Climate models still need to be improved in their capability of reproducing the present climate at both global and regional scale. The assessment of their performance depends on the datasets used as comparators. Reanalysis and gridded (homogenized or not homogenized) observational datasets have been frequently used for this purpose. However, none of these can be considered a reference dataset. Here, for the first time, using in-situ measurements from NOAA U.S. Climate Reference Network (USCRN), a network of 139 stations with high-quality instruments deployed across the continental U.S, daily temperature, and precipitation from a suite of dynamically downscaled regional climate models (RCMs; driven by ERA-Interim) involved in NA-CORDEX are assessed. The assessment is extended also to the most recent and modern widely used reanalysis (ERA5, ERA-Interim, MERRA2, NARR) and gridded observational datasets (Daymet, PRISM, Livneh, CPC). Results show that biases for the different datasets are mainly seasonal and subregional dependent. On average, reanalysis and in-situ-based datasets are generally warmer than USCRN year-round, while models are colder (warmer) in winter (summer). In-situ-based datasets provide the best performance in most of the CONUS regions compared to reanalysis and models, but still have biases in regions such as the Midwest mountains and the Northwestern Pacific. Results also highlight that reanalysis does not outperform RCMs in most of the U.S. subregions. Likewise, for both reanalysis and models, temperature and precipitation biases are also significantly depending on the orography, with larger temperature biases for coarser model resolutions and precipitation biases for reanalysis.

The RHARM (Radiosounding HARMonization) algorithm is the first to provide homogenized radiosonde-based records of temperature, relative humidity and wind profiles since 1978, alongside an estimation of the observational uncertainty for each observation and pressure level. The algorithm and the dataset are presented in the companion paper. In this paper we assess the performance of the dataset through comparison with some of the most widely used climate data records. The RHARM adjustments reduce the difference with the reanalysis especially in the northern hemisphere for temperature and relative humidity. The study of temperature trends at different pressure levels reveals a good agreement between RHARM, reanalysis and existing radiosounding homogenized datasets (<0.1K per decade above 300 hPa, 0.25 K per decade below). For relative humidity, the discrepancies among the datasets are more significant, although RHARM trends are most similar to the reanalysis. For wind speed, the comparison indicates a good agreement above 300 hPa. Compared to IGRA, RHARM also improves by 50% the agreement with the estimated trends in the lower stratosphere from MSU (Microwave Sounding Unit) deep layer averages. For water vapour, the good performance of post-2004 RHARM data is quantified from the comparison of the 300 hPa monthly means in tropics between RHARM and AURA/MLS, as the absolute mean difference is 0.01 g/kg for RHARM and 0.03 g/kg for IGRA, and correlation increases from 0.95 to 0.99. A validation of the observational uncertainties of RHARM is also presented showing that they provide a good estimate or overestimate the theoretical distribution.

Observational records are more often than not influenced by residual non‐climatic factors which must be detected and adjusted for prior to their usage. Moreover, measurement uncertainties should be properly quantified and validated. In this work we present a novel approach, named RHARM (Radiosounding HARMonization), to provide a homogenized dataset of temperature, humidity and wind profiles along with an estimation of the measurement uncertainties for 700 radiosounding stations globally. The RHARM method has been used to adjust twice daily (0000 and 1200 UTC) radiosonde data holdings at 16 pressure levels from 1000 to 10 hPa from 1978 to the present from the Integrated Global Radiosonde Archive (IGRA). Relative humidity (RH) data are limited to 250 hPa. The applied adjustments are interpolated to all reported significant levels. RHARM is the first dataset to provide homogenized time series of temperature, relative humidity and wind profiles alongside an estimation of the observational uncertainty for each observation at each pressure level. The comparison of RHARM and unadjusted profiles highlights a median temperature warmer by 0.6 K in the boreal hemisphere, while in the tropics RHARM is cooler by 0.1 K. For RH, the difference is -2.1%, while in the tropics it is reduced to 0.3%. For wind speed, adjustments largely improve the data homogeneity locally. Analysis of decadal trends for temperature, RH and winds highlights increased the geographical coherency of trends. In a companion paper, the performances of the RHARM dataset are assessed through comparison with the reanalysis, satellite and other homogenized radiosonde datasets.