Resolution of Satellite Ocean Data Using Discretized Satellite Gridding Neural Networks

Ocean satellite data are often impeded by intrinsic limitations in resolution and accuracy. However, conventional data reconstruction approaches encounter substantial challenges when facing the nonlinear oceanic system and high-resolution fusion of variables. This research presents a Discrete Satellite Gridding Neural Network (DSGNN), a new machine learning method that processes satellite data within a discrete grid framework. By transforming the positional information of grid elements into a standardized vector format, the DSGNN significantly elevates the accuracy and resolution of data fusion through a neural network model. This method’s innovative aspect lies in its discretization and fusion technique, which not only enhances the spatial resolution of oceanic data but also, through the integration of multi-element datasets, better reflects the true physical state of the ocean. A comprehensive analysis of the reconstructed datasets indicates the DSGNN’s consistency and reliability across different seasons and oceanic regions, especially in its adept handling of complex nonlinear interactions and small-scale oceanic features. The DSGNN method has demonstrated exceptional competence in reconstructing global ocean datasets, maintaining small error variance, and achieving high congruence with in situ observations, which is almost equivalent to 1/12° hybrid coordinate ocean model (HYCOM) data. This study offers a novel and potent strategy for the high-resolution reconstruction and fusion of ocean satellite datasets.

A global coastal permeability dataset (CoPerm 1.0)

The permeability of aquifers strongly influences groundwater flow characteristics. Worldwide, coastal groundwater is often the primary freshwater source for coastal communities and ecosystems but is also particularly vulnerable to abstraction since saltwater intrusion may threaten its quality. Thus, understanding coastal permeability is crucial to the sustainable use of coastal groundwater. Here, we present the first global dataset of coastal permeability (CoPerm 1.0), which provides data on coasts’ landward, shoreline, and seaward permeability. CoPerm accounts for shoreline characteristics such as cliffs and beaches and contains information on four million segments representing more than two million kilometers of global coastline. Rocky Shores are the most abundant shoreline class, followed by mangroves, beaches, and muddy coasts. Permeability differs between the immediate shoreline (median permeability: 10−12.3 m2), the seaward (median: 10−13.3 m2), and the landward (median: 10−13 m2) sides of the coast. CoPerm provides input data for global coastal groundwater assessments and regional studies of submarine groundwater discharge or saltwater intrusion that can radiate into ecological and economic studies.

EDARA: An ERA5-based Dataset for Atmospheric River Analysis

Atmospheric Rivers (ARs) are long and narrow bands of strong horizontal water vapour transport concentrated in the lower troposphere. ARs play an important role in producing some high-impact weather events such as extreme precipitation and flooding, damaging winds, and temperature anomalies. To facilitate various studies on the short- and long-term variability of ARs and their impacts, I compiled a multi-decade global dataset containing 12 relevant meteorological variables for AR analysis. These variables were derived from the European Centre for Medium-Range Weather Forecasts atmospheric reanalysis version 5 (ERA5). They are available at 6-hour intervals from 1940 to present. Also included in the dataset is an interactive web browser-based graphical tool for visualising the AR evolution on regional (North America) and global scales. This ERA5-based Dataset for Atmospheric River Analysis (EDARA) may serve as a valuable resource for many AR-related research and applications.

