U.S. Climate Reference Network Research Articles

Using machine learning to produce near surface soil moisture estimates from deeper in situ records at U.S. Climate Reference Network (USCRN) locations: Analysis and applications to AMSR-E satellite validation

Surface soil moisture is a critical parameter for understanding the energy flux at the land atmosphere boundary. Weather modeling, climate prediction, and remote sensing validation are some of the applications for surface soil moisture information. The most common in situ measurement for these purposes are sensors that are installed at depths of approximately 5cm. There are however, sensor technologies and network designs that do not provide an estimate at this depth. If soil moisture estimates at deeper depths could be extrapolated to the near surface, in situ networks providing estimates at other depths would see their values enhanced. Soil moisture sensors from the U.S. Climate Reference Network (USCRN) were used to generate models of 5cm soil moisture, with 10cm soil moisture measurements and antecedent precipitation as inputs, via machine learning techniques. Validation was conducted with the available, in situ, 5cm resources. It was shown that a 5cm estimate, which was extrapolated from a 10cm sensor and antecedent local precipitation, produced a root-mean-squared-error (RMSE) of 0.0215m3/m3. Next, these machine-learning-generated 5cm estimates were also compared to AMSR-E estimates at these locations. These results were then compared with the performance of the actual in situ readings against the AMSR-E data. The machine learning estimates at 5cm produced an RMSE of approximately 0.03m3/m3 when an optimized gain and offset were applied. This is necessary considering the performance of AMSR-E in locations characterized by high vegetation water contents, which are present across North Carolina. Lastly, the application of this extrapolation technique is applied to the ECONet in North Carolina, which provides a 10cm depth measurement as its shallowest soil moisture estimate. A raw RMSE of 0.028m3/m3 was achieved, and with a linear gain and offset applied at each ECONet site, an RMSE of 0.013m3/m3 was possible.

Deploying temporary networks for upscaling of sparse network stations

Soil observations networks at the national scale play an integral role in hydrologic modeling, drought assessment, agricultural decision support, and our ability to understand climate change. Understanding soil moisture variability is necessary to apply these measurements to model calibration, business and consumer applications, or even human health issues. The installation of soil moisture sensors as sparse, national networks is necessitated by limited financial resources. However, this results in the incomplete sampling of the local heterogeneity of soil type, vegetation cover, topography, and the fine spatial distribution of precipitation events. To this end, temporary networks can be installed in the areas surrounding a permanent installation within a sparse network. The temporary networks deployed in this study provide a more representative average at the 3km and 9km scales, localized about the permanent gauge. The value of such temporary networks is demonstrated at test sites in Millbrook, New York and Crossville, Tennessee. The capacity of a single U.S. Climate Reference Network (USCRN) sensor set to approximate the average of a temporary network at the 3km and 9km scales using a simple linear scaling function is tested. The capacity of a temporary network to provide reliable estimates with diminishing numbers of sensors, the temporal stability of those networks, and ultimately, the relationship of the variability of those networks to soil moisture conditions at the permanent sensor are investigated. In this manner, this work demonstrates the single-season installation of a temporary network as a mechanism to characterize the soil moisture variability at a permanent gauge within a sparse network.

An Assessment of HIRS Surface Air Temperature with USCRN Data

The surface air temperature retrievals from the High Resolution Infrared Radiation Sounder (HIRS) are evaluated by using observations from the U.S. Climate Reference Network (USCRN) for the period of 2006 to 2013. One year of the USCRN data is also used as ground truth in calibrating retrieval biases. The final retrieval results show that mean biases of HIRS retrievals from comparisons to all surface stations for each year are mostly in the range of ±0.2 °C, and the root mean square difference (RMSD) values are 3.2–3.5 °C. Results for biases of individual stations are mostly within ±2 °C. In average, RMSDs are smaller over the eastern U.S. than over the western U.S., smaller at nighttime than at daytime, and smaller at lower elevations. The comparison patterns are consistent from year to year and for different satellites, showing the potential of HIRS data for long-term studies.

