Search

Credit risk refers to the possibility of borrower default, and its assessment is crucial for maintaining financial stability. However, the journey of credit risk data generation is often gradual, and machine learning techniques may not be readily applicable for crafting evaluations at the initial stage of the data accumulation process. This article proposes a credit risk modeling methodology, TED-NN, that first constructs an indicator system based on expert experience, assigns initial weights to the indicator system using the Analytic Hierarchy Process, and then constructs a neural network model based on the indicator system to achieve a smooth transition from an empirical model to a data-driven model. TED-NN can automatically adapt to the gradual accumulation of data, which effectively solves the problem of risk modeling and the smooth transition from no to sufficient data. The effectiveness of this methodology is validated through a specific case of credit risk assessment. Experimental results on a real-world dataset demonstrate that, in the absence of data, the performance of TED-NN is equivalent to the AHP and better than untrained neural networks. As the amount of data increases, TED-NN gradually improves and then surpasses the AHP. When there are sufficient data, its performance approaches that of a fully data-driven neural network model.

The accurate prediction of global horizontal irradiance (GHI) is crucial for optimizing solar power generation systems, particularly in mountainous areas with complex topography and unique microclimates. These regions face significant challenges due to limited reliable data and the dynamic nature of local weather conditions, which complicate accurate GHI measurement. The scarcity of precise data impedes the development of reliable solar energy prediction models, impacting both economic and environmental outcomes. To address these data scarcity challenges in solar energy prediction, this paper focuses on various locations in Europe and Asia Minor, predominantly in mountainous regions. Advanced machine learning techniques, including random forest (RF) and extreme gradient boosting (XGBoost) regressors, are employed to effectively predict GHI. Additionally, optimizing training data distribution based on cloud opacity values and integrating synthetic data significantly enhance predictive accuracy, with R2 scores ranging from 0.91 to 0.97 across multiple locations. Furthermore, substantial reductions in root mean square error (RMSE), mean absolute error (MAE), and mean bias error (MBE) underscore the improved reliability of the predictions. Future research should refine synthetic data generation, optimize additional meteorological and environmental parameter integration, extend methodology to new regions, and test for predicting global tilted irradiance (GTI). The studies should expand training data considerations beyond cloud opacity, incorporating sky cover and sunshine duration to enhance prediction accuracy and reliability.

Tumor boundary identification during colorectal cancer surgery can be challenging, and incomplete tumor removal occurs in approximately 10% of the patients operated for advanced rectal cancer. In this paper, a deep learning framework for automatic tumor segmentation in colorectal ultrasound images was developed, to provide real-time guidance on resection margins using intra-operative ultrasound. A colorectal ultrasound dataset was acquired consisting of 179 images from 74 patients, with ground truth tumor annotations based on histopathology results. To address data scarcity, transfer learning techniques were used to optimize models pre-trained on breast ultrasound data for colorectal ultrasound data. A new custom gradient-based loss function (GWDice) was developed, which emphasizes the clinically relevant top margin of the tumor while training the networks. Lastly, ensemble learning methods were applied to combine tumor segmentation predictions of multiple individual models and further improve the overall tumor segmentation performance. Transfer learning outperformed training from scratch, with an average Dice coefficient over all individual networks of 0.78 compared to 0.68. The new GWDice loss function clearly decreased the average tumor margin prediction error from 1.08 mm to 0.92 mm, without compromising the segmentation of the overall tumor contour. Ensemble learning further improved the Dice coefficient to 0.84 and the tumor margin prediction error to 0.67 mm. Using transfer and ensemble learning strategies, good tumor segmentation performance was achieved despite the relatively small dataset. The developed US segmentation model may contribute to more accurate colorectal tumor resections by providing real-time intra-operative feedback on tumor margins.

Several recent studies have evidenced the relevance of machine-learning for soil salinity mapping using Sentinel-2 reflectance as input data and field soil salinity measurement (i.e., Electrical Conductivity-EC) as the target. As soil EC monitoring is costly and time consuming, most learning databases used for training/validation rely on a limited number of soil samples, which can affect the model consistency. Based on the low soil salinity variation at the Sentinel-2 pixel resolution, this study proposes to increase the learning database’s number of observations by assigning the EC value obtained on the sampled pixel to the eight neighboring pixels. The method allowed extending the original learning database made up of 97 field EC measurements (OD) to an enhanced learning database made up of 691 observations (ED). Two classification machine-learning models (i.e., Random Forest-RF and Support Vector Machine-SVM) were trained with both OD and ED to assess the efficiency of the proposed method by comparing the models’ outcomes with EC observations not used in the models´ training. The use of ED led to a significant increase in both models’ consistency with the overall accuracy of the RF (SVM) model increasing from 0.25 (0.26) when using the OD to 0.77 (0.55) when using ED. This corresponds to an improvement of approximately 208% and 111%, respectively. Besides the improved accuracy reached with the ED database, the results showed that the RF model provided better soil salinity estimations than the SVM model and that feature selection (i.e., Variance Inflation Factor-VIF and/or Genetic Algorithm-GA) increase both models´ reliability, with GA being the most efficient. This study highlights the potential of machine-learning and Sentinel-2 image combination for soil salinity monitoring in a data-scarce context, and shows the importance of both model and features selection for an optimum machine-learning set-up.

