Downscaling Of Precipitation Research Articles

Investigation of impact of climate change on small catchments using different climate models and statistical approaches

Abstract The use of a statistical downscaling technique is needed to investigate the hydrological consequences of climate change on the local hydropower capacity. Global Circulation Models (GCMs) are crucial tools used in various simulations for potential climate change effects, including precipitation and temperature. Statistical downscaling methods comprise the improvement of relations between the large-scale climatic parameters and the local variables. This study presents the trend analysis of the observed variables compared to the statistically downscaled emission scenarios that are adopted from the Canadian Second Generation Earth Systems Model (CanESM2) in the basin of Göksu River which is located in Turkey. The key purpose of the research is to evaluate both the predicted monthly precipitation and the projections of GCMs within the three simulated scenarios of RCP2.6, RCP4.5, and RCP8.5 by Gene Expression Programming (GEP). In addition, the findings of statistical downscaling of monthly mean precipitation will be compared to the Linear Regression (LR) model. The R-value is 0.827 and 0.755 for precipitation of the GEP model for the periods of calibrating and validation. In comparison with the LR model for the validation and calibration periods (1971–2005), the results of the GEP model prove its applicability in projecting the data for monthly mean rainfall. Generally, in the simulated periods of 2021–2100, the mentioned scenarios forecast a decline in the monthly mean precipitation in the basin. Moreover, the scenario of RCP8.5 projected more suitably for the case study than expected under the scenarios RCP4.5 and RCP2.6. The mean statistically downscaled CanESM2 model was compared with the trend analysis of the areal mean precipitation over the case study area, and the trend was shown as decreasing. However, the RCP 8.5 scenario was the more quasi-asymptotic for trend.

Impact of climate change on hydro-climatological parameters in North Cyprus: application of artificial intelligence-based statistical downscaling models

Abstract There are many environmental challenges in water-limited places in the 21st century, particularly in dry and semi-arid regions, due to the threat of climate change caused by the greenhouse effect. This study intends to explore and assess the influence of climate change on hydro-climatological parameters using statistical downscaling and future forecasts of mean monthly precipitation and temperature throughout Famagusta (Mağusa), Nicosia (Lefkoşa), and Kyrenia (Girne) stations, North Cyprus. To achieve the study's goal, 13 predictors of BNU-ESM GCMs from CMIP5 were used at a grid point in the Karfas region. To find the primary predictors, GCM data were screened using mutual information (MI) and correlation coefficient (CC) feature extraction methods prior to downscaling modeling. A neural network (ANN), an adaptive neuro fuzzy inference system (ANFIS), and multiple linear regression (MLR) models were employed as the downscaling models. We used the best downscaling model as a benchmark for future precipitation and temperature estimates for the period 2018–2040 under the RCP4.5 scenario. In the future, Famagusta and Nicosia would have up to 22% less rain, and Famagusta and Kyrenia will have 2.9% greater heat. The findings of this research could be useful in decision-making, as well as water resource management and climate change.

Effects of the spatial and temporal resolution of meteorological data on the accuracy of precipitation estimation by means of CNN

Future climate projections are valuable datasets to investigate the impacts of future climate changes on natural disasters such as intense precipitation, severe flood, and drought. However, they are too coarse depending on the purpose, and downscaling is required in such a case. There is nowadays a downscaling technique using deep learning such as CNN. Atmospheric information can be used as an input for precipitation downscaling by means of Convolutional Neural Network (CNN). For such precipitation downscaling, the spatial and temporal resolution of the atmospheric information may be important. This study obtained atmospheric information from a coarser reanalysis dataset and a finer reanalysis dataset as input for precipitation downscaling. As a coarser reanalysis dataset, ERA-Interim was selected. As a finer reanalysis dataset, ERA5 was utilized. Then, this study investigated the effect of spatial and temporal resolution of input data on the estimation accuracy of precipitation downscaling by CNN. For simplification, daily average precipitation at a watershed was used as the target data. The results show advantage of the use of a higher resolution as input can improve the model accuracy.

