THE ATMOSPHERIC DYNAMICS MISSION FOR GLOBAL WIND FIELD MEASUREMENT

Abstract. Understanding the vertical wind profile is paramount for design and operation of wind turbines. It is needed not only for extrapolation of the wind velocity to hub height but also for structural load calculations, to name the most obvious issues. As wind turbines grow in size and development transitions offshore, issues such as shallow surface layers, low-level jets (LLJs) and internal boundary layers (IBLs) are raising questions concerning the applicability of the commonly used Monin–Obukhov similarity theory to accurately describe the vertical wind development at modern wind turbine hub heights. In this study the 10 min averaged vertical wind profile up to a minimum elevation of 100 m is analyzed through measurements collected from seven sites which represent a span of conditions. Three sites are located offshore in the North Sea and Baltic Sea (FINO1, FINO2 and FINO3) with varying fetch, two onshore by the Norwegian coast (Frøya and Valsneset), one further onshore by the Danish coast (Høvsøre), and one inland at a forested site in Sweden (Ryningsnäs). Through analysis of data series ranging from 8 months to several years depending on the site, the wind profile has been quantitatively categorized according to the number of exhibited local maxima which is not possible within the Monin–Obukhov similarity theory. The results reveal that the occurrence of local maxima is higher at sites of low surface roughness and a high prevalence of unstable atmospheric conditions, causing up to 17 % severely inflected abnormal profiles at the most exposed offshore site, which decreases as the location transitions from offshore to coastal to further inland and is lowest at the forested site. The results indicate that issues in predicting the vertical wind profile are most prevalent offshore, where very stable inflections cause severe deviations which may be related to an offshore internal boundary layer. These findings suggest that there is evident need of an improved vertical wind profile description in order to improve the accuracy of power predictions and load calculations, especially at offshore and coastal sites.

A discussion of the best mode of expressing the facts of the mean westerly flow in temperate latitudes parametrically leads to the result that over North America the shape of the vertical geostrophic westerly wind profile is nearly invariant with respect to time and place. The vertical wind profile can be expressed in terms either of the change pressure-gradient with height, or of contour-height gradient with pressure-level. In either case the profile shape is characteristic, and varies little with season, longitude or latitude over North America. The intensity of the mean circulation varies rythmically with season, but that for any one month may vary markedly from year to year. The profile shape for an oceanic region differs from that of a land region such as eastern North America in exibiting a lower ratio of high-level to surface wind, and in being much more variable from month to month. It is shown that the pressure-difference at 7 km between two latitudes, or the difference in mean height of the 200 mb surface gives a measure of the total geostrophic momentum between those latitudes. The square of these quantities also gives a measure of the total kinetic energy of the westerlies between these two latitudes. DOI: 10.1111/j.2153-3490.1951.tb00808.x

The 'Living Planet Programme' of the European Space Agency (ESA) is gaining momentum. Six Earth Explorer missions are currently being implemented including ADM- Aeolus, ESA's Doppler Wind Lidar (DWL) mission. The Aeolus mission will demonstrate the capability of a space- borne high spectral resolution DWL to accurately measure wind profiles in the troposphere and the lower stratosphere (0- 30 km). The Mission thus addresses one of the main identified deficiencies of the current Global Observing System (GOS). From the backscattered frequency-shifted laser light it will be possible to obtain about 3,000 globally distributed horizontal line-of-sight (HLOS) wind profiles daily. The accuracy of the Aeolus winds, in cloud-free regions and above thick clouds, is expected to be comparable to that of radiosonde wind measurements. Additional geophysical products that will be retrieved from the Aeolus measurements are cloud and aerosol optical properties. Aeolus HLOS wind profiles will find wide application in Numerical Weather Prediction (NWP) and climate studies, improving the accuracy of numerical weather forecasting, advancing our understanding of tropical dynamics and processes relevant to climate variability and climate modelling. Impact experiments, assimilating synthetic Aeolus wind data into operational NWP models, have already been performed. One focus has been the forecast performance in regions known to be particularly sensitive to the accuracy of the initial conditions. The tentative results show that the largest benefits from Aeolus HLOS winds can be expected over the oceans and in the Tropics. In view of other lidar missions in space, the potential of a long-term database of cloud and aerosol optical properties is being studied. An additional topic is the potential benefit of wind profile observations in the lower stratosphere. Ground-based and airborne campaigns are being prepared for the validation of the Aeolus Airborne Demonstrator (A2D) - a high spectral resolution DWL instrument with technology very similar to Aeolus. This paper provides an overview of the Aeolus mission, the science and application activities being performed in support of the mission and the potential benefit of such observations in a wider context.

