Global Historical Climatology Network Research Articles

Multi-Source Dataset Assessment and Variation Characteristics of Snow Depth in Eurasia from 1980 to 2018

Snow is an indicator of climate change. Its variation can affect surface energy, water balance, and atmospheric circulation, providing important feedback on climate change. There is a lack of assessment of the spatial characteristics of multi-source snow data in Eurasia, and these data exhibit high spatial variability and other differences. Therefore, using data obtained from the Global Historical Climatology Network Daily (GHCND) from 1980 to 2018, snow depth information from ERA5, MERRA2, and GlobSnow is assessed in this study. The spatiotemporal variation characteristics and the primary spatial modes of seasonal variations in snow depth are analyzed. The results show that the snow depth, according to GlobSnow data, is closer to that of the measured site data, while the ERA5_Land and MERRA2 data are overestimated. The annual variations in snow depth are consistent with seasonal variations in winter and spring, with an increasing trend in the mountains of Central Asia and Siberia and a decreasing trend in most of the rest of Eurasia. The dominant patterns of snow depth in late autumn, winter, and spring are all north–south dipole patterns, and there is overall consistency in summer.

An Improved Ensemble of Land Surface Air Temperatures Since 1880 Using Revised Pair-Wise Homogenization Algorithms Accounting for Autocorrelation

Abstract Land surface air temperatures (LSAT) inferred from weather station data differ among major research groups. The estimate by NOAA’s monthly Global Historical Climatology Network (GHCNm) averages 0.02°C cooler between 1880 and 1940 than Berkeley Earth’s and 0.14°C cooler than the Climate Research Unit estimates. Such systematic offsets can arise from differences in how poorly documented changes in measurement characteristics are detected and adjusted. Building upon an existing pairwise homogenization algorithm used in generating the fourth version of NOAA’s GHCNm(V4), PHA0, we propose two revisions to account for autocorrelation in climate variables. One version, PHA1, makes minimal modification to PHA0 by extending the threshold used in breakpoint detection to be a function of LSAT autocorrelation. The other version, PHA2, uses penalized likelihood to detect breakpoints through optimizing a model-selection problem globally. To facilitate efficient optimization for series with more than 1000 time steps, a multiparent genetic algorithm is proposed for PHA2. Tests on synthetic data generated by adding breakpoints to CMIP6 simulations and realizations from a Gaussian process indicate that PHA1 and PHA2 both similarly outperform PHA0 in recovering accurate climatic trends. Applied to unhomogenized GHCNmV4, both revised algorithms detect breakpoints that correspond with available station metadata. Uncertainties are estimated by perturbing algorithmic parameters, and an ensemble is constructed by pooling 50 PHA1- and 50 PHA2-based members. The continental-mean warming in this new ensemble is consistent with that of Berkeley Earth, despite using different homogenization approaches. Relative to unhomogenized data, our homogenization increases the 1880–2022 trend by 0.16 [0.12, 0.19]°C century−1 (95% confidence interval), leading to continental-mean warming of 1.65 [1.62, 1.69]°C over 2010–22 relative to 1880–1900. Significance Statement Accurately correcting for systematic errors in observational records of land surface air temperature (LSAT) is critical for quantifying historical warming. Existing LSAT estimates are subject to systematic offsets associated with processes including changes in instrumentation and station movement. This study improves a pairwise homogenization algorithm by accounting for the fact that climate signals are correlated over time. The revised algorithms outperform the original in identifying discontinuities and recovering accurate warming trends. Applied to monthly station temperatures, the revised algorithms adjust trends in continental mean LSAT since the 1880s to be 0.16°C century−1 greater relative to raw data. Our estimate is most consistent with that from Berkeley Earth and indicates lesser and greater warming than estimates from NOAA and the Met Office, respectively.