Evaluating the impact of U.S. Historical Climatology Network homogenization using the U.S. Climate Reference Network

AbstractNumerous inhomogeneities including station moves, instrument changes, and time of observation changes in the U.S. Historical Climatological Network (USHCN) complicate the assessment of long‐term temperature trends. Detection and correction of inhomogeneities in raw temperature records have been undertaken by NOAA and other groups using automated pairwise neighbor comparison approaches, but these have proven controversial due to the large trend impact of homogenization in the United States. The new U.S. Climate Reference Network (USCRN) provides a homogenous set of surface temperature observations that can serve as an effective empirical test of adjustments to raw USHCN stations. By comparing nearby pairs of USHCN and USCRN stations, we find that adjustments make both trends and monthly anomalies from USHCN stations much more similar to those of neighboring USCRN stations for the period from 2004 to 2015 when the networks overlap. These results improve our confidence in the reliability of homogenized surface temperature records.

Multi‐Profile Analysis of Soil Moisture within the US Climate Reference Network

Core Ideas Hydraprobe measurements can be quantified. USCRN gauges distributed nationally are analyzed. Random error magnitudes are a function of how recently precipitation has occurred. Soil moisture estimates are crucial for hydrologic modeling and agricultural decision‐support efforts. These measurements are also pivotal for long‐term inquiries regarding the impacts of climate change and the resulting droughts over large spatial and temporal scales. However, it has only been the past decade during which ground‐based soil moisture sensory resources have become sufficient to tackle these important challenges. Despite this progress, random and systematic errors remain in ground‐based soil moisture observations. Such errors must be quantified (and/or adequately minimized) before such observations can be used with full confidence. In response, this paper calibrates and analyzes US Climate Reference Network (USCRN) profile estimates at each of three sensors collocated at each USCRN location. With each USCRN location consisting of three independent, Hydraprobe measurements, triple collocation analysis of these sensory triads reveals the random error associated with this particular sensing technology in each individual location. This allows quantification of the accuracy of these individual profiles, the random errors associated with these measurements in different geographic locations, and offers the potential for more adept quality control procedures in near real time. Averaged over USCRN gauge locations nationally, this random error is determined to be approximately 0.012 m3/m3.

Future measurements for climate monitoring

Future measurements for climate monitoring

Comparing AMSR-E soil moisture estimates to the extended record of the U.S. Climate Reference Network (USCRN)

Soil moisture plays an integral role in multi-scale hydrologic modeling, agricultural decision analysis, climate change assessments, and drought prediction/prevention. The broad availability of soil moisture estimates has only occurred within the past decade through a combination of in situ networks and satellite-driven remote sensing. The U.S. Climate Reference Network (USCRN) has provided a nationwide in situ resource since 2009. The Advanced Microwave Scanning Radiometer (AMSR-E), launched in 2002, is one of the satellite products available for comparison, but there are a limited number of years where the data records overlap. This study compares the results of modeled historical soil moisture estimates derived using USCRN precipitation data to the remotely sensed estimates provided by the AMSR-E satellite between 2002 and 2011. First, this work assesses the calibrated model's similarity to in situ estimates. Next, the model estimates and in situ measurements are shown to perform comparably well against the AMSR-E satellite product, suggesting that it may be possible to utilize modeled estimates at times and locations where satellite estimates are unavailable and further extend the soil moisture record spatially and temporally.

Evaporation from weighing precipitation gauges: impacts on automated gauge measurements and quality assurance methods