Assessing the impacts of climate change on drought-prone regions in Bhima sub-basin (India) using the Standard Precipitation Index

Abstract In Bhima sub-basin, the water sector is at high demand and in critical stress due to rapid urbanization. The past few decades have witnessed extreme events and seasonal shifts due to anthropogenic activity triggered climate change. Thus, to evaluate the variability of extreme events, assessing the historical and future trends of precipitation in climate change scenarios is vital for developing comprehensive mitigation and adaptation strategies. This paper examines the drought-prone regions by studying spatio-temporal variation of drought scenarios using the Standard Precipitation Index (SPI). The change factor method is used to downscale precipitation data from general circulation model (GCM) outputs under four Representative Concentration Pathway (RCP) scenarios to project future downscaled precipitation, to be input to examine the drought for 12 months. GCM and scenario uncertainty in climate change impact assessments are examined using box-whisker plots. Temporal variation for 12-month SPI shows significant changes over RCP scenarios. For the beginning of the period, 2021 precipitation is scanty for RCP 2.6 and 4.5 scenarios. Mild to moderate and severe drought events for the RCP 2.6 scenario are more predominant. Severe drought events under the RCP 6.0 scenario dominate over others. Lastly, the inconsistent pattern of drought events for RCP 8.5 is reported.

Spatio-temporal variations of land surface temperature and precipitation due to climate change in the Jhelum river basin, India

The present study investigated the spatio-temporal variations of precipitation and temperature for the projected period (2011-2100) in the Jhelum basin, India. The precipitation and temperature variables are projected under RCP 8.5 scenario using statistical down scaling techniques such as Artificial Neural Network (ANN) and Wavelet Artificial Neural Network (WANN) models. Firstly, the screened predictors were downscaled to predictand using ANN and WANN models for all the study stations. On the basis of the performance criteria, the WANN model is selected as an efficient model for downscaling of precipitation and temperature. The future screened predictor data pertaining to RCP 8.5 of CanESM2 model were downscaled to monthly temperature and precipitation for future periods (2011-2100) using WANN models. The investigation of the future projections revealed an average increase of 17-25% in the mean annual precipitation and 20-25% average increase in the monthly mean precipitation for all the selected stations towards the end of 21st century. The monthly mean temperature also showed an increase of 2-3 °C for all the study stations towards the end of 21st century. The mean seasonal temperature of the projected period is found to be increasing for all the four seasons in most parts of the basin.

Statistical downscaling of daily precipitation over southeastern South America: Assessing the performance in extreme events

AbstractThe performance of multiple empirical statistical downscaling (ESD) methods was assessed for simulating daily precipitation during 1979–2017 over southeastern South America (SESA), a region where extremes are remarkable. Meteorological stations were used as reference and three gridded precipitation products were included to account for observational uncertainty. The set of ESD models involved different configurations of the analogue method (ANs), deterministic and stochastic versions of neural networks (NNs) and generalized linear models (GLMs) and circulation‐conditioned GLMs (GLM_WTs). The years with the largest number of extreme events (wet years) were calibrated separately. The spatio‐temporal variability of extremes was assessed by analysing their intensity, spatial extent, frequency and interannual variability. An overall good performance of the ESD models was found for several aspects of daily precipitation. ESD performance dispersion was usually contained in the observational spread. No particular model configuration was found to perform best in all aspects, indicating the advantage of considering a multi‐model ensemble. The ANs tended to follow the stations, satisfactorily simulating daily precipitation and its extremes. The deterministic GLMs strongly underestimated precipitation estimates and were not able to represent extreme frequencies and intensities, but this was alleviated by employing a stochastic version of the method. Furthermore, the use of weather types to condition the GLMs (GLM_WTs) considerably improved model performance, particularly for the annual cycle and the spatial structure of extreme precipitation. The NNs adequately reproduced the spatial behaviour and intra‐annual variability of extreme precipitation, although they underestimated its intensity in their deterministic version. An analysis of the regional time series of extremes showed consistency among datasets and evidenced the influence of the ENSO teleconnection on the wet years, which were commonly well‐simulated by the statistical models. ESD models presented good skills in simulating a wetter climate over SESA, which is of particular importance in a climate change scenario.