In this paper monthly trends of vertical wind profiles within and above an urban area with complex topography (Tehran) have been investigated using data from a Sodar, a 100-m and two 28-m towers and some surface stations. It includes the occurrence, evolution, dissipation time, peak time and maximum height for katabatic–anabatic winds due to topography, return flow and urban circulation. Vertical distributions of wind and the heat and momentum fluxes up to 600 m were also considered. The vertical wind profiles which have diurnal and seasonal variations show southwesterly daily anabatic wind and northeasterly nocturnal katabatic wind. Daily vertical wind profile structure which has two layers and two jets contains a decrease approximately at 300–400 m which may be due to the return flow and a daily thermal convective flow in the urban convergent terrain. At night the nocturnal wind profile structure, in the majority of months, has two layers and sometimes three layers containing northeasterly flows. In two layers structure, a decrease can be seen between two layers. In three layers structure, the middle layer has a different direction that indicates the return flow and urban circulation more clearly. Katabatic flows also develop in successive phases varying from 1 to 3 phases in different months. Investigation on surface wind demonstrates that in cold months a drainage flow from a valley located in the west of the station can affect wind speed and direction especially at evening transition time. These flow patterns can influence the way air pollutants disperse in this area.

Introducing two Random Forest based methods for cloud detection in remote sensing images

To identify the influence of turbulence and vertical wind profile in wind turbine performance, wind speed measurements at different heights have been performed. Measurements have been developed using a cup anemometer and a LIDAR equipment, specifically a pulsed wave one. The wind profile has been recorded to study the effect of the atmospheric conditions over the energy generated by a wind turbine located close to the LIDAR system. The changes in the power production of the wind turbine are relevant.

In the framework of the Firestorm project, four portable weather stations were installed in the Lousã-Estrela mountainous ridge. The data observed with this equipment, along with two other stations from IPMA’s surface network, allows the monitoring of the vertical profile of temperature, relative humidity and wind in the region. Many forest fires in mainland Portugal affect areas with complex terrain, with hills or mountains in the 400-1200m asl range, where weather conditions strongly influence the fire propagation and lifetime. Therefore, it becomes even more pertinent to study the vertical profiles of temperature, humidity and wind as they often exhibit large variability, namely in the night-time period. This paper provides the first study of the available observations by providing statistics for the summer 2021 and winter 2021/22. It has also allowed the identification of some periods in which significant temperature inversions with very dry air at medium levels or even the valley have occurred. Further details of the vertical profile and the weather pattern that led to these events are shown in three selected events.

Cloud detection in remote sensing images is a challenging but significant task. Due to the variety and complexity of underlying surfaces, most of the current cloud detection methods have difficulty in detecting thin cloud regions. In fact, it is quite meaningful to distinguish thin clouds from thick clouds, especially in cloud removal and target detection tasks. Therefore, we propose a method based on multiscale features-convolutional neural network (MF-CNN) to detect thin cloud, thick cloud, and noncloud pixels of remote sensing images simultaneously. Landsat 8 satellite imagery with various levels of cloud coverage is used to demonstrate the effectiveness of our proposed MF-CNN model. We first stack visible, near-infrared, short-wave, cirrus, and thermal infrared bands of Landsat 8 imagery to obtain the combined spectral information. The MF-CNN model is then used to learn the multiscale global features of input images. The high-level semantic information obtained in the process of feature learning is integrated with low-level spatial information to classify the imagery into thick, thin and noncloud regions. The performance of our proposed model is compared to that of various commonly used cloud detection methods in both qualitative and quantitative aspects. Compared to other cloud detection methods, the experimental results show that our proposed method has a better performance not only in thick and thin clouds but also in the entire cloud regions.