Abstract. Evaporation from a precipitation gauge can cause errors in the amount of measured precipitation. For automated weighing-bucket gauges, the World Meteorological Organization (WMO) suggests the use of evaporative suppressants and frequent observations to limit these biases. However, the use of evaporation suppressants is not always feasible due to environmental hazards and the added cost of maintenance, transport, and disposal of the gauge additive. In addition, research has suggested that evaporation prior to precipitation may affect precipitation measurements from auto-recording gauges operating at sub-hourly frequencies. For further evaluation, a field campaign was conducted to monitor evaporation and its impacts on the quality of precipitation measurements from gauges used at U.S. Climate Reference Network (USCRN) stations. Two Geonor gauges were collocated, with one gauge using an evaporative suppressant (referred to as Geonor-NonEvap) and the other with no suppressant (referred to as Geonor-Evap) to evaluate evaporative losses and evaporation biases on precipitation measurements. From June to August, evaporative losses from the Geonor-Evap gauge exceeded accumulated precipitation, with an average loss of 0.12 mm h−1. The impact of evaporation on precipitation measurements was sensitive to the choice of calculation method. In general, the pairwise method that utilized a longer time series to smooth out sensor noise was more sensitive to gauge evaporation (−4.6% bias with respect to control) than the weighted-average method that calculated depth change over a smaller window (

Methods to Calculate Precipitation from Weighing-Bucket Gauges with Redundant Depth Measurements

AbstractThe U.S. Climate Reference Network (USCRN) monitors precipitation using a well-shielded Geonor T-200B gauge. To ensure the quality and continuity of the data record, the USCRN adopted an innovative approach to monitor precipitation using redundant technology: three vibrating-wire load sensors measuring the liquid depth of a weighing-bucket gauge. In addition to detecting and flagging suboptimally operating sensors, quality assurance (QA) approaches also combine the redundant observations into a precipitation measurement. As an early adopter of this technology, USCRN has pioneered an effort to develop QA strategies for such precipitation systems.The initial USCRN approach to calculating precipitation from redundant depth observations, pairwise calculation (pairCalc), was found to be sensitive to sensor noise and gauge evaporation. These findings led to the development of a new approach to calculating precipitation that minimized these nonprecipitation impacts using a weighted average calculation (wavgCalc). The two calculation approaches were evaluated using station data and simulated precipitation scenarios with a known signal. The new QA system had consistently lower measures of error for simulated precipitation events. Improved handling of sensor noise and gauge evaporation led to increases in network total precipitation of 1.6% on average. These results indicate the new calculation system will improve the quality of USCRN precipitation measurements, making them a more reliable reference dataset with the capacity to monitor the nation’s precipitation trends (mean and extremes). In addition, this study provides valuable insight into the development and evaluation of QA systems, particularly for networks adopting redundant approaches to monitoring precipitation.

Observational Perspectives from U.S. Climate Reference Network (USCRN) and Cooperative Observer Program (COOP) Network: Temperature and Precipitation Comparison

AbstractThe U.S. Cooperative Observer Program (COOP) network was formed in the early 1890s to provide daily observations of temperature and precipitation. However, manual observations from naturally aspirated temperature sensors and unshielded precipitation gauges often led to uncertainties in atmospheric measurements. Advancements in observational technology (ventilated temperature sensors, well-shielded precipitation gauges) and measurement techniques (automation and redundant sensors), which improve observation quality, were adopted by NOAA’s National Climatic Data Center (NCDC) into the establishment of the U.S. Climate Reference Network (USCRN). USCRN was designed to provide high-quality and continuous observations to monitor long-term temperature and precipitation trends, and to provide an independent reference to compare to other networks. The purpose of this study is to evaluate how diverse technological and operational choices between the USCRN and COOP programs impact temperature and precipitation observations. Naturally aspirated COOP sensors generally had warmer (+0.48°C) daily maximum and cooler (−0.36°C) minimum temperatures than USCRN, with considerable variability among stations. For precipitation, COOP reported slightly more precipitation overall (1.5%) with network differences varying seasonally. COOP gauges were sensitive to wind biases (no shielding), which are enhanced over winter when COOP observed (10.7%) less precipitation than USCRN. Conversely, wetting factor and gauge evaporation, which dominate in summer, were sources of bias for USCRN, leading to wetter COOP observations over warmer months. Inconsistencies in COOP observations (e.g., multiday observations, time shifts, recording errors) complicated network comparisons and led to unique bias profiles that evolved over time with changes in instrumentation and primary observer.