Evaluation of Daily Temperature Extremes in the ECMWF Operational Weather Forecasts and ERA5 Reanalysis

In weather forecasting and climate monitoring, daily maximum and minimum air temperatures (TMAX and TMIN) are fundamental for operational and research purposes, from early warning of extreme events to climate change studies. This study provides an integrated evaluation of TMAX and TMIN from two European Centre for Medium-range Weather Forecasts (ECMWF) products: ERA5 reanalysis (1980–2019) and operational weather forecasts (2017–2021). Both products are evaluated using in situ observations from the Global Historical Climatology Network (GHCN). While the analyses span globally, emphasis is given to four key regions: Europe, East and West United States, and Australia. Results reveal a general underestimation of TMAX and overestimation of TMIN in both operational forecasts and ERA5, highlighting the limitation of the ECMWF model in estimating the amplitude of the diurnal cycle of air temperature. ERA5′s accuracy has improved over the past decade, due to enhanced constrain of land–atmosphere analysis streaming from more and higher-quality satellite data. Furthermore, ERA5 outperforms one-day-ahead weather forecasts, indicating that non-real-time dependent studies should rely on ERA5 instead of real-time operational forecasts. This study underscores the importance of ongoing research in model and data assimilation, considering the relevance of daily temperature extremes forecasting and reanalysis for operational meteorology and climate monitoring.

Modelled Climate Change Impacts on Spring Canola Production Across British Columbia, Canada

This paper investigates the impact of projected near future climate trends on the production of spring planted canola (Brassica napus, Brassica rapa, and Brassica juncea of canola classification) across British Columbia, Canada. Analysis of historic climate trends from 2001-2020, informed by the Global Historical Climatology Network daily (GHCNd) database, establishes patterns of warming temperatures and increasing precipitation values across the province over the early 21st century. Near future climate trends were modelled using CMIP6 climate models, from 2021-2040 under SSP1-2.6, SSP2-4.5 and SSP5-8.5, with the projections downscaled using ClimateBC. The projected trends mirrored those of the observed historic record, while an observable relationship between rising levels of climate change and increasing projected annual precipitation and temperature is recorded. The subsequent crop modelling using the CSM-CROPGRO-Canola and DNDC models, fed with the modelled climate trends, highlighted the expectation for near future climate change to cause significant decreases in spring canola production across British Columbia.

Employing higher density lower reliability weather data from the Global Historical Climatology Network monitors to generate serially complete weather data for watershed modelling

AbstractHydrological models require complete and accurate weather data time series to represent watershed‐scale responses adequately. The Global Historical Climatology Network (GHCN) is the most comprehensive weather database used in hydrological modelling studies globally. Since higher‐density, lower‐reliability precipitation measurements from private citizens collected by the Community Collaborative Rain, Hail, and Snow (CoCoRaHS) network data were integrated into the GHCN, hydrological modellers in the United States have access to a much greater amount of weather data. However, the benefit of using CoCoRaHS data has not been assessed. The objectives of this work were to develop a method for generating a complete weather data time series based on the combination of data from multiple GHCN monitors and to assess several methods for the estimation of missing weather data. Weather data from GHCN monitors located within a specific radius of a watershed were obtained and interpolated using three estimation methods (Inverse Distance Weighting (IDW), Inverse Distance and Elevation Weighting (IDEW) and Closest Station), creating a seamless time series of weather observations. To evaluate the performance of the methodologies, weather data obtained from each estimation method was used to force the Soil and Water Assessment Tool (SWAT) and Thornthwaite‐Mather models for 21 US Department of Agriculture‐Conservation Effects Assessment Project watersheds in different climate regions to simulate daily streamflow for 2010–2021. Except for three watersheds, all of the SWAT models had Nash‐Sutcliffe Efficiency above 0.5, the ratio of the root mean square error to the standard deviation of observations below 0.7, and percent bias from −25% to 25% with a satisfactory performance rating. IDEW and IDW performed similarly, and the Closest Station method resulted in the poorest streamflow simulation. A comparison with published SWAT model results further corroborated improved model performance using novel combined GHCN data with all Closest Station, IDW and IDEW methods.

