ESTIMATING NEAR-SURFACE WIND FIELD BY AERIAL THERMAL IMAGE VELOCIMETRY

By adjusting the distribution of vertical layers and increasing its number in WRF model, this paper mainly studies the effects of vertical stratification on the near surface wind field and vertical profile simulation. The test outcomes show that moderately increasing vertical layers can effectively improve the near surface wind field simulation results, while it has little influence on the numeral and changing trend of high vertical wind profile. Considering both accuracy and efficiency, it is recommended to set 10~15 layers below 300m. On the basis of this research, instead of USGS data by using the MODIS_30S data, the data underlying surface land in Shenzhen and HK area are updated. Comparative results between the two schemes, due to the roughness and drag coefficient of difference types of surface are not identical; the surface data has a significant impact on wind field and wind profile simulation. Using the MODIS land use data which is more consistent with the actual situation can improve the accuracy of numerical simulation.

Download Full-text

Reconstruction of Near-Surface Tornado Wind Fields from Forest Damage

Journal of Applied Meteorology and Climatology ◽

10.1175/2010jamc2254.1 ◽

2010 ◽

Vol 49 (7) ◽

pp. 1517-1537 ◽

Cited By ~ 33

Author(s):

Veronika Beck ◽

Nikolai Dotzek

Keyword(s):

Wind Field ◽

Surface Wind ◽

Forest Damage ◽

Wind Fields ◽

Tree Model ◽

Translation Speed ◽

Near Surface ◽

Additional Information ◽

Secondary Vortices ◽

Surface Wind Field

Abstract Tornado intensity is usually inferred from the damage produced. To foster postevent tornado intensity assessments, the authors present a model to reconstruct near-surface wind fields from forest damage patterns. By comparing the structure of observed and simulated damage patterns, essential parameters to describe a tornado near-surface wind field are derived, such as the ratio Gmax between circular and translational velocity, and the deflection angle α between peak wind and pressure gradient. The model consists of a wind field module following the Letzmann analytical tornado model and a tree module based on the mechanistic HWIND tree model to assess tree breakage. Using this method, the velocity components of the near-surface wind field, the track of the tornado center, and the spatial distribution of the Fujita scale along and across the damage path can be assessed. Necessary requirements to apply the model are knowledge of the tornado translation speed (e.g., from radar observations) and a detailed analysis of the forest damage patterns. One of the key findings of this analysis is that the maximum intensity of the tornado is determinable with an uncertainty of only (Gmax + 1) times the variability of the usually well-known tornado translation speed. Further, if Letzmann’s model is applied and the translation speed of the tornado is known, the detailed tree model is unnecessary and could be replaced by an average critical velocity for stem breakage υcrit independent of the tree species. Under this framework, the F3 and F2 ratings of the tornadoes in Milosovice, Czech Republic, on 30 May 2001 and Castellcir, Spain, on 18 October 2006, respectively, could be verified. For the Milosovice event, the uncertainty in peak intensity was only ±6.0 m s−1. Additional information about the structure of the near-surface wind field in the tornado and several secondary vortices was also gained. Further, this model allows for distinguishing downburst damage patterns from those of tornadoes.

Download Full-text

Adaptive Thermal Image Velocimetry of spatial wind movement on landscapes using near target infrared cameras

10.5194/amt-2021-335 ◽

2021 ◽

Author(s):

Benjamin Schumacher ◽

Marwan Katurji ◽

Jiawei Zhang ◽

Peyman Zawar-Reza ◽

Benjamin Adams ◽

...

Keyword(s):

Spatial Information ◽

Turbulent Transport ◽

Surface Wind ◽

Infrared Camera ◽

Thermal Image ◽

Transport Processes ◽

Near Surface ◽

Remote Sensing Technique ◽

Sonic Anemometers ◽

Image Velocimetry

Abstract. Thermal Image Velocimetry (TIV) is a near-target remote sensing technique for estimating two- dimensional near-surface wind velocity based on spatiotemporal displacement of fluctuations in surface brightness temperature captured by an infrared camera. The addition of an automated parameterization and the combination of ensemble TIV results into one output made the method more suitable to changing meteorological conditions and less sensitive to noise stemming from the airborne sensor platform. Three field campaigns were carried out to evaluate the algorithm over turf, dry grass and wheat stubble. The derived velocities were validated with independently acquired observations from fine wire thermocouples and sonic anemometers. It was found that the TIV technique correctly derives atmospheric flow patterns close to the ground. Moreover, the modified method resolves wind speed statistics close to the surface at a higher resolution than the traditional measurement methods. Adaptive Thermal Image Velocimetry (A-TIV) is capable of providing contact-less spatial information about near-surface atmospheric motion and can help to be a useful tool in researching turbulent transport processes close to the ground.

Download Full-text

Wind field reconstruction with adaptive random Fourier features

Proceedings of The Royal Society A Mathematical Physical and Engineering Sciences ◽

10.1098/rspa.2021.0236 ◽

2021 ◽

Vol 477 (2255) ◽

Author(s):

Jonas Kiessling ◽

Emanuel Ström ◽

Raúl Tempone

Keyword(s):

Wind Field ◽

Spatial Interpolation ◽

Surface Wind ◽

Sobolev Norm ◽

Near Surface ◽

Distance Weighting ◽

Random Fourier Features ◽

Sparse Set ◽

Surface Wind Field ◽

Inverse Distance

We investigate the use of spatial interpolation methods for reconstructing the horizontal near-surface wind field given a sparse set of measurements. In particular, random Fourier features is compared with a set of benchmark methods including kriging and inverse distance weighting. Random Fourier features is a linear model β ( x ) = ∑ k = 1 K β k e i ω k x approximating the velocity field, with randomly sampled frequencies ω k and amplitudes β k trained to minimize a loss function. We include a physically motivated divergence penalty | ∇ ⋅ β ( x ) | 2 , as well as a penalty on the Sobolev norm of β . We derive a bound on the generalization error and a sampling density that minimizes the bound. We then devise an adaptive Metropolis–Hastings algorithm for sampling the frequencies of the optimal distribution. In our experiments, our random Fourier features model outperforms the benchmark models.

Download Full-text