Extending the soil moisture data record of the U.S. Climate Reference Network (USCRN) and Soil Climate Analysis Network (SCAN)

Soil moisture estimates are valuable for hydrologic modeling, drought prediction and management, climate change analysis, and agricultural decision support. However, in situ measurements of soil moisture have only become available within the past few decades with additional sensors being installed each year. Comparing newer in situ resources with older resources, previously required a period of cross-calibration, often requiring several years of data collection. One new technique to improve this issue is to develop a methodology to extend the in situ record backwards in time using a soil moisture model and ancillary available data sets. This study will extend the soil moisture record of the U.S. Climate Reference Network (USCRN) by calibrating a precipitation-driven model during the most recent few years when soil moisture data are available and applying that model backwards temporally in years where precipitation data are available and soil moisture data are not. This approach is validated by applying the technique to the Soil Climate Analysis Network (SCAN) where the same model is calibrated in recent years and validated during preceding years at locations with a sufficiently long soil moisture record. Results suggest that if two or three years of concurrent precipitation and soil moisture time series data are available, the calibrated model’s parameters can be applied historically to produce RMSE values less than 0.033m3/m3. With this approach, in locations characterized by in situ sensors with short or intermittent data records, a model can now be used to fill the relevant gaps and improve the historical record as well.

U.S. Climate Reference Network Soil Moisture and Temperature Observations

AbstractThe U.S. Climate Reference Network (USCRN) is a network of climate-monitoring stations maintained and operated by the National Oceanic and Atmospheric Administration (NOAA) to provide climate-science-quality measurements of air temperature and precipitation. The stations in the network were designed to be extensible to other missions, and the National Integrated Drought Information System program determined that the USCRN could be augmented to provide observations that are more drought relevant. To increase the network’s capability of monitoring soil processes and drought, soil observations were added to USCRN instrumentation. In 2011, the USCRN team completed at each USCRN station in the conterminous United States the installation of triplicate-configuration soil moisture and soil temperature probes at five standards depths (5, 10, 20, 50, and 100 cm) as prescribed by the World Meteorological Organization; in addition, the project included the installation of a relative humidity sensor at each of the stations. Work is also under way to eventually install soil sensors at the expanding USCRN stations in Alaska. USCRN data are stewarded by the NOAA National Climatic Data Center, and instrument engineering and performance studies, installation, and maintenance are performed by the NOAA Atmospheric Turbulence and Diffusion Division. This article provides a technical description of the USCRN soil observations in the context of U.S. soil-climate–measurement efforts and discusses the advantage of the triple-redundancy approach applied by the USCRN.

U.S. Climate Reference Network Soil Moisture Observations with Triple Redundancy: Measurement Variability

Between 2009 and 2011, the U.S. Climate Reference Network (USCRN) was augmented with soil moisture/soil temperature probes and atmospheric relative humidity instruments as part of a programmatic expansion in support of the National Integrated Drought Information System (NIDIS). The 114 sites in this sparse network are well distributed across the conterminous United States in open, rural locations expected to remain unchanged in land use for many decades into the future. Soil probes are installed in triplicate redundancy, similar to the air temperature and precipitation measurements, at either five standard World Meteorological Organization (WMO) depths (5, 10, 20, 50, and 100 cm) or only two depths (5 and 10 cm) depending on the nature of the underlying materials. Stations also measure air temperature, surface skin temperature, precipitation, solar radiation, and 1.5‐m wind speed. In addition to sensor failure, the triplicate design of USCRN soil probes have allowed for an initial characterization of variability of soil moisture measurements. Nationwide analysis of soil moisture during early‐to‐mid growing season in 2011 and 2012 was performed to examine the differences in response to the widespread drought of 2012. The redundancy of the network helps retain the continuity of the record over time, and also provides key insights into the variations of measurements at a single location that are related to a combination of installation effects and the impacts of soil differences at the local level. This article highlights the usefulness of deploying triplicate configurations of soil probes for detecting faulty sensors and for better understanding the nature of soil moisture measurement variability.