ANALYSIS OF UZBEKISTAN'S RELATIONS WITH CHINA, RUSSIA, AND SOUTH KOREA: UTILIZING TEXT MINING BASED ON GDELT BIG DATA

GDELT (Global Dataset on Events, Language, and Tone), a comprehensive event dataset developed in 2013, provides quantifiable data on cooperative and conflictual relations between countries, as well as events related to specific phenomena. Its utility has been widely recognized, and it is actively used in international relations and foreign policy research. This study also utilizes GDELT data to introduce a method for analyzing the relationships between Uzbekistan and China, Russia, and South Korea through keyword correlation indicators derived from text mining, and to present the actual analysis results. By conducting a keyword correlation analysis based on event data between Uzbekistan and the three analyzed countries—China, Russia, and South Korea—it was possible to identify diplomatic action patterns through the correlation of core keywords such as "engage," "express intent," and "make" with keywords indicating specific actions. Uzbekistan’s foreign policy under the Mirziyoyev government has shown a diplomatic action pattern of seeking or participating in negotiations, expressing intentions for negotiation or cooperation, and pursuing meetings for negotiation and cooperation with the analyzed countries, while inviting or visiting counterpart countries as part of these processes. This demonstrates that Uzbekistan has actively worked towards establishing cooperative relations, which were set as the goal or direction of its foreign relations, named the “Good Neighbor Policy” by President Mirziyoyev. Such keyword correlation indicator analysis facilitates the explanation of cooperative relationships.

Mapping safe drinking water use in low- and middle-income countries.

Safe drinking water access is a human right, but data on safely managed drinking water services (SMDWS) is lacking for more than half of the global population. We estimate SMDWS use in 135 low- and middle-income countries (LMICs) at subnational levels with a geospatial modeling approach, combining existing household survey data with available global geospatial datasets. We estimate that only one in three people used SMDWS in LMICs in 2020 and identified fecal contamination as the primary limiting factor affecting almost half of the population of LMICs. Our results are relevant for raising awareness about the challenges and limitations of current global monitoring approaches and demonstrating how globally available geospatial data can be leveraged to fill data gaps and identify priority areas in LMICs.

Analyzing public sentiment toward economic stimulus using natural language processing

Purpose This study aims to investigate public sentiment toward economic stimulus using textual analysis. Specifically, it analyzes Twitter’s public opinion, emotion-based sentiment and topics related to COVID-19 economic stimulus packages. Design/methodology/approach This study applies natural language processing techniques, such as sentiment analysis, t-distributed stochastic neighbor embedding and semantic network analysis, to a global data set of 88,441 tweets from January 2020 to December 2021 extracted from the Twitter platform, discussing COVID-19 economic stimulus packages. Findings Results show that in the fourth quarter of 2021, there is a declining trend of positive sentiment (−5%) and an increasing trend of negative sentiment (14%), which may indicate the perceived inadequacy of COVID-19 stimulus measures. Topic modeling identifies seven topics, highlighting the necessity of stimulus in the education sector. Practical implications The big-data findings of this study provide a better understanding of public sentiment about economic stimulus for regulators and policymakers, which can help in formulating more effective fiscal and monetary policies. Originality/value Public sentiment is a significant concern for regulators because of its associated ambiguity, such as how to design stimulus packages and evaluate the effectiveness of previous measures. This study applies natural language processing, contributing to the growing literature on designing effective economic stimulus.

Shrub encroachment leads to accumulation of C, N, and P in grassland soils and alters C:N:P stoichiometry: A meta-analysis

Soil stoichiometry of carbon (C), nitrogen (N), and phosphorus (P) are indicators for nutrient balance. Shrub encroachment into grasslands could change nutrient concentrations and stoichiometry in soils, but the general patterns remain unclear. With a meta-analysis of a global dataset covering 344 observations from 68 studies, we examined the responses of grassland soil C:N:P stoichiometry to shrub encroachment under various environmental conditions. Our results show that: 1) Shrub encroachment significantly increased the concentrations of soil C (+29 %), N (+25 %), P (+20 %), C:N (+5 %), C:P (+12 %), and N:P (+6 %). The magnitude of such effects varied with climate, soil texture, and soil layer. 2) Increases in SOC and TN concentrations mainly occurred in Mediterranean and very humid climate zones. Soil C:P and N:P decreased in semi-humid climate zone after shrub encroachment. 3) The increases in SOC and TN concentrations and in the C:N, C:P, and N:P ratios after shrub encroachment were greater in the topsoil than in deeper soil layers. 4) Both finest-textured soil (clay) and coarsest-textured soil (sand) are beneficial for increase of soil nutrient concentrations following shrub encroachment. 5) The magnitude of the change in soil C:N was negatively correlated with the duration of shrub encroachment, due to greater increases in soil TN than in SOC concentrations with longer durations of encroachment. Our results indicate that soil stoichiometric shifts in shrub-encroached grasslands are relatively sensitive to environmental factors, including soil texture, soil pH, and climate. These findings help us to better understand the effects of shrub encroachment on biogeochemical cycling, functioning, and services in grasslands across a broad range of spatio-temporal scales.