Statistical downscaling of precipitation using inclusive multiple modelling (IMM) at two levels

Abstract Topical research on hydrological behaviour of climate change in terms of downscaling of monthly precipitation is investigated in this paper by formulating an inclusive multiple modelling (IMM) strategy. IMM strategies manage multiple models at two levels and the paper uses statistical downscaling model, Sugeno fuzzy logic and support vector machine at Level 1 and feeds their outputs to a neuro-fuzzy model at Level 2. In the downscaling stage, large-scale NCEP (National Centres for Environmental Prediction)/NCAR (National Centre for Atmospheric Research) data are used for a station with local data record from 1961 to 2005 for training and testing Level 1 models. The results are found to be ‘fit-for-purpose’, but the variations between them signify some room for improvements. The model at Level 2 combines outputs of those at Level 1 and produces Level 2 results, which improve compared with those at the Level 1 models in terms of dispersion of residual errors. In this way, IMM provides a more defensible modelling strategy for application in the projection stage. The comparison between observed and projected precipitation indicates that precipitation will be likely to reduce compared with observed precipitation in cold seasons (October–February), but the projected precipitation will be likely to increase slightly in wet seasons (April and May).

TRU-NET: a deep learning approach to high resolution prediction of rainfall

Climate models (CM) are used to evaluate the impact of climate change on the risk of floods and heavy precipitation events. However, these numerical simulators produce outputs with low spatial resolution that exhibit difficulties representing precipitation events accurately. This is mainly due to computational limitations on the spatial resolution used when simulating multi-scale weather dynamics in the atmosphere. To improve the prediction of high resolution precipitation we apply a Deep Learning (DL) approach using input data from a reanalysis product, that is comparable to a climate model’s output, but can be directly related to precipitation observations at a given time and location. Further, our input excludes local precipitation, but includes model fields (weather variables) that are more predictable and generalizable than local precipitation. To this end, we present TRU-NET (Temporal Recurrent U-Net), an encoder-decoder model featuring a novel 2D cross attention mechanism between contiguous convolutional-recurrent layers to effectively model multi-scale spatio-temporal weather processes. We also propose a non-stochastic variant of the conditional-continuous (CC) loss function to capture the zero-skewed patterns of rainfall. Experiments show that our models, trained with our CC loss, consistently attain lower RMSE and MAE scores than a DL model prevalent in precipitation downscaling and outperform a state-of-the-art dynamical weather model. Moreover, by evaluating the performance of our model under various data formulation strategies, for the training and test sets, we show that there is enough data for our deep learning approach to output robust, high-quality results across seasons and varying regions.

Comparison of CMIP5 and CMIP6 Multi-Model Ensemble for Precipitation Downscaling Results and Observational Data: The Case of Hanjiang River Basin

Evaluating global climate model (GCM) outputs is essential for accurately simulating future hydrological cycles using hydrological models. The GCM multi-model ensemble (MME) precipitation simulations of the Climate Model Intercomparison Project Phases 5 and 6 (CMIP5 and CMIP6, respectively) were spatially and temporally downscaled according to a multi-site statistical downscaling method for the Hanjiang River Basin (HRB), China. Downscaled precipitation accuracy was assessed using data collected from 14 meteorological stations in the HRB. The spatial performances, temporal performances, and seasonal variations of the downscaled CMIP5-MME and CMIP6-MME were evaluated and compared with observed data from 1970–2005. We found that the multi-site downscaling method accurately downscaled the CMIP5-MME and CMIP6-MME precipitation simulations. The downscaled precipitation of CMIP5-MME and CMIP6-MME captured the spatial pattern, temporal pattern, and seasonal variations; however, precipitation was slightly overestimated in the western and central HRB and precipitation was underestimated in the eastern HRB. The precipitation simulation ability of the downscaled CMIP6-MME relative to the downscaled CMIP5-MME improved because of reduced biases. The downscaled CMIP6-MME better simulated precipitation for most stations compared to the downscaled CMIP5-MME in all seasons except for summer. Both the downscaled CMIP5-MME and CMIP6-MME exhibit poor performance in simulating rainy days in the HRB.