U.S. Climate Reference Network after One Decade of Operations: Status and Assessment

The year 2012 marks a decade of observations undertaken by the U.S. Climate Reference Network (USCRN) under the auspices of NOAA's National Climatic Data Center and Atmospheric Turbulence and Diffusion Division. The network consists of 114 sites across the conterminous 48 states, with additional sites in Alaska and Hawaii. Stations are installed in open (where possible), rural sites very likely to have stable land-cover/use conditions for several decades to come. At each site a suite of meteorological parameters are monitored, including triple redundancy for the primary air temperature and precipitation variables and for soil moisture/temperature. Instrumentation is regularly calibrated to National Institute for Standards and Technology (NIST) standards and maintained by a staff of expert engineers. This attention to detail in USCRN is intended to ensure the creation of an unimpeachable record of changes in surface climate over the United States for decades to come. Data are made available without restriction for all public, private, and government use. This article describes the rationale for the USCRN, its implementation, and some of the highlights of the first decade of operations. One critical use of these observations is as an independent data source to verify the existing U.S. temperature record derived from networks corrected for nonhomogenous histories. Future directions for the network are also discussed, including the applicability of USCRN approaches for networks monitoring climate at scales from regional to global. Constructive feedback from end users will allow for continued improvement of USCRN in the future and ensure that it continues to meet stakeholder requirements for precise climate measurements.

Critical assessment of surface incident solar radiation observations collected by SURFRAD, USCRN and AmeriFlux networks from 1995 to 2011

Surface incident solar radiation (Rs) drives weather and climate changes. Observations of Rs have been widely used as reference data to evaluate climate model simulations and satellite retrievals. However, few have studied uncertainties of Rs observations, especially long term. This paper compares Rs from 1995 to 2011 at collocated sites collected by the Surface Radiation Budget Network (SURFRAD), the U.S. Climate Reference Network (USCRN) and the AmeriFlux network. SURFRAD stations have measured separately the diffuse and direct components of Rs as well as Rs by a pyranometer, while Rs was measured by a pyranometer or a net radiometer at the USCRN and AmeriFlux sites. Rs can be calculated by summing the diffuse and direction radiation measurements. Rs measured by the summation technique was compared those measured by a pyranometer or a net radiometer at collocated sites. Agreement among these four independent Rs measurements is good with correlation coefficients higher than 0.98 and an average error (one standard deviation) of about 4% at both hourly and monthly time scales. Rs has a large spatial variability at the hourly time scale, even exceeding 100 W m−2 in ∼6 km. This spatial variability is substantially reduced at the monthly time scale. The two independent measurement systems at the SURFRAD sites agree rather well in annual variability of Rs with an average relative standard deviation error of 34%. The errors are 71% and 85% for the USCRN and AmeriFlux sites. Evidently, caution should be taken when using the Rs data collected at the USCRN and AmeriFlux sites to study annual variability of Rs.

Observing Climate at High Elevations Using United States Climate Reference Network Approaches

AbstractThe U.S. Climate Reference Network (USCRN) was deployed between 2001 and 2008 for the purpose of yielding high-quality and temporally stable in situ climate observations in pristine environments over the twenty-first century. Given this mission, USCRN stations are engineered to operate largely autonomously with great reliability and accuracy. A triplicate approach is used to provide redundant measurements of temperature and precipitation at each location, allowing for observations at a specific time to be compared for quality control. This approach has proven to be robust in the most extreme environments, from extreme cold (−49°C) to extreme heat (+52°C), in areas of heavy precipitation (4700 mm yr−1), and in locations impacted by strong winds, freezing rain, and other hazards. In addition to a number of stations enduring extreme winter environments in Alaska and the northern United States, seven of the USCRN stations are located at elevations over 2000 m, including stations on Mauna Loa, Hawaii (3407 m) and on Niwot Ridge above Boulder, Colorado (2996 m). The USCRN temperature instruments and radiation shield have also been installed and run successfully at a station on the Quelccaya Ice Cap in Peru (5670 m). This paper reviews the performance of the USCRN station network during its brief lifetime and the potential utility of its triplicate temperature instrument configuration for measuring climate change at elevation.