Geographic Gradients in Species Interactions: From Latitudinal Patterns to Ecological Mechanisms

The idea that species interactions are more ecologically and evolutionarily important toward lower latitudes underpins seminal theories in ecology and evolution. Recent global studies have found the predicted latitudinal gradients in interactions, particularly predation. However, latitudinal patterns alone do not reveal why interactions vary geographically and so do not provide strong predictions in space (e.g., for specific ecosystems) or time (e.g., forecasting responses to global change). Here, I review theory to identify a clearer, mechanistic, and testable framework for predicting geographic variation in the importance of species interactions. I review competing metrics of importance, proximate mechanisms that can increase interaction importance, and environmental gradients that could generate predictable geographic patterns (climate extremes and stability, warmth, productivity, and biodiversity). Strong empirical tests are accumulating thanks to the rise of global experiments and datasets; renewed focus on testing why interactions vary spatially will help move the field from identifying latitudinal patterns to understanding broader mechanisms.

P-band radiometry for enhanced vegetation optical depth (VOD) and soil moisture retrieval in dense crop canopies

P-band (0.3–1 GHz) radiometry shows promise in enhancing the next generation of radiometer missions for global soil moisture and vegetation optical depth (VOD) monitoring, particularly in moderately to densely vegetated areas, which remain challenging for current L-band (1.4 GHz) missions such as Soil Moisture and Ocean Salinity (SMOS) and Soil Moisture Active Passive (SMAP). However, the superiority of P-band over L-band in this regard has not been experimentally verified, and the quantitative differences remain unexplored. To address this gap, the P-band Radiometer Inferred Soil Moisture (PRISM) tower-based experiment was conducted, pioneering a comprehensive evaluation of P- (0.75 GHz) and L-band (1.4 GHz) radiometry in densely vegetated areas with vegetation water content (VWC) up to 20 kg/m2. The results revealed that P-band VOD was consistently lower than L-band VOD, suggesting reduced vegetation attenuation and thus a higher VWC threshold for reliable soil moisture sensing at P-band. The VWC thresholds for P- and L-band were estimated at 7.7 kg/m2 and 4.0 kg/m2, respectively, suggesting that P-band radiation could go through the vegetation approximately twice as effectively as L-band radiation. Additionally, the Microwave Polarization Difference Index (MPDI) was identified as a strong predictor of retrieval performance because it established a good link between retrieval errors and VWC. Regarding VOD retrieval, robust logarithmic relationships (R2 > 0.9) between VOD and VWC were established for both bands within the 0–20 kg/m2 VWC range, challenging the previously assumed linear relationship. Moreover, P-band demonstrated greater retrieval robustness as its dual-polarized TB observations contain more independent information than those of L-band. This research establishes the foundation for a successful passive microwave mission at P-band, aiming to improve the quality of global VOD and soil moisture datasets.

Nitrite-oxidizing bacteria adapted to low-oxygen conditions dominate nitrite oxidation in marine oxygen minimum zones.

