Variations in the spectral scattering coefficient of marine particles [b_p(λ)] were measured at 241 locations in oceanic (case 1) and coastal (case 2) waters around Europe. The scattering coefficient at 555 nm normalized to the dry mass of particles [b^m_p (555)] was, on average, 1.0 and 0.5 m² g^‐1 in case 1 and case 2 waters, respectively. Spectral variations in b_p(λ) were, on average, small in all investigated waters. To understand the observed variations in particle scattering, we performed calculations based on the Mie theory with various values for the slope for the Junge‐type size distribution, refractive index, and apparent density of particles (dry weight/wet volume). The latter two were varied according to the type of particles (organic versus mineral) and, for organic ones, as a function of the water content. The higher b^m_p(555) values in case 1 waters are mainly due to the low apparent density, which results from the organic nature of particles and their elevated water content. In all investigated coastal regions, except for the Baltic Sea, the low b^m_p(555) values can be explained by the dominant presence of mineral particles, characterized by a high density that counterbalances the effect of a higher refractive index. In the Baltic Sea, b^m_p(555) was similar to values found in other coastal waters despite the fact that particles were dominantly organic, which may result from higher absorption relative to scattering. A smaller than expected increase of b_p(λ) toward short wavelengths is attributed to significant absorption that increases toward the shorter wavelengths and reduces scattering, whether particles are living, detrital, or mineral. Our analyses suggest that the determination of b^m_p may be significantly sensitive to the porosity of the filter used to assess the dry mass of particles.

The Global Energy and Water cycle Exchanges (GEWEX) Data and Assessments Panel (GDAP) initiated the GEWEX Water Vapor Assessment (G-VAP), which has the main objectives to quantify the current state of the art in water vapour products being constructed for climate applications and to support the selection process of suitable water vapour products by GDAP for its production of globally consistent water and energy cycle products. During the construction of the G-VAP data archive, freely available and mature satellite and reanalysis data records with a minimum temporal coverage of 10 years were considered. The archive contains total column water vapour (TCWV) as well as specific humidity and temperature at four pressure levels (1000, 700, 500, 300 hPa) from 22 different data records. All data records were remapped to a regular longitude–latitude grid of 2^∘ × 2^∘. The archive consists of four different folders: 22 TCWV data records covering the period 2003–2008, 11 TCWV data records covering the period 1988–2008, as well as 7 specific humidity and 7 temperature data records covering the period 1988–2009. The G-VAP data archive is referenced under the following digital object identifier (doi): https://doi.org/10.5676/EUM_SAF_CM/GVAP/V001. Within G-VAP, the characterization of water vapour products is, among other ways, achieved through intercomparisons of the considered data records, as a whole and grouped into three classes of predominant retrieval condition: clear-sky, cloudy-sky and all-sky. Associated results are shown using the 22 TCWV data records. The standard deviations among the 22 TCWV data records have been analysed and exhibit distinct maxima over central Africa and the tropical warm pool (in absolute terms) as well as over the poles and mountain regions (in relative terms). The variability in TCWV within each class can be large and prohibits conclusions about systematic differences in TCWV between the classes.

A model developed recently by Loisel and Stramski [Appl. Opt. 39, 3001–3011 (2000)] for estimating the spectral absorption a(λ), scattering b(λ), and backscattering b _b(λ) coefficients in the upper ocean from the irradiance reflectance just beneath the sea surface R(λ, z = 0^-) and the diffuse attenuation of downwelling irradiance within the surface layer 〈K _d(λ)〉₁ is compared with measurements. Field data for this comparison were collected in different areas including off-shore and near-shore waters off southern California and around Europe. The a(λ) and b _b(λ) values predicted by the model in the blue-green spectral region show generally good agreement with measurements that covered a broad range of conditions from clear oligotrophic waters to turbid coastal waters affected by river discharge. The agreement is still good if the model estimates of a(λ) and b _b(λ) are based on R(λ, z = 0^-) used as the only input to the model available from measurements [as opposed to both R(λ, z = 0^-) and 〈K _d(λ)〉₁ being measured]. This particular mode of operation of the model is relevant to ocean-color remote-sensing applications. In contrast to a(λ) and b _b(λ) the comparison between the modeled and the measured b(λ) shows large discrepancies. These discrepancies are most likely attributable to significant variations in the scattering phase function of suspended particulate matter, which were not included in the development of the model.

To date, a large variety of water vapour data records from satellite and reanalysis are available. It is key to understand the quality and uncertainty of these data records in order to fully exploit these records and to avoid data being employed incorrectly or misinterpreted. Therefore, it is important to inform users on accuracy and limitations of these data records based on consistent inter-comparisons carried out in the framework of international assessments. Addressing this challenge is the major objective of the Global Water and Energy Exchanges (GEWEX) water vapor assessment (G-VAP) which was initiated by the GEWEX Data and Assessments Panel (GDAP). Here, an overview of G-VAP objectives and an introduction to the results from G-VAP’s first phase are given. After this overview, a summary of available data records on water vapour and closely related variables and a short introduction to the utilized methods are presented. The results from inter-comparisons, homogeneity testing and inter-comparison of trend estimates, achieved within G-VAP’s first phase are summarized. The conclusions on future research directions for the wider community and for G-VAP’s next phase are outlined and recommendations have been formulated. For instance, a key recommendation is the need for recalibration and improved inter-calibration of radiance data records and subsequent reprocessing in order to increase stability and to provide uncertainty estimates. This need became evident from a general disagreement in trend estimates (e.g., trends in TCWV ranging from −1.51 ± 0.17 kg/m²/decade to 1.22 ± 0.16 kg/m²/decade) and the presence of break points on global and regional scale. It will be a future activity of G-VAP to reassess the stability of updated or new data records and to assess consistency, i.e., the closeness of data records given their uncertainty estimates.