On the suitability of deep convolutional neural networks for continental-wide downscaling of climate change projections

In a recent paper, Baño-Medina et al. (Configuration and Intercomparison of deep learning neural models for statistical downscaling. preprint, 2019) assessed the suitability of deep convolutional neural networks (CNNs) for downscaling of temperature and precipitation over Europe using large-scale ‘perfect’ reanalysis predictors. They compared the results provided by CNNs with those obtained from a set of standard methods which have been traditionally used for downscaling purposes (linear and generalized linear models), concluding that CNNs are well suited for continental-wide applications. That analysis is extended here by assessing the suitability of CNNs for downscaling future climate change projections using Global Climate Model (GCM) outputs as predictors. This is particularly relevant for this type of “black-box” models, whose results cannot be easily explained based on physical reasons and could potentially lead to implausible downscaled projections due to uncontrolled extrapolation artifacts. Based on this premise, we analyze in this work the two key assumptions that are made in perfect prognosis downscaling: (1) the predictors chosen to build the statistical model should be well reproduced by GCMs and (2) the statistical model should be able to reliably extrapolate out of sample (climate change) conditions. As a first step to test the suitability of these models, the latter assumption is assessed here by analyzing how the CNNs affect the raw GCM climate change signal (defined as the difference, or delta, between future and historical climate). Our results show that, as compared to well-established generalized linear models (GLMs), CNNs yield smaller departures from the raw GCM outputs for the end of century, resulting in more plausible downscaling results for climate change applications. Moreover, as a consequence of the automatic treatment of spatial features, CNNs are also found to provide more spatially homogeneous downscaled patterns than GLMs.

An Uncertainty-Based Regional Comparative Analysis on the Performance of Different Bias Correction Methods in Statistical Downscaling of Precipitation

Statistical downscaling of General Circulation Models (GCM) simulations is widely used for projecting precipitation at different spatiotemporal scales. However, the downscaling process is linked with different source of uncertainty including structural/parametric uncertainty of the model and output uncertainty. This research proposes a novel framework to assess the parametric uncertainty of downscaling model, and used this framework to assess the performance of different bias correction methods linked to the regression-based statistical downscaling model. The used downscaling framework in the current paper is Statistical Downscaling Model (SDSM). The conventional bias correction method linked with SDSM is the Variance InFlation method (VIF), this paper substitutes this method with three different bias correction methods including Local Intensity Scaling (LOCI), Power Transformation (PT), and Quantile Mapping (QM) to assess the associated parametric and global uncertainty of each method in different climate by using a new approach. The proposed method is applied to six different stations located in Iran and United States with different climate status. Average Relative Interval Length (ARIL), P-level, and Normalized Uncertainty Efficiency (NUE) are used as uncertainty indicators to evaluate the results. Results represent that in every assessed climate class, LOCI, and PT, work better than conventional VIF in both amount and occurrence modules of SDSM framework. More precisely, LOCI works better in station that has wet summer, while PT performs well in the stations where there is no or very limited precipitation in summer. Substituting LOCI with VIF, result in increasing the value of NUE by at least factor of 3 in occurrence and amount model which means the significant reduction in structural uncertainty. Also applying PT in arid regions improves the NUE indicator at least by factor 2 in occurrence and amount model and by factor 3 in output uncertainty assessment, and results in less parametric and output uncertainty. Results illustrate the important role of bias correction approaches in reducing structural, and output uncertainty and improving the statistical efficiency of the downscaling model.