Climatological trends of mean and extreme daily precipitation in Arizona (USA)

Precipitation in Arizona, USA, represents a fundamental resource for the needs of agriculture, people, and the environment. However, due to the risk associated with flash flooding, extreme precipitation also constitutes a natural hazard. Available research on precipitation in Arizona has been limited in scope and focus and there currently exists no comprehensive assessment of statewide, historical, mean, and extreme precipitation. We use daily precipitation records from 43 Global Historical Climate Network daily (GHCNd) weather stations to examine trends in mean and extreme precipitation across Arizona from 1950 to 2020. We use a suite of standardized precipitation indices and explore the statistical significance of historical changes at the annual, monthly, and seasonal scales. Our analysis returns a motley collection of results displaying a great degree of spatial variability and sensitivity to the temporal scale of analysis. Mean total precipitation at the statewide scale underwent a small decreasing trend of 0.15 mm year−1, with considerable interannual variability being a dominant feature. The majority of stations experienced no statistically significant change in extreme precipitation at the annual scale. Monsoon season mean precipitation followed a similar pattern to annual means, with an overall reduction of 0.14 mm year−1 across the state. One fourth of stations, all located at elevations lower than 650 m, recorded decreases in monsoon precipitation intensity. Winter season analysis presents a different picture, displaying a positive statewide trend of 0.11 mm year−1 in mean precipitation. One tenth of stations recorded increases in extreme precipitation intensity and decreases in the number of consecutive dry days. Results of our observational analysis indicate a lack of a clear consensus on the climatological trends of mean/extreme precipitation in Arizona during the study period. Our findings are in contrast to those from other similarly arid areas across the world and highlight the role of regional differences in modulating the potential hydrometeorological impacts associated with climate change. However, while we do not find widespread, significant, modification in the spectrum of precipitation changes examined, the hydrologic system of the state has been – and will continue to be - impacted by the ongoing temperature increase associated with the buildup of greenhouse gases and continued population growth.

Ensemble Learning for Blending Gridded Satellite and Gauge-Measured Precipitation Data

Regression algorithms are regularly used for improving the accuracy of satellite precipitation products. In this context, satellite precipitation and topography data are the predictor variables, and gauged-measured precipitation data are the dependent variables. Alongside this, it is increasingly recognised in many fields that combinations of algorithms through ensemble learning can lead to substantial predictive performance improvements. Still, a sufficient number of ensemble learners for improving the accuracy of satellite precipitation products and their large-scale comparison are currently missing from the literature. In this study, we work towards filling in this specific gap by proposing 11 new ensemble learners in the field and by extensively comparing them. We apply the ensemble learners to monthly data from the PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) and IMERG (Integrated Multi-satellitE Retrievals for GPM) gridded datasets that span over a 15-year period and over the entire contiguous United States (CONUS). We also use gauge-measured precipitation data from the Global Historical Climatology Network monthly database, version 2 (GHCNm). The ensemble learners combine the predictions of six machine learning regression algorithms (base learners), namely the multivariate adaptive regression splines (MARS), multivariate adaptive polynomial splines (poly-MARS), random forests (RF), gradient boosting machines (GBM), extreme gradient boosting (XGBoost) and Bayesian regularized neural networks (BRNN), and each of them is based on a different combiner. The combiners include the equal-weight combiner, the median combiner, two best learners and seven variants of a sophisticated stacking method. The latter stacks a regression algorithm on top of the base learners to combine their independent predictions. Its seven variants are defined by seven different regression algorithms, specifically the linear regression (LR) algorithm and the six algorithms also used as base learners. The results suggest that sophisticated stacking performs significantly better than the base learners, especially when applied using the LR algorithm. It also beats the simpler combination methods.

Fusing Satellite Precipitation Products Based on Top–Down and Bottom–Up Approaches and an Improved Double Instrumental Variable Method for the Chuanyu Region, China, from 2007 to 2019

Precipitation is one of the crucial variables in the hydrological and ecological cycles. High-quality precipitation data are of great importance for climate change, water resource management, and agricultural research over complex terrains. Recently, satellite precipitation products have been produced based on different retrieval algorithms, mainly the top–down and bottom–up approaches. Fusing satellite precipitation products based on these two different approaches may combine their advantages and achieve a better data quality for describing precipitation. In this paper, an improved double instrumental variable (IMDIV) method is proposed for data quality enhancement by merging IMERG (integrated multi-satellite retrievals for global precipitation measurement), which is based on the top–down approach, and SM2RAIN (soil moisture to rain), which is based on the bottom–up approach. In detail, IMERG-Early (IMERG early run) and IMERG-Final (IMERG final run) are merging with SM2RAIN at a daily scale, respectively. Rain gauge station records from GHCNd (Global Historical Climatology Network Daily) are used to evaluate the original and fused precipitation products for the Chuanyu region, China, from 2007 to 2019. The results show that the proposed IMDIV method outperforms the original DIV method on precipitation fusion tasks. Moreover, the proposed IMDIV-EAS (fusing IMERG-Early and SM2RAIN) and IMDIV-FIS (fusing IMERG-Final and SM2RAIN) products both outperform the original precipitation products IMERG and SM2RAIN, with higher correlation coefficients (R) of 0.603 and 0.634; better RMSEs of 5.136 and 5.088 mm/day; and better biases of 0.514 and 0.509 mm/day. The results demonstrate the effectiveness of the proposed method and the high quality of the fused products, which could be useful for hydrology and climate studies.