The ground deformation rate is an important index for evaluating the stability and degradation of permafrost. Due to limited accessibility, in-situ measurement of the ground deformation of permafrost areas on the Tibetan Plateau is a challenge. Thus, the technique of time-series interferometric synthetic aperture radar (InSAR) is often adopted for measuring the ground deformation rate of the permafrost area, the effectiveness of which is, however, degraded in areas with geometric distortions in synthetic aperture radar (SAR) images. In this study, a method that integrates InSAR and the random forest method is proposed for an improved permafrost stability mapping on the Tibetan Plateau; to demonstrate the application of the proposed method, the permafrost stability mapping in a small area located in the central region of the Tibetan Plateau is studied. First, the ground deformation rate in the concerned area is studied with InSAR, in which 67 Sentinel-1 scenes taken in the period from 2014 to 2020 are collected and analyzed. Second, the relationship between the environmental factors (i.e., topography, land cover, land surface temperature, and distance to road) and the permafrost stability is mapped with the random forest method based on the high-quality data extracted from the initial InSAR analysis. Third, the permafrost stability in the whole study area is mapped with the trained random forest model, and the issue of data scarcity in areas where the terrain visibility of SAR images is poor or InSAR results are not available in permafrost stability mapping can be overcome. Comparative analyses demonstrate that the integration of the InSAR and the random forest method yields a more effective permafrost stability mapping compared with the sole application of InSAR analysis.

Satellite remote sensing provides useful gridded data for the conceptual modelling of hydrological processes such as precipitation–runoff relationship. Structurally flexible and computationally advanced AI-assisted data-driven (DD) models foster these applications. However, without linking concepts between variables from many grids, the DD models can be too large to be calibrated efficiently. Therefore, effectively formulized, collective input variables and robust verification of the calibrated models are desired to leverage satellite data for the strategic DD modelling of catchment runoff. This study formulates new satellite-based input variables, namely, catchment- and event-specific areal precipitation coverage ratios (CCOVs and ECOVs, respectively) from the Global Precipitation Mission (GPM) and evaluates their usefulness for monthly runoff modelling from five mountainous Karkheh sub-catchments of 5000–43,000 km2 size in west Iran. Accordingly, 12 different input combinations from GPM and MODIS products were introduced to a generalized deep learning scheme using artificial neural networks (ANNs). Using an adjusted five-fold cross-validation process, 420 different ANN configurations per fold choice and 10 different random initial parameterizations per configuration were tested. Runoff estimates from five hybrid models, each an average of six top-ranked ANNs based on six statistical criteria in calibration, indicated obvious improvements for all sub-catchments using the new variables. Particularly, ECOVs were most efficient for the most challenging sub-catchment, Kashkan, having the highest spacetime precipitation variability. However, better performance criteria were found for sub-catchments with lower precipitation variability. The modelling performance for Kashkan indicated a higher dependency on data partitioning, suggesting that long-term data representativity is important for modelling reliability.

Assessing how climate change will affect hydrological ecosystem services (HES) provision is necessary for long-term planning and requires local comprehensive climate information. In this study, we used SWAT to evaluate the impacts on four HES, natural hazard protection, erosion control regulation and water supply and flow regulation for the Laguna del Sauce catchment in Uruguay. We used downscaled CMIP-5 global climate models for Representative Concentration Pathways (RCP) 2.6, 4.5 and 8.5 projections. We calibrated and validated our SWAT model for the periods 2005–2009 and 2010–2013 based on remote sensed ET data. Monthly NSE and R2 values for calibration and validation were 0.74, 0.64 and 0.79, 0.84, respectively. Our results suggest that climate change will likely negatively affect the water resources of the Laguna del Sauce catchment, especially in the RCP 8.5 scenario. In all RCP scenarios, the catchment is likely to experience a wetting trend, higher temperatures, seasonality shifts and an increase in extreme precipitation events, particularly in frequency and magnitude. This will likely affect water quality provision through runoff and sediment yield inputs, reducing the erosion control HES and likely aggravating eutrophication. Although the amount of water will increase, changes to the hydrological cycle might jeopardize the stability of freshwater supplies and HES on which many people in the south-eastern region of Uruguay depend. Despite streamflow monitoring capacities need to be enhanced to reduce the uncertainty of model results, our findings provide valuable insights for water resources planning in the study area. Hence, water management and monitoring capacities need to be enhanced to reduce the potential negative climate change impacts on HES. The methodological approach presented here, based on satellite ET data can be replicated and adapted to any other place in the world since we employed open-access software and remote sensing data for all the phases of hydrological modelling and HES provision assessment.