Intercomparison of downscaling methods for daily precipitation with emphasis on wavelet-based hybrid models

Downscaling of local daily precipitation from large-scale climatic variables is required for assessing the impact of climate change on hydrology and water resources. This study proposes wavelet transform (WT)-based Feed-Forward Neural Network (FF-NN) and Nonlinear Auto Regressive with exogenous inputs Network (NARX-NN) models for downscaling daily precipitation. The models are applied to a large river basin, the Krishna River basin, in the Indian subcontinent. Several climatic variables, including geo-potential heights, wind direction, vorticity, humidity, air temperature, mean sea level pressure, meridional velocity at surface, and 500hpa and 850hpa levels, are considered based on their statistical correlations. The results are evaluated using different performance measures and the ability of the models to capture the extreme events at five selected grid points (in different locations) having varying climatic characteristics is assessed. The performance of the proposed wavelet-based models is also compared with that of four different traditional and recent downscaling methods: Multiple Linear Regression (MLR), Statistical Downscaling Model (SDSM), Genetic Programming (GP), and Artificial Neural Networks (ANNs). The results reveal that the wavelet-based neural network models (WT-FF-NN and WT-NARX-NN) are robust compared to the other methods in terms of their ability to capture the regional precipitation patterns and the extreme events. The improvement in the wavelet-based models can be attributed to their ability to unravel the hidden relationship between the predictors and precipitation. It is also observed that there is considerable increase in the correlation between precipitation and the decomposed climatic variables. All these results suggest that wavelets aid in unravelling the relationship between local precipitation and large-scale climatic variables and improving the overall performance of the downscaling models.

A temporal downscaling approach for sub-daily gridded extreme rainfall intensity estimation under climate change

Study regionCanada. Study focusIntensity-duration-frequency curves need to be updated to account for the potential effect of climate change. However, sub-daily regional climate model (RCM) simulations required for this purpose are commonly unavailable. The scaling approach is often applied for temporal downscaling of point daily precipitation; nevertheless, difference in spatial scale between simulations and observations may complicate its application at the grid scale. This study proposes a grid-scale temporal downscaling approach for estimating sub-daily gridded extreme rainfall intensity under climate change at regions with a relatively high station density. The approach uses daily annual maximum rainfall intensity series from RCM simulations and grid-scale scaling properties from observations. The study evaluates two methods for estimating gridded scaling exponents needed in temporal downscaling. The assessment is performed on observation-based and RCM-based temporally downscaled quantiles. The effect of temporal downscaling and bias correction on sub-daily quantiles and relative changes is also analysed. The approach is applied to three case studies across Canada. New hydrological insights for the regionPiecewise simple scaling is found over the study domains, indicating caution in using simple scaling. The performance of the approach in estimating sub-daily observation-based gridded rainfall intensity quantiles is suitable, with preference for point to grid transfer at the annual maximum series stage. Bias correction is necessary for properly estimating sub-daily RCM-based quantiles, presenting less impact on relative changes.

High-resolution dynamical downscaling of ERA-Interim temperature and precipitation using WRF model for Greece

This study presents the results of high-resolution dynamical downscaling of 5 km on maximum (TX) and minimum (TN) air temperature and precipitation, for Greece, with the Weather Research and Forecasting (WRF) model. The ERA-Interim (ERA-I) reanalysis dataset is used for initial and boundary conditions. The model results (WRF_5) are evaluated against available ground observations for the period 1980–2004 through the calculation of mean climatology, statistical metrics, and distributions of extreme events on daily, monthly and seasonal scales. WRF_5 model captures very well the geographical distribution of TX and TN of the study area, and illustrates finely the seasonal differences. Statistical results for TX (TN) indicate a cold (warm) bias of − 0.6 °C (1 °C) regarding WRF_5 and − 3 °C (0.5 °C) for ERA-I. The efficiency metrics for temperatures showed a highly improved performance of the model compared to reanalysis for all temporal scales investigated. The observed mean annual cycle and inter-annual variability of precipitation are also well represented by model simulation. Although WRF_5 overestimates rainfall during most of the year, the seasonal pattern of WRF_5 presented similar correlation coefficients for all stations with a range of 0.6–0.85, showing a good model ability to simulate the precipitation in Greece. The results reveal the capability of the configured WRF high resolution model to reproduce the main climatological variables of the study area, outperforming the coarse resolution ERA-Interim in a region that is dominated by highly variable topographic characteristics. This is deemed necessary for undertaking any further studies concerning future climate change impacts in various sectors.

