Abstract:

This dataset includes experimental observations of a single-phase (saltwater) plume injected in a homogeneous rotating environment.

Suggested Citation:

Julien Landel, Stuart Dalziel, Paul Linden. 2018. Laboratory experiments on single-phase plumes in a rotating environment. Distributed by: GRIIDC, Harte Research Institute, Texas A&M University–Corpus Christi. doi:10.7266/N7XS5SR0

Publications:

Frank, D., Landel, J. R., Dalziel, S. B., & Linden, P. F. (2017). Anticyclonic precession of a plume in a rotating environment. Geophysical Research Letters, 44(18), 9400–9407. doi:10.1002/2017gl074191

Purpose:

To investigate how the rotation of the environment affects the dynamics of the plume.

Data Parameters and Units:

Water depth (H, m), Nozzle radius (r_0 = d_0/2, m), Flow rate (q_0, l/min), Initial density (rho_0, g/cm^3), Rotation rate (omega, rad/s), Richardson number (Gamma), Buoyancy flux (B, cm^4/s^3), Rossby number

Methods:

This data set contains experimental recordings of a salt-water plume in a homogeneous rotating environment. The most remarkable observation is the anticyclonic plume precession. The results have been published in "Frank D., J. R. Landel, S. B. Dalziel, and P. F. Linden (2017), Anticyclonic precession of a plume in a rotating environment, Geophys. Res. Lett., 44, doi:10.1002/2017GL074191". This article also contains the description of the experimental setup and of the technique for the data analysis. "paramaters.xls" contains a table of experimental parameters. Two tanks were used for the experiments: tank A and tank B. The analysis to extract the precession frequency is conducted is a separate folder for each experiment: "TankX/PrecessionFrequency/SeriesX/xxx". The number at the very beginning of the folder name indicates the rotation rate in rad/s. For example, "0x4increasedB" means that the experiment was conducted for Omega=0.4 rad/s. The folder for each experiment contains a digigflow .dfi movie file. This is the sideview recording of the plume cropped to the region of 300x150 pixels just below the plume nozzle. The name of the video file is the date on which the experiment was conducted. The same recording is also provided as an .avi movie. The subfolder "Images" contains the recording decomposed into individual frames. The subfolders "ProcIm" and "ProcImSmooth" contain the images which show the automatically determined plume centreline using the method described in the published article and after smoothing the data, respectively. Also, the matlab file "Plume_Loc_