Effect of modelled areas and inflow turbulence on the wind characteristics over a densely built-up city

New calculations of the relative optical air mass function are made over the 0°–87° range of apparent solar zenith angle θ, for various vertical profiles of background aerosol, diamond dust and thin cirrus cloud particle extinction coefficient in the Arctic and Antarctic atmospheres. The calculations were carried out by following the Tomasi and Petkov (2014) procedure, in which the above-mentioned vertical profiles derived from lidar observations were used as weighting functions. Different sets of lidar measurements were examined, recorded using: (i) the Koldewey-Aerosol-Raman Lidar (KARL) system (AWI, Germany) at Ny-Ålesund (Spitsbergen, Svalbard) in January, April, July and October 2013; (ii) the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) satellite-based sensor over Barrow (Alaska), Eureka (Nunavut, Canada) and Sodankylä (northern Finland), and Neumayer III, Mario Zucchelli and Mirny coastal stations in Antarctica in the local summer months of the last two years; (iii) the National Institute of Optics (INO), National Council of Research (CNR) Antarctic lidar at Dome C on the Antarctic Plateau for a typical “diamond dust” case; and (iv) the KARL lidar at Ny-Ålesund and the University of Rome/National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA) lidar at Thule (northwestern Greenland) for some cirrus cloud layers in the middle and upper troposphere. The relative optical air mass calculations are compared with those obtained by Tomasi and Petkov (2014) to define the seasonal changes produced by aerosol particles, diamond dust and cirrus clouds. The results indicate that the corresponding air mass functions generally decrease as angle θ increases with rates that are proportional to the increase in the pure aerosol, diamond dust and cirrus cloud particle optical thickness.

The most turbulent vortices that populate the forest-atmosphere interface have canopy height length scales. These vortices are mainly responsible for turbulent exchanges between inside and above canopy region. Thus, we used the vertical wind profiles obtained by 10 anemometers installed inside and above the forest canopy of the Rebio-Jarú experimental site, in the Amazon Rainforest. A third degree polynomial function was developed to better fit the wind profile and therefore estimate the inflection point height of the vertical wind profile (zi) a length scale associated with wind shear (Ls), and the wind speed at height zi. These length and velocity scales were used to obtain better fits for the dimensional wind profiles and turbulence statistical moments. Three dimensionless profile models were compared using friction velocity, wind velocity in zi and wind velocity at canopy height. It was observed that the dimensionless profiles using the velocity and shear calculated at zi provided support for the elaboration of more realistic parameterization of the turbulent exchange processes that occur both at the forest-atmosphere interface and inside the canopy.

Fire behaviour is strongly affected by wind speed and direction. Wind affects fire spread rate bending the flames closer to unburnt fuel, increasing the rate of combustion and by blowing embers ahead of the main fire front lighting spotfires. Stronger winds blow flames forward, lowering the flame angle and subsequent distance between the flame and the fuel. In open areas there is typical a logarithmic vertical wind profile; winds become faster as height increases. However forest vegetation disrupts wind flow, and a logarithmic profile cannot be assumed. When predicting forest fire spread, wind reduction factors are typically used to reduce open area wind speeds to the more sheltered sub-canopy winds that directly affect fire spread. These methods typically assume a consistent vertical wind profile that does not change with wind speed. In this study we investigated these assumptions. We characterised the variation in wind speed at different heights in different vegetation types and at different open area wind speeds. To do this, we used cup anemometers to measure the horizontal wind speed at different heights in four differently structured forest areas. Our results show that the use on a single wind reduction factor is a gross over-simplification; horizontal wind speeds can vary by a factor of three depending of the height above ground. We have also found that even for a given height, the wind reduction factor can vary by a factor of two or more depending on the strength of the winds in the open. This is in addition to a factor of 10 or more variation in wind speed between different vegetation types. The variation in wind speed in different vegetation type, height above ground and wind speed in the open means that there is potentially a large margin of error in predicting fire behaviour using wind speeds measured or forecast for 10 m in the open for fires in forests or shrublands. Some strategies for improving estimates of wind speed for fire spread prediction are discussed.