Evidence of Urban Blending in Homogenized Temperature Records in Japan and in the United States: Implications for the Reliability of Global Land Surface Air Temperature Data

Abstract To reduce the amount of nonclimatic biases of air temperature in each weather station’s record by comparing it with neighboring stations, global land surface air temperature datasets are routinely adjusted using statistical homogenization to minimize such biases. However, homogenization can unintentionally introduce new nonclimatic biases due to an often-overlooked statistical problem known as “urban blending” or “aliasing of trend biases.” This issue arises when the homogenization process inadvertently mixes urbanization biases of neighboring stations into the adjustments applied to each station record. As a result, urbanization biases of the original unhomogenized temperature records are spread throughout the homogenized data. To evaluate the extent of this phenomenon, the homogenized temperature data for two countries (Japan and the United States) are analyzed. Using the Japanese stations in the widely used Global Historical Climatology Network (GHCN) dataset, it is first confirmed that the unhomogenized Japanese temperature data are strongly affected by urbanization bias (possibly ∼60% of the long-term warming). The U.S. Historical Climatology Network (USHCN) dataset contains a relatively large amount of long, rural station records and therefore is less affected by urbanization bias. Nonetheless, even for this relatively rural dataset, urbanization bias could account for ∼20% of the long-term warming. It is then shown that urban blending is a major problem for the homogenized data for both countries. The IPCC’s estimate of urbanization bias in the global temperature data based on homogenized temperature records may have been low as a result of urban blending. Recommendations on how future homogenization efforts could be modified to reduce urban blending are discussed. Significance Statement Most weather station–based global land temperature datasets currently use a process called “statistical homogenization” to reduce the amount of nonclimatic biases. However, using temperature data from two countries (Japan and the United States), we show that the homogenization process unintentionally introduces new nonclimatic biases into the data as a result of an “urban blending” problem. Urban blending arises when the homogenization process inadvertently mixes the urbanization (warming) bias of the neighboring stations into the adjustments applied to each station record. As a result, the urbanization biases of the unhomogenized temperature records are spread throughout all of the homogenized data. The net effect tends to artificially add warming to rural stations and subtract warming from urban stations until all stations have about the same amount of urbanization bias.

Observed Changes in Interannual Precipitation Variability in the United States

AbstractCharacterizing changes in precipitation patterns over time is critical for hydrologically dependent fields like water resource management and agriculture. Here, we explore observed trends in interannual precipitation variability using a suite of metrics that describe changes in precipitation over time. We analyze daily in situ Global Historical Climatology Network precipitation data from 1970 to present over 17 internally consistent sub‐national United States domains using a regional Mann‐Kendall trend test. We find robustly increasing trends in annual mean precipitation and wet day frequency for most of the central and eastern U.S., but decreasing trends in the western U.S. Importantly, we identify widespread significant trends in interannual precipitation variability, with increasing variability in the southeast, decreasing variability in the far west, and mixed signals in the Rocky Mountains and north‐central U.S. Our results provide important context for water resource managers and a new observational standard for climate model performance assessments.