Nitrite is a central molecule in the nitrogen cycle because nitrite oxidation to nitrate (an aerobic process) retains fixed nitrogen in a system and its reduction to dinitrogen gas (anaerobic) reduces the fixed nitrogen inventory. Despite its acknowledged requirement for oxygen, nitrite oxidation is observed in oxygen-depleted layers of the ocean's oxygen minimum zones (OMZs), challenging the current understanding of OMZ nitrogen cycling. Previous attempts to determine whether nitrite-oxidizing bacteria in the anoxic layer differ from known nitrite oxidizers in the open ocean were limited by cultivation difficulties and sequencing depth. Here, we construct 31 draft genomes of nitrite-oxidizing bacteria from global OMZs. The distribution of nitrite oxidation rates, abundance and expression of nitrite oxidoreductase genes, and relative abundance of nitrite-oxidizing bacterial draft genomes from the same samples all show peaks in the core of the oxygen-depleted zone (ODZ) and are all highly correlated in depth profiles within the major ocean oxygen minimum zones. The ODZ nitrite oxidizers are not found in the Tara Oceans global dataset (the most complete oxic ocean dataset), and the major nitrite oxidizers found in the oxygenated ocean do not occur in ODZ waters. A pangenomic analysis shows the ODZ nitrite oxidizers have distinct gene clusters compared to oxic nitrite oxidizers and are microaerophilic. These findings all indicate the existence of nitrite oxidizers whose niche is oxygen-deficient seawater. Thus, specialist nitrite-oxidizing bacteria are responsible for fixed nitrogen retention in marine oxygen minimum zones, with implications for control of the ocean's fixed nitrogen inventory.

Co‐occurrence of surf breaks and carbon‐dense ecosystems suggests opportunities for coastal conservation

AbstractSurf breaks are increasingly recognized as socio‐environmental phenomena that provide opportunities for biodiversity conservation and sustained benefits for local communities. Here, we examine an additional benefit from improved conservation of the ecosystems that host and surround surf breaks—their coincidence with carbon dense coastal ecosystems. Using global spatial datasets of irrecoverable carbon (defined as carbon stocks that, if lost today, could not be recovered within 30 years' time), surf break locations, ecosystem types, protected areas, and Key Biodiversity Areas, we identified 88.3 million tonnes of irrecoverable carbon held in surf ecosystems. Of this total, 17.2 million tonnes are found in Key Biodiversity Areas without formal measures of protection. These results highlight surf conservation as a potential avenue to simultaneously mitigate climate change, protect biodiversity, and promote sustainable development in coastal communities.

The global distribution and climate resilience of marine heterotrophic prokaryotes

Heterotrophic Bacteria and Archaea (prokaryotes) are a major component of marine food webs and global biogeochemical cycles. Yet, there is limited understanding about how prokaryotes vary across global environmental gradients, and how their global abundance and metabolic activity (production and respiration) may be affected by climate change. Using global datasets of prokaryotic abundance, cell carbon and metabolic activity we reveal that mean prokaryotic biomass varies by just under 3-fold across the global surface ocean, while total prokaryotic metabolic activity increases by more than one order of magnitude from polar to tropical coastal and upwelling regions. Under climate change, global prokaryotic biomass in surface waters is projected to decline ~1.5% per °C of warming, while prokaryotic respiration will increase ~3.5% ( ~ 0.85 Pg C yr−1). The rate of prokaryotic biomass decline is one-third that of zooplankton and fish, while the rate of increase in prokaryotic respiration is double. This suggests that future, warmer oceans could be increasingly dominated by prokaryotes, diverting a growing proportion of primary production into microbial food webs and away from higher trophic levels as well as reducing the capacity of the deep ocean to sequester carbon, all else being equal.