Evaluation of the Relationship between Air and Land Surface Temperature under Clear- and Cloudy-Sky Conditions

Abstract Clear and cloudy daytime comparisons of land surface temperature (LST) and air temperature (Tair) were made for 14 stations included in the U.S. Climate Reference Network (USCRN) of stations from observations made from 2003 through 2008. Generally, LST was greater than Tair for both the clear and cloudy conditions; however, the differences between LST and Tair were significantly less for the cloudy-sky conditions. In addition, the relationships between LST and Tair displayed less variability under the cloudy-sky conditions than under clear-sky conditions. Wind speed, time of the observation of Tair and LST, season, the occurrence of precipitation at the time of observation, and normalized difference vegetation index values were all considered in the evaluation of the relationship between Tair and LST. Mean differences between LST and Tair of less than 2°C were observed under cloudy conditions for the stations, as compared with a minimum difference of greater than 2°C (and as great as 7+°C) for the clear-sky conditions. Under cloudy conditions, Tair alone explained over 94%—and as great as 98%—of the variance observed in LST for the stations included in this analysis, as compared with a range of 81%–93% for clear-sky conditions. Because of the relatively homogeneous land surface characteristics encouraged in the immediate vicinity of USCRN stations, and potential regional differences in surface features that might influence the observed relationships, additional analyses of the relationships between LST and Tair for additional regions and land surface conditions are recommended.

Impacts of Land Use/Land Cover Change on Climate and Future Research Priorities

Several recommendations have been proposed for detecting land use and land cover change (LULCC) on the environment from, observed climatic records and to modeling to improve its understanding and its impacts on climate. Researchers need to detect LULCCs accurately at appropriate scales within a specified time period to better understand their impacts on climate and provide improved estimates of future climate. The US Climate Reference Network (USCRN) can be helpful in monitoring impacts of LULCC on near-surface atmospheric conditions, including temperature. The USCRN measures temperature, precipitation, solar radiation, and ground or skin temperature. It is recommended that the National Climatic Data Center (NCDC) and other climate monitoring agencies develop plans and seek funds to address any monitoring biases that are identified and for which detailed analyses have not been completed.

Estimating precipitation normals for USCRN stations

Normals estimation in this study is based on the fact that monthly precipitation anomalies at any target location are similar to those of neighboring stations. Precipitation normals have been generated for stations of the newly developed U.S. Climate Reference Network (USCRN) by using data from this network and 4629 stations of the National Weather Service Cooperative Station network (COOP). This analysis evaluated several variations of estimation approaches. The best method that was used in USCRN normals estimation was one that included the use of monthly departure data from ∼11 neighboring COOP stations (within ∼78 km of a USCRN station), and a weighting scheme that used the inverse square difference in monthly precipitation totals between the COOP and USCRN stations. When the estimated normals are based on a few years of data (e.g., 3 years of data), their errors are about 9–12% of true normals, and are about 19–28% of the typical magnitude of year‐to‐year variability. This suggests that these estimated USCRN normals can be useful in climate monitoring and assessment.

Validation of GOES-Based Insolation Estimates Using Data from the U.S. Climate Reference Network

Abstract Reliable procedures that accurately map surface insolation over large domains at high spatial and temporal resolution are a great benefit for making the predictions of potential and actual evapotranspiration that are required by a variety of hydrological and agricultural applications. Here, estimates of hourly and daily integrated insolation at 20-km resolution, based on Geostationary Operational Environmental Satellite (GOES) visible imagery are compared to pyranometer measurements made at 11 sites in the U.S. Climate Reference Network (USCRN) over a continuous 15-month period. Such a comprehensive survey is necessary in order to examine the accuracy of the satellite insolation estimates over a diverse range of seasons and land surface types. The relatively simple physical model of insolation that is tested here yields good results, with seasonally averaged model errors of 62 (19%) and 15 (10%) W m−2 for hourly and daily-averaged insolation, respectively, including both clear- and cloudy-sky conditions. This level of accuracy is comparable, or superior, to results that have been obtained with more complex models of atmospheric radiative transfer. Model performance can be improved in the future by addressing a small elevation-related bias in the physical model, which is likely the result of inaccurate model precipitable water inputs or cloud-height assessments.