<strong class="journal-contentHeaderColor">Abstract.</strong> Aerosol particle optical, physical and chemical properties have been previously studied in the United Arab Emirates (UAE), but there is still a gap in the knowledge of particle sources, and in the horizontal and vertical transport of aerosol particles and their precursors in the area. To investigate how aerosol particle and SO₂ concentrations at the surface responded to changes in horizontal and vertical transport, we used data from a one-year measurement campaign at a background site where local sources of SO₂ where expected to be minimal. The measurement campaign provided a combination of in-situ measurements at the surface, and the boundary layer evolution from vertical and horizontal wind profiles measured by a Doppler lidar. The diurnal structure of the boundary layer in the UAE was very similar from day to day, with deep well-mixed boundary layer during the day transitioning to a shallow nocturnal layer, with the maximum boundary layer height usually being reached around 1400 local time. Both SO₂ and nucleation mode aerosol particle concentrations were elevated for surface winds coming from the east or western sectors. We attribute this to oil refineries located on the eastern and western coasts of the UAE. The concentrations of larger cloud condensation nuclei (CCN) sized particles and their activation fraction did not show any clear dependence on wind direction, but the CCN number concentration showed some dependence on wind speed, with higher concentrations coinciding with the weakest surface winds. Peaks in SO₂ concentrations were also observed despite low surface wind speeds and wind directions unfavourable for transport. However, winds aloft were much stronger, with wind speeds of 10 m s^-1 at 1 km common at night, and with wind directions favourable for transport, and surface-measured concentrations increased rapidly once these particular layers started to be entrained into the growing boundary layer, even if the surface wind direction was from a clean sector. These conditions also displayed higher nucleation mode aerosol particle concentrations, i.e. new particle formation events occurring due to the increase in the gaseous precursor.

Optical satellite images are the main source of geophysical information which can be used for various land surface studies. But one of the most challenging issue faced by these images are the severe cloud contamination. As far as optical satellite images are of concern, thick clouds will completely obstructs the observation of landscape and uneven illumination caused by thin clouds will further limits the processing of optical satellite images. So a method to eliminate the impact of clouds in optical satellite images is necessary to improve the effectiveness and availability of remote sensing data for various land surface studies. In this paper, a novel methodology for removing the uneven illumination caused by thick and thin clouds is intro-duced. For this approach, multitemporal cloud contaminated optical satellite images are required with the assumption that in each image cloud cover change significantly over a short duration of time. A cloud detection algorithm in frequency domain is used to detect the cloud contaminated portions in multitemporal optical satellite images. Suitable cloud free patch from reference images for reconstructing cloud contaminated portions in target image are selected based on lower amount of cloud rate and quality assessment technique. Some experimental analysis is conducted on collected Aqua MODIS sensor captured multitemporalsatellite images. Experimental result shows that the proposed algorithm can effectively remove majority of cloud contamination within the optical satellite images. Finally quantitative measures are performed to evaluate the feasibility of the methodology.

It is generally assumed that viscosity smooths out vertical gradients of the horizontal thermospheric winds in the upper thermosphere, and thus observations of neutral winds at one height can be used at other altitudes in this region. In this paper we present neutral wind simulations of the May 1997 geomagnetic storm using the Coupled Magnetosphere‐Ionosphere‐Thermosphere (CMIT) model. The model results show that during quiet periods, the assumption of a shearless vertical profile of the horizontal winds is generally valid in low and middle latitudes, although vertical shears do occur in wind profiles in the upper thermosphere in some locations at higher latitudes. During disturbed periods, large variations in the vertical profiles of the upper thermospheric winds are seen globally in the model simulations. A diagnostic analysis of the forcing processes in the neutral momentum equations shows that (1) during quiet time, there are shearing forces, most noticeably the pressure gradient and ion drag, in the upper thermosphere that result in a net momentum forcing that changes with height; this induces altitude variations in the wind profiles at high latitudes and sometime even at middle latitudes. (2) During storm time, momentum advection, which is relatively weak during the quiet time, becomes a dominant force globally. Pressure gradient forces are also significantly enhanced. Ion drag, on the other hand, can be enhanced or suppressed, depending on the location of positive or negative effects. All these forces exhibit significant altitude variations that lead to a net force that is greatly enhanced and has large vertical shears. This produces globally enhanced neutral winds that vary with height. (3) Viscosity is less important than other forcing processes during both the quiet and storm periods and thus cannot prevent shears from occurring in the vertical profiles of the horizontal winds. Viscosity has an insignificant effect on vertical shears that change with height linearly. It, however, does restrict vertical shears that vary nonlinearly with height. The effectiveness of the viscosity in restricting such shears depends on its magnitude. In our simulations, viscosity is weaker than other forcing processes and thus is a relatively slow process, so it will take a few hours for viscosity to reduce such shears.