CARDIOVASCULAR DISEASES IN UKRAINE UNDER CONDITIONS OF GLOBAL WARMING

Рurpose: to establish the relationship between cardiovascular diseases in Ukraine and an increase in the average annual temperature.
 Materials and Мethods. The data of the State Statistics Service of Ukraine on the number of cases of cardiovascular diseases from 1990 to 2013, as well as on the share of the urban and rural population in various regions for 1990, 2000 and 2013 were analyzed. Statistical data starting from 2014 were not used due to their absence from the temporarily occupied territories of Ukraine. Average annual temperatures were obtained from the global database – The Global Historical Climatology Network. With the help of the National reports on the state of the natural environment in Ukraine, atmospheric air pollution in the regions of Ukraine was assessed. Statistical processing of the obtained results was carried out using correlation and regression analysis with determination of the reliability of the approximation.
 Results. A direct average correlation between average annual temperatures and the number of newly registered cases of cardiovascular diseases in Ukraine between 1990 and 2013 was established (correlation coefficient +0.5). After carrying out an analysis between the regions of Ukraine, it was established that in the regions with direct strong and medium correlations between the number of newly detected diseases of the circulatory system and average annual air temperatures (Dnipropetrovsk, Zaporizhzhya, Luhansk, Kharkiv, Donetsk, Autonomous Republic of Crimea, Kyiv, Sumy, Kherson, Odesa, Mykolaiv, Kirovohrad, Lviv, Zhytomyr regions) the urban population prevails, as well as a higher air pollution index. In other regions of Ukraine, where the rural population predominates, a weak straight line is established, as well as inverse correlation dependences and a lower level of atmospheric air pollution.
 Conclusions. In recent decades, the average annual temperature of the environment has been increasing, which has a negative effect on human health, in particular on the cardiovascular system, causing its damage. Therefore, it is important to start acting now to avoid the negative consequences of global warming.
 
 

Meteorological Driving Datasets for the U.S. Midwest and Great Lakes Region Incorporating Precipitation Gauge Undercatch Corrections

Abstract Precipitation (P) gauge undercatch (PUC) is an important source of error when using observed meteorological datasets for hydrologic modeling studies in regions with cold and windy winters. Preliminary simulations using the Variable Infiltration Capacity (VIC) hydrological model forced with different meteorological datasets showed significant underprediction of simulated streamflow throughout the domain. A new hybrid gridded meteorological dataset at 1/16° resolution based on observed station data was assembled over the U.S. Midwest and Great Lakes region from 1915 to 2021 at a daily time step. Correction of primary station data using existing techniques is generally difficult or infeasible in the United States due to missing station metadata and lack of local wind speed (WS) measurements. We developed and tested several different postprocessing adjustment techniques using regridded WS obtained from the NCEP–NCAR reanalysis. The most effective approach corrected rain or mixed P using WS alone, and P as snow using a regressed snow-to-P ratio from a group of wind-shielded reference stations (to account for different and generally unknown snow measurement techniques). The PUC-corrected gridded products were validated against high-quality shielded stations and corrected Global Historical Climatology Network stations with in situ WS, showing good overall agreement. Observed monthly streamflow at 40 river basins was also compared to hydrologic model simulations forced by datasets with and without PUC corrections. The best PUC-corrected dataset produced improvements in streamflow simulations in at least 80% of the streamflow locations for three validation metrics (r2, Nash–Sutcliff efficiency, bias in the mean), demonstrating its value for hydrometeorological studies in the greater Midwest region. Significance Statement Many applications in hydrology require in situ precipitation (P) measurements, which are known to have a systematic low bias due to the effects of wind, also known as precipitation undercatch (PUC). Addressing PUC is problematic in the United States due to limited access to detailed station metadata (SMD) and local wind speed (WS) measurements. In this paper we develop a set of procedures to create gridded precipitation datasets for the U.S. Midwest region that incorporate corrections for PUC without needing either (i) detailed SMD or (ii) local WS measurements. Among other tests, results in 40 test basins throughout the Midwest show substantial improvements in simulated streamflow in 32 out of 40 basins when PUC corrections are included in meteorological driving datasets.

Newly reconstructed Arctic surface air temperatures for 1979–2021 with deep learning method

A precise Arctic surface air temperature (SAT) dataset, that is regularly updated, has more complete spatial and temporal coverage, and is based on instrumental observations, is critically important for timely monitoring and improving understanding of the rapid change in the Arctic climate. In this study, a new monthly gridded Arctic SAT dataset dated back to 1979 was reconstructed with a deep learning method by combining surface air temperatures from multiple data sources. The source data include the observations from land station of GHCN (Global Historical Climatology Network), ICOADS (International Comprehensive Ocean-Atmosphere Data Set) over the oceans, drifting ice station of Russian NP (North Pole), and buoys of IABP (International Arctic Buoy Programme). The last two are crucial for improving the representation of the in-situ observed temperatures within the Arctic. The newly reconstructed dataset includes monthly Arctic SAT beginning in 1979 and daily Arctic SAT beginning in 2011. This dataset would represent a new improvement in developing observational temperature datasets and can be used for a variety of applications.

