A constant value of air density based on its annual average value at a given location is commonly used for the computation of the annual energy production in wind industry. Thus, the correction required in the estimation of daily, monthly or seasonal wind energy production, due to the use of air density, is ordinarily omitted in existing literature. The general method, based on the implementation of the wind speed’s Weibull distribution over the power curve of the turbine, omits it if the power curve is not corrected according to the air density of the site. In this study, the seasonal variation of air density was shown to be highly relevant for the computation of offshore wind energy potential around the Iberian Peninsula. If the temperature, pressure, and moisture are taken into account, the wind power density and turbine capacity factor corrections derived from these variations are also significant. In order to demonstrate this, the advanced Weather Research and Forecasting mesoscale Model (WRF) using data assimilation was executed in the study area to obtain a spatial representation of these corrections. According to the results, the wind power density, estimated by taking into account the air density correction, exhibits a difference of 8% between summer and winter, compared with that estimated without the density correction. This implies that seasonal capacity factor estimation corrections of up to 1% in percentage points are necessary for wind turbines mainly for summer and winter, due to air density changes.

Several sensitivity experiments with the Weather Research and Forecasting (WRF) model version 3.8.1 have been performed to find the optimal parameterization setup for precipitation amounts and patterns around Mount Kenya at a convection-permitting scale of 1 km. Hereby, the focus is on the cumulus scheme, with tests of the Kain–Fritsch, the Grell–Freitas, and no cumulus parameterizations. In addition, two longwave radiation schemes and two planetary boundary layer parameterizations are evaluated, and different nesting ratios and numbers of nests are tested. The precipitation amounts and patterns are compared against a large amount of weather station data and three gridded observational data sets. The temporal correlation of monthly precipitation sums show that fewer nests lead to a more constrained simulation, and hence the correlation is higher. The pattern correlation with weather station data confirms this result, but when comparing it to the most recent gridded observational data set the difference between the number of nests and nesting ratios is marginal. The precipitation patterns further reveal that using the Grell–Freitas cumulus parameterization in the domains with resolutions >5 km provides the best results when it comes to precipitation patterns and amounts. If no cumulus parameterization is used in any of the domains, the temporal correlation between gridded and in situ observations and simulated precipitation is especially poor with more nests. Moreover, even if the patterns are captured reasonably well, a clear overestimation in the precipitation amounts is simulated around Mount Kenya when using no cumulus scheme in all domains. The experiment with the Grell–Freitas cumulus parameterization in the domains with resolutions >5 km also provides reasonable results for 2 m temperature with respect to gridded observational and weather station data.

There have been numerous statistical and dynamical downscaling model comparisons. However, differences in model skill can be distorted by inconsistencies in experimental set-up, inputs and output format. This paper harmonizes such factors when evaluating daily precipitation downscaled over the Iberian Peninsula by the Statistical DownScaling Model (SDSM) and two configurations of the dynamical Weather Research and Forecasting Model (WRF) (one with data assimilation (D) and one without (N)). The ERA-Interim reanalysis at \(0.75^{\circ }\) resolution provides common inputs for spinning-up and driving the WRF model and calibrating SDSM. WRF runs and SDSM output were evaluated against ECA&D stations, TRMM, GPCP and EOBS gridded precipitation for 2010–2014 using the same suite of diagnostics. Differences between WRF and SDSM are comparable to observational uncertainty, but the relative skill of the downscaling techniques varies with diagnostic. The SDSM ensemble mean, WRF-D and ERAI have similar correlation scores (\(\hbox {r}= 0.45\)–0.7), but there were large variations amongst SDSM ensemble members (\(\hbox {r}=0.3\)–0.6). The best Linear Error in Probability Space (\(\hbox {LEPS}= 0.001\)–0.007) and simulations of precipitation amount were achieved by individual members of the SDSM ensemble. However, the Brier Skill Score shows these members do not improve the prediction by ERA-Interim, whereas precipitation occurrence is reproduced best by WRF-D. Similar skill was achieved by SDSM when applied to station or gridded precipitation data. Given the greater computational demands of WRF compared with SDSM, clear statements of expected value-added are needed when applying the former to climate impacts and adaptation research.

The performance of the Weather Research and Forecasting (WRF) model version 3.8.1 at convection-permitting scale is evaluated by means of several sensitivity simulations over southern Peru down to a grid resolution of 1 km, whereby the main focus is on the domain with 5 km horizontal resolution. Different configurations of microphysics, cumulus, longwave radiation, and planetary boundary layer schemes are tested. For the year 2008, the simulated precipitation amounts and patterns are compared to gridded observational data sets and weather station data gathered from Peru, Bolivia, and Brazil. The temporal correlation of simulated monthly accumulated precipitation against in situ and gridded observational data show that the most challenging regions for WRF are the slopes along both sides of the Andes, i.e. elevations between 1000 and 3000 m above sea level. The pattern correlation analysis between simulated precipitation and station data suggests that all tested WRF setups perform rather poorly along the northeastern slopes of the Andes during the entire year. In the southwestern region of the domain the performance of all setups is better except for the driest period (May–September). The results of the pattern correlation to the gridded observational data sets show that all setups perform reasonably well except along both slopes during the dry season. The precipitation patterns reveal that the typical setup used over Europe is too dry throughout the entire year, and that the experiment with the combination of the single-moment 6-class microphysics scheme and the Grell–Freitas cumulus parameterization in the domains with resolutions larger than 5 km, suitable for East Africa, does not perfectly apply to other equatorial regions such as the Amazon basin in southeastern Peru. The experiment with the Stony Brook University microphysics scheme and the Grell-Freitas cumulus parameterization tends to overestimate precipitation over the northeastern slopes of the Andes, but enforces a positive feedback between the soil moisture, air temperature, relative humidity, mid-level cloud cover and, finally, precipitation. Hence, this setup provides the most accurate results over the Peruvian Amazon, and particularly over the department of Madre de Dios, which is a region of interest because it is considered a biodiversity hotspot of Peru. The robustness of this particular configuration of the model is backed up by similar results obtained during wet climate conditions observed in 2012.