Challenges and potential solutions in statistical downscaling of precipitation

Downscaling is an effective technique to bridge the gap between climate model outputs and data requirements of most crop and hydrologic models for assessing local and site-specific climate change impacts, especially on future food security. However, downscaling of temporal sequences, extremes in daily precipitation, and handling of nonstationary precipitation in future conditions are considered common challenges for most statistical downscaling methods. This study reviewed the above key challenges in statistical downscaling and proposed potential solutions. Ten weather stations located across the globe were used as proof of concept. The use of a stochastic Markov chain to generate daily precipitation occurrences is an effective approach to simulate the temporal sequence of precipitation. Also, the downscaling of precipitation extremes can be achieved by adjusting the skewness coefficient of a probability distribution, as they are highly correlated. Nonstationarity in precipitation downscaling can be handled by adjusting parameters of a probability distribution according to future precipitation change signals projected by climate models. The perspectives proposed in this study are of great significance in using climate model outputs for assessing local and site-specific climate change impacts, especially on future food security.

Deep Learning for Daily Precipitation and Temperature Downscaling

AbstractDownscaling is a critical step to bridge the gap between large‐scale climate information and local‐scale impact assessment. This study presents a novel deep learning approach: Super Resolution Deep Residual Network (SRDRN) for downscaling daily precipitation and temperature. This approach was constructed based on an advanced deep convolutional neural network with residual blocks and batch normalizations. The data augmentation technique was utilized to address overfitting that is due to highly imbalanced precipitation and nonprecipitation days and sparse precipitation extremes. Synthetic experiments were designed to downscale daily maximum/minimum temperature and precipitation data from coarse resolutions (25, 50, and 100 km) to a high resolution (4 km). The results showed that, during the validation period, the SRDRN approach not only captured the spatial and temporal patterns remarkably well, but also reproduced both precipitation and temperature extremes in different locations and time at the local scale. Through transfer learning, the trained SRDRN model in one region was directly applied to downscale precipitation in another region with a different environment, and the results showed notable improvement compared to classic statistical downscaling methods. The outstanding performance of the SRDRN approach stemmed from its ability to fully extract spatial features without suffering from degradation and overfitting issues due to the incorporations of residual blocks, batch normalizations, and data augmentations. The SRDRN approach is thus a powerful tool for downscaling daily precipitation and temperature and can potentially be leveraged to downscale any hydrologic, climate, and earth system data.

Statistical downscaling (Delta method) of precipitation and temperature for Bilate watershed, Ethiopia

It is important to investigate potential changes in temperature and precipitation for assessing the impacts of future climate change on agricultural production for specific regions. In this study, climate scenarios of precipitation and temperature for the bilate watershed were generated. Statistical downscaling techniques of the Delta method analysis protocol of the Agricultural Intercomparison and Improvement Project (AgMIP) was used to project the future climate state in the farm lands of Bilate Watershed. In this study, climate change scenarios were generated for two Representative Concentration Pathways (RCPs): RCP 4.5 and RCP 8.5 using 20 GCMs from CMIP5 bias-corrected under three future time slices; near-term (2010-2039), mid-century (2040-2069) and end-century (2071-2099). Rainfall is projected to increase in total amount under all-time slices and emissions pathways but with pronounced inter and intra-variability. Minimum temperature will significantly increase during mid-century by 1.81°C (RCP 4.5) and 2.55°C (RCP 8.5) and by 2.1°C (RCP 4.5) and 4.27°C (RCP8.5) during end-century relative to the baseline. The projected increase in maximum temperature during mid-century is 1.43°C under RCP 4.5 and 1.99°C under RCP 8.5 and during end-century by 1.65°C under RCP 4.5 and 3.5°C under RCP8.5 during end-century. Key words: Agricultural Intercomparison and Improvement Project (AgMIP), climate change, Delta method, Global climate models (GCMs), statistical downscaling, watershed.