Agreement in extreme precipitation exposure assessment is modified by race and social vulnerability.

Epidemiologic investigations of extreme precipitation events (EPEs) often rely on observations from the nearest weather station to represent individuals' exposures, and due to structural factors that determine the siting of weather stations, levels of measurement error and misclassification bias may differ by race, class, and other measures of social vulnerability. Gridded climate datasets provide higher spatial resolution that may improve measurement error and misclassification bias. However, similarities in the ability to identify EPEs among these types of datasets have not been explored. In this study, we characterize the overall and temporal patterns of agreement among three commonly used meteorological data sources in their identification of EPEs in all census tracts and counties in the conterminous United States over the 1991-2020 U.S. Climate Normals period and evaluate the association between sociodemographic characteristics with agreement in EPE identification. Daily precipitation measurements from weather stations in the Global Historical Climatology Network (GHCN) and gridded precipitation estimates from the Parameter-elevation Relationships on Independent Slopes Model (PRISM) and the North American Land Data Assimilation System (NLDAS) were compared in their ability to identify EPEs defined as the top 1% of precipitation events or daily precipitation >1 inch. Agreement among these datasets is fair to moderate from 1991 to 2020. There are spatial and temporal differences in the levels of agreement between ground stations and gridded climate datasets in their detection of EPEs in the United States from 1991 to 2020. Spatial variation in agreement is most strongly related to a location's proximity to the nearest ground station, with areas furthest from a ground station demonstrating the lowest levels of agreement. These areas have lower socioeconomic status, a higher proportion of Native American population, and higher social vulnerability index scores. The addition of ground stations in these areas may increase agreement, and future studies intending to use these or similar data sources should be aware of the limitations, biases, and potential for differential misclassification of exposure to EPEs. Most importantly, vulnerable populations should be engaged to determine their priorities for enhanced surveillance of climate-based threats so that community-identified needs are met by any future improvements in data quality.

Changes in Seasonal Large-Scale Extreme Precipitation in the Mid-Atlantic and Northeast United States, 1979–2019

Abstract This study uses Global Historical Climate Network (GHCN) data in each season to identify the days with the most extreme precipitation (“EP days”) in the mid-Atlantic and northeast United States between 1979 and 2019. These days are sorted according to the fraction of extreme precipitation attributed to tropical cyclone (TC), atmospheric river (AR), and extreme integrated vapor transport (IVT) influences. In winter and spring, there have been increases in seasonal precipitation from the most extreme days, associated with a combination of frequency and intensity changes. These increasing trends come primarily from atmospheric rivers. In summer and fall, there have also been large increases in precipitation on extreme days, in this case due entirely to increased event frequency. These changes come from a combination of AR, TC, and extreme IVT influences. Synoptic characteristics of AR-related EP days in winter and spring have changed significantly. In winter, there has been an amplification of the Atlantic ridge and a deepening of the upstream trough over the upper Great Plains, as well as enhanced AR detection and IVT on these days. The composite low has shifted north and intensified. In spring, the trough has weakened and 1000–500-hPa thickness has increased broadly to the south. These changes are related to changes in the large-scale flow. In winter and spring, the North Atlantic subtropical high (NASH) has strengthened and shifted west, leading to increased southwesterly IVT over the mid-Atlantic and Northeast United States. In summer, southerly IVT along the east coast has increased, and 1000–500-hPa climatological thickness has increased broadly in all seasons. Significance Statement This paper studies the days with the most extreme precipitation over the mid-Atlantic and Northeast United States in each season and finds that there have been large changes in the frequency, intensity, and characteristics of some of these days. Extreme days associated with atmospheric rivers (ARs) and tropical cyclones (TCs) have become more frequent. Precipitation on AR-related days has also become more intense in some areas. On winter extreme days associated with ARs, the pattern has become more amplified and moisture fluxes have become stronger; these changes are likely associated with strengthened high pressure over the Atlantic. Throughout all seasons, there is evidence of warming on extreme days. This is important because it adds to our understanding of how total and AR-related extreme precipitation is changing in a warming climate.