The Data Scarcity problem is repeatedly encountered in environmental research. This may induce an inadequate representation of the response’s complexity in any environmental system to any input/change (natural and human-induced). In such a case, before getting engaged with new expensive studies to gather and analyze additional data, it is reasonable first to understand what enhancement in estimates of system performance would result if all the available data could be well exploited. The purpose of this Special Issue, “Overcoming Data Scarcity in Earth Science” in the Data journal, is to draw attention to the body of knowledge that leads at improving the capacity of exploiting the available data to better represent, understand, predict, and manage the behavior of environmental systems at meaningful space-time scales. This Special Issue contains six publications (three research articles, one review, and two data descriptors) covering a wide range of environmental fields: geophysics, meteorology/climatology, ecology, water quality, and hydrology.

The Kabul River basin (KRB) of Afghanistan, a lifeline of around 10 million people, has multiplicity of governance, management, and development-related challenges leading to inequity, inadequacy, and unreliability of irrigation water distribution. Prior to any uplifting intervention, there is a need to evaluate the performance of irrigation system on a long term basis to identify the existing bottlenecks. Although there are several indicators available for the performance evaluation of the irrigation schemes, we used the coefficient of variation (CV) of actual evapotranspiration (ETa) in space (basin, sub-basin, and provincial level), relative evapotranspiration (RET), and temporal CV of RET, to assess the equity, adequacy, and reliability of water distribution, respectively, from 2003 to 2013. The ETa was estimated through a surface energy balance system (SEBS) algorithm and the ETa estimates were validated using the advection aridity (AA) method with a R2 value of 0.81 and 0.77 at Nawabad and Sultanpur stations, respectively. The global land data assimilation system (GLDAS) and moderate-resolution imaging spectroradiometer (MODIS) products were used as main inputs to the SEBS. Results show that the mean seasonal sub-based RET values during summer (May–September) (0.37 ± 0.06) and winter (October–April) (0.40 ± 0.08) are below the target values (RET ≥ 0.75) during 2003–2013. The CV of the mean ETa, within sub-basins and provinces for the entire study period, has an equitable distribution of water from October–January (0.09 ± 0.04), whereas the highest inequity (0.24 ± 0.08) in water distribution is during early summer. The range of the CV of the mean ETa (0.04–0.06) on a monthly and seasonal basis shows the unreliability of water supplies in several provinces or sub-basins. The analysis of the temporal CV of mean RET highlights the unreliable water supplies across the entire basin. The maximum ETa during the study period was estimated for the Shamal sub-basin (552 ± 43 mm) while among the provinces, Kunar experienced the highest ETa (544 ± 39 mm). This study highlights the dire need for interventions to improve the irrigation performance in time and space. The proposed methodology can be used as a framework for monitoring and implementing water distribution plans in future.

Insect infestation behaves as a complex system, characterized by non-linear spatial dynamics and emergent patterns that evolve from smaller to larger spatial scales. The emerald ash borer (EAB) is an invasive species that has infested and killed millions of ash trees across North America. Existing EAB models use traditional statistical approaches that often cannot address the spatiotemporal complexity emerging from EAB infestation processes. Moreover, these studies of insect infestation are limited by a lack of sufficient time series data. The objective of this study is to develop a geosimulation approach to overcome the challenge of data scarcity and represent EAB infestation at a regional scale. Geographic information systems (GIS), multi-criteria evaluation (MCE), and cellular automata (CA) are used to model EAB spread across different hypothetical landscape types. Simulation results represent EAB propagation and indicate different dynamics of spread for each landscape. Urban environments are identified as being at the greatest risk to the infestation. The proposed approach offers a theoretical framework and a modeling tool to represent the propagation of EAB infestation that can be applied with real geospatial datasets and potentially used in forest management strategies.