Impact of Automated Statistical Downscaling and Delta Downscaling methods on projecting future climate change in the northeast Tibetan Plateau

The accuracy of different downscaling methods in projecting future precipitation and air temperature from general circulation models (GCMs) has rarely been addressed with regards to the Tibetan Plateau, and this information is important for future water resource management in the region. The performance of automated statistical downscaling (ASD) and Delta downscaling methods in predicting precipitation and air temperature was evaluated at 19 meteorological stations in the Qilian Mountains and Hexi Corridor (QM-HC) by comparing with in situ observations from 2006-2015. These comparisons, based on Representative Concentration Pathway 4.5 (RCP4.5), suggest that the difference in annual precipitation between the ASD model and the Delta method is 17 mm. Testing different weights of the 2 downscaling methods indicates that combining the 2 methods results in lower uncertainty. The downscaling of annual precipitation projected by weighting the results of the 2 methods suggested that, based on RCP4.5, precipitation will not increase significantly from 2021-2100 compared to the past (1961-2005) and will fluctuate steadily in the coming decades. These projections are in contrast with previous projections of a significant increase. Air temperature is projected to increase by approximately 0.2°C decade-1 from 2021-2100 according to the weighted average of the ASD model and Delta method based on RCP4.5. This study indicates that management measures based on projected increased precipitation should be carefully reconsidered in different regions.

Deep Precipitation Downscaling

Precipitation downscaling, which is similar to the mechanism of single-image super-resolution (SR), aims to improve the spatial resolution of rain maps. It is of great practical value and theoretical significance. This letter presents a new deep precipitation downscaling (DPD) method, named auxiliary guided spatial distortion (AGSD) network, motivated by SR techniques. Specifically, an auxiliary guided module (AGM), which takes multiple meteorological elements (e.g., temperature, relative humidity, and wind) as input, is proposed for getting more accurate rain map features. Meanwhile, a simple but effective spatial distortion module (SDM) is proposed. Benefitting from SDM, the DPD model can rectify the rain map via terrain correlation. Furthermore, to improve the model performance among various rain intensity (including small rain, moderate rain, heavy rain, and storm), a threat score-driven pseudo threat-score (PTS) loss is presented. Experimental results compared with state-of-the-art methods demonstrate the superiority of the proposed method.

An Efficient Downscaling Scheme for High-Resolution Precipitation Estimates over a High Mountainous Watershed

Satellites are capable of observing precipitation over large areas and are particularly suitable for estimating precipitation in high mountains and poorly gauged regions. However, the coarse resolution and relatively low accuracy of satellites limit their applications. In this study, a downscaling scheme was developed to obtain precipitation estimates with high resolution and high accuracy in the Heihe watershed. Shannon’s entropy, together with a semi-variogram, was applied to establish the optimal precipitation station network. A combination of the random forest (RF) method and the residual correction approach with the established rain gauge network was applied to downscale monthly precipitation products from Integrated Multi-satellitE Retrievals for Global Precipitation Measurement (IMERG). The results indicated that the RF model showed little improvement in the accuracy of IMERG-based precipitation downscaling. Including residual modification could improve the results of the RF model. The mean absolute error (MAE) and root mean square error (RMSE) values decreased by 19% and 21%, respectively, after residual corrections were added to the RF approach. Moreover, we found that enough rain gauge records are necessary for and remain an important component of tuning model performance. The application of more rain gauges improves the performance of the combined RF and residual modification methods, with the MAE and RMSE values reduced by 8% and 9%, respectively. Residual correction, together with enough precipitation stations, can effectively enhance the quality of the precipitation patterns and magnitudes obtained in the RF downscaling process. The proposed downscaling scheme is an effective tool for increasing the accuracy and spatial resolution of precipitation fields in the Heihe watershed.