Comparison of Tree-Based Ensemble Algorithms for Merging Satellite and Earth-Observed Precipitation Data at the Daily Time Scale

Merging satellite products and ground-based measurements is often required for obtaining precipitation datasets that simultaneously cover large regions with high density and are more accurate than pure satellite precipitation products. Machine and statistical learning regression algorithms are regularly utilized in this endeavor. At the same time, tree-based ensemble algorithms are adopted in various fields for solving regression problems with high accuracy and low computational costs. Still, information on which tree-based ensemble algorithm to select for correcting satellite precipitation products for the contiguous United States (US) at the daily time scale is missing from the literature. In this study, we worked towards filling this methodological gap by conducting an extensive comparison between three algorithms of the category of interest, specifically between random forests, gradient boosting machines (gbm) and extreme gradient boosting (XGBoost). We used daily data from the PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) and the IMERG (Integrated Multi-satellitE Retrievals for GPM) gridded datasets. We also used earth-observed precipitation data from the Global Historical Climatology Network daily (GHCNd) database. The experiments referred to the entire contiguous US and additionally included the application of the linear regression algorithm for benchmarking purposes. The results suggest that XGBoost is the best-performing tree-based ensemble algorithm among those compared. Indeed, the mean relative improvements that it provided with respect to linear regression (for the case that the latter algorithm was run with the same predictors as XGBoost) are equal to 52.66%, 56.26% and 64.55% (for three different predictor sets), while the respective values are 37.57%, 53.99% and 54.39% for random forests, and 34.72%, 47.99% and 62.61% for gbm. Lastly, the results suggest that IMERG is more useful than PERSIANN in the context investigated.

Comparison of Machine Learning Algorithms for Merging Gridded Satellite and Earth-Observed Precipitation Data

Gridded satellite precipitation datasets are useful in hydrological applications as they cover large regions with high density. However, they are not accurate in the sense that they do not agree with ground-based measurements. An established means for improving their accuracy is to correct them by adopting machine learning algorithms. This correction takes the form of a regression problem, in which the ground-based measurements have the role of the dependent variable and the satellite data are the predictor variables, together with topography factors (e.g., elevation). Most studies of this kind involve a limited number of machine learning algorithms and are conducted for a small region and for a limited time period. Thus, the results obtained through them are of local importance and do not provide more general guidance and best practices. To provide results that are generalizable and to contribute to the delivery of best practices, we here compare eight state-of-the-art machine learning algorithms in correcting satellite precipitation data for the entire contiguous United States and for a 15-year period. We use monthly data from the PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) gridded dataset, together with monthly earth-observed precipitation data from the Global Historical Climatology Network monthly database, version 2 (GHCNm). The results suggest that extreme gradient boosting (XGBoost) and random forests are the most accurate in terms of the squared error scoring function. The remaining algorithms can be ordered as follows, from the best to the worst: Bayesian regularized feed-forward neural networks, multivariate adaptive polynomial splines (poly-MARS), gradient boosting machines (gbm), multivariate adaptive regression splines (MARS), feed-forward neural networks and linear regression.

Scaling of Floods With Geomorphologic Characteristics and Precipitation Variability Across the Conterminous United States

AbstractAccurate flood risk assessment requires a comprehensive understanding of flood sensitivity to regional drivers and climate factors. This paper presents the scaling of floods (duration, peak, volume) with geomorphologic characteristics of the basin (i.e., drainage area, slope, elevation) and precipitation patterns (rainfall accumulation, variability). Long‐term daily streamflow observations over the 20th and early 21st centuries from Hydro‐Climatic Data Network streamgages across the conterminous United States are used to create a flood event database based on their flood stage information. Antecedent daily rainfall accumulation and variability corresponding to these floods are computed using Global Historical Climatology Network daily data set. Two Bayesian scaling models are developed, and the spatial organization of scaling exponents is investigated. The baseline model quantifies the scaling of floods to geomorphologic characteristics. The dynamic model quantifies the scaling of floods to antecedent precipitation distribution which is further conditioned on geomorphologic characteristics. Results show that small and low‐elevation basins have a stronger response to antecedent rainfall distribution in amplifying flood peaks, while high‐elevation steeper basins have a lower response for flood duration and volume. The dynamic models demonstrate that there are significant variations in the flood scaling rates, with the largest rates up to 40% and 4.5% for flood duration, 64% and 44% for peak, and 98% and 40% for volume found across the Northeast, Coastal Southeast, and Northwest with intensifying rainfall accumulation and variability, respectively. This study advances flood predictions by better informing the flood attributes in the context of dynamical land‐atmosphere perturbations.