Global response of soil biodiversity to climate and land use changes

Soil biodiversity (SB) is experiencing significant changes worldwide, yet its responses to climate and land use changes remain unclear. In light of this, we combined multiple environmental factors to construct a global SB dataset with a spatial resolution of 0.25° × 0.25° from 2000 to 2019. We determined that the global mean SB value was 0.66 during 2000–2019, and it slowly increased at a rate of 6.24 × 10−4/yr. The highest SB value (1.20) occurred within the Amazon Plain. Among the climate factors, temperature (T) was found to be responsible for the majority of the SB changes, accounting for approximately 40% of them. Compared to the results of other studies, our findings indicated that the SB increased across all types of land use and that urbanization had a positive impact on the increase in the SB. However, the most significant increase occurred in the pasture/range land. Precipitation (P) had similar effects on the SB in the pasture/range land and unmanaged grass/shrubland, while the areas with sparse or no vegetation experienced significant variations in temperature. These findings provide additional insights into the global pattern of SB and highlight the role of climate and land use changes in driving global and regional changes in SB.

Global patterns in the growth potential of soil bacterial communities

Despite the growing catalogue of studies detailing the taxonomic and functional composition of soil bacterial communities, the life history traits of those communities remain largely unknown. This study analyzes a global dataset of soil metagenomes to explore environmental drivers of growth potential, a fundamental aspect of bacterial life history. We find that growth potential, estimated from codon usage statistics, was highest in forested biomes and lowest in arid latitudes. This indicates that bacterial productivity generally reflects ecosystem productivity globally. Accordingly, the strongest environmental predictors of growth potential were productivity indicators, such as distance to the equator, and soil properties that vary along productivity gradients, such as pH and carbon to nitrogen ratios. We also observe that growth potential was negatively correlated with the relative abundances of genes involved in carbohydrate metabolism, demonstrating tradeoffs between growth and resource acquisition in soil bacteria. Overall, we identify macroecological patterns in bacterial growth potential and link growth rates to soil carbon cycling.

GLocal: A global development dataset of subnational administrative areas

The purpose of the GLocal dataset is to enable research in international development that requires both global scope and local precision. Leveraging modern geospatial analysis tools, we process a diverse array of sources to provide researchers with a growing set of economic, demographic, ecological and socio-political variables for geographic units relevant to public policy. We provide separate data files for different levels of administrative and periodic aggregation, along with ad-hoc files with more detailed information on specific topics. In this data descriptor paper, we discuss both our data processing methodologies and validation pipelines, and provide a short case study to illustrate the research potential of the dataset. We also introduce a simple web app, glocal.streamlit.app, which offers a user-friendly interface for exploring and visualizing the dataset. Given the growing number of public and granular sources of relevance for international development research, we hope to continue adding features and expand the GLocal dataset in the future.

Deep Learning-Based Amyloid PET Harmonization to Predict Cognitive Decline in Non-Demented Elderly

Abstract Background The robustness of conventional amyloid PET harmonization across tracers has been questioned. Purpose To evaluate deep learning-based harmonization of amyloid PET in predicting conversion from cognitively unimpaired (CU) to mild cognitive impairment (MCI) and MCI to Alzheimer’s disease (AD). Methods We developed an amyloid PET-based deep-learning model to classify participants with a clinical diagnosis of AD-dementia vs. CU across different tracers from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), Japanese ADNI, and Australian Imaging, Biomarker, and Lifestyle cohorts (n = 1,050). The model output (DL-ADprob), with other prognostic factors, was evaluated for predicting cognitive decline in ADNI-MCI (n = 451) and Harvard Aging Brain Study (HABS)-CU (n = 271) participants using Cox regression and area under time-dependent receiver operating characteristics curve (tdAUC) at 4-year follow-up. Subgroup analyses were performed in the ADNI-MCI group for conversion from amyloid-positive to AD and from amyloid negative to positive. Intraclass correlation coefficient (ICC) of DL-ADprob between tracers was calculated in the Global Alzheimer's Association Interactive Network dataset (n = 155). Results DL-ADprob was independently prognostic in both ADNI-MCI (P < 0.001) and HABS-CU (P = 0.048) sets. Adding DL-ADprob to other factors increased prognostic performances in both ADNI-MCI (tdAUC 0.758 [0.721–0.792] vs. 0.782 [0.742–0.818], tdAUC difference 0.023 [0.007–0.038]) and HABS-CU (tdAUC 0.846 [0.755–0.925] vs. 0.870 [0.773–0.943], tdAUC difference 0.022 [-0.004–0.053]). DL-ADprob was independently prognostic in amyloid-positive (P < 0.001) and amyloid-negative subgroups (P = 0.007). DL-ADprob showed incremental prognostic value in amyloid-positive (tdAUC 0.666 [0.623–0.713] vs. 0.706 [0.657–0.755], tdAUC difference 0.039 [0.016–0.064]), but not in amyloid-negative (tdAUC 0.818 [0.757–0.882] vs. 0.816 [0.751–0.880], tdAUC difference -0.002 [-0.031–0.029]) subgroup. The pairwise ICCs of DL-ADprob between Pittsburgh compound B and florbetapir, florbetaben, and flutemetamol respectively ranged from 0.913 to 0.935. Conclusion Deep learning-based harmonization of amyloid PET improves cognitive decline prediction in non-demented elderly, suggesting it could complement conventional amyloid PET measures.