An Optimal Interpolation–Based Snow Data Assimilation for NOAA’s Unified Forecast System (UFS)

Abstract Within the National Weather Service’s Unified Forecast System (UFS), snow depth and snow cover observations are assimilated once daily using a rule-based method designed to correct for gross errors. While this approach improved the forecasts over its predecessors, it is now quite outdated and is likely to result in suboptimal analysis. We have then implemented and evaluated a snow data assimilation using the 2D optimal interpolation (OI) method, which accounts for model and observation errors and their spatial correlations as a function of distances between the observations and model grid cells. The performance of the OI was evaluated by assimilating daily snow depth observations from the Global Historical Climatology Network (GHCN) and the Interactive Multisensor Snow and Ice Mapping System (IMS) snow cover data into the UFS, from October 2019 to March 2020. Compared to the control analysis, which is very similar to the method currently in operational use, the OI improves the forecast snow depth and snow cover. For instance, the unbiased snow depth root-mean-squared error (ubRMSE) was reduced by 45 mm and the snow cover hit rate increased by 4%. This leads to modest improvements to globally averaged near-surface temperature (an average reduction of 0.23 K in temperature bias), with significant local improvements in some regions (much of Asia, the central United States). The reduction in near-surface temperature error was primarily caused by improved snow cover fraction from the data assimilation. Based on these results, the OI DA is currently being transitioned into operational use for the UFS. Significance Statement Weather and climate forecasting systems rely on accurate modeling of the evolution of atmospheric, oceanic, and land processes. In addition, model forecasts are substantially improved by continuous incorporation of observations to models, through a process called data assimilation. In this work, we upgraded the snow data assimilation used in the U.S. National Weather Service (NWS) global weather prediction system. Compared to the method currently in operational use, the new snow data assimilation improves both the forecasted snow quantity and near-surface air temperatures over snowy regions. Based on the positive results obtained in the experiments presented here, the new snow data assimilation method is being implemented in the NWS operational forecast system.

Daily High-Resolution Temperature and Precipitation Fields for the Contiguous United States from 1951 to Present

Abstract In this paper, a new set of daily gridded fields and area averages of temperature and precipitation is introduced that covers the contiguous United States (CONUS) from 1951 to present. With daily updates and a grid resolution of approximately 0.0417° (nominally 5 km), the product, named nClimGrid-Daily, is designed to be used particularly in climate monitoring and other applications that rely on placing event-specific meteorological patterns into a long-term historical context. The gridded fields were generated by interpolating morning and midnight observations from the Global Historical Climatology Network–Daily dataset using thin-plate smoothing splines. Additional processing steps limit the adverse effects of spatial and temporal variations in station density, observation time, and other factors on the quality and homogeneity of the fields. The resulting gridded data provide smoothed representations of the point observations, although the accuracy of estimates for individual grid points and days can be sensitive to local spatial variability and the ability of the available observations and interpolation technique to capture that variability. The nClimGrid-Daily dataset is therefore recommended for applications that require the aggregation of estimates in space and/or time, such as climate monitoring analyses at regional to national scales. Significance Statement Many applications that use historical weather observations require data on a high-resolution grid that are updated daily. Here, a new dataset of daily temperature and precipitation for 1951–present is introduced that was created by interpolating irregularly spaced observations to a regular grid with a spacing of 0.0417° across the contiguous United States. Compared to other such datasets, this product is particularly suitable for monitoring climate and drought on a daily basis because it was processed so as to limit artificial variations in space and time that may result from changes in the types and distribution of observations used.