Mesoscale eddies exert inverse latitudinal effects on global industrial squid fisheries

Squid species, as a burgeoning global food source, has garnered significant concerns due to expanding fisheries and little regulation. Elucidating the dynamics of squid fisheries and their biophysical coupling mechanisms is crucial for predicting spatiotemporal variations in squid fisheries and their sustainable management. Mesoscale eddies are discrete rotating oceanographic features that dominate local environmental variations and have been shown to modulate top predators. However, given controls of both predators and environmental factors, it remains unknown how eddies impact mid-trophic level species such as squids. Using satellite-based global squid fishery datasets, we showed an inverse latitudinal pattern of eddy-induced squid fisheries, where fishing activities are aggregated in (repelled from) cyclonic (anticyclonic) eddy cores in tropical waters and anticyclonic (cyclonic) eddy cores in temperate waters, and this pattern can be significantly enhanced with increasing eddy amplitude. Regarding solely the satellite-based global squid fisheries, eddy-induced environmental variations may generate a trade-off between food intake and energy expenditure, causing these oceanic squids to prefer cool cyclonic eddies in hot but food-limited waters, and warm anticyclonic eddies in nutritious but heat-limited waters. Given that eddy activity is projected to continuously enhance under global warming, our finding of eddy-driven bottom-up control for squid fisheries highlights an increasingly important hotspot for squid stock predictions and ecosystem-based ocean management in a changing climate.

Combining global precipitation data and machine learning to predict flood peaks in ungauged areas with similar climate

Increasing flood risk due to urbanization and climate change poses a significant challenge to societies at global scale. Hydrologic information that is required for understanding flood processes and for developing effective warning procedures is currently lacking in most parts of the world. Procedures that can combine global climate dataset from satellite and reanalysis with fast and low computational cost prediction systems, are attractive solutions for addressing flood predictions in ungauged areas. This work develops and tests a prediction framework that relies on two fundamental components. First, meteorological data from global datasets (IMERG and ERA5-Land) provide key input variables and second, ML models trained in the data-rich contiguous US, are applied in climatically similar regions in other parts of the world. Catchments in Australia, Brazil, Chile, Switzerland, and Great Britain were used as pseudo-ungauged regions for testing. Results indicate acceptable performance for both IMERG and ERA5-Land forced models with relative difference in flood peak prediction within 30 % and similar overall performance to locally trained ML models. Specific climate regions for which ML models have revealed good performance include Mediterranean climates like the US West Coast, subtropical areas like the Southern Atlantic Gulf, and mild temperate regions like the Mid-Atlantic Basin. This work highlights the potential of combining global precipitation dataset with pre-trained ML models in data-rich areas, for flood prediction in ungauged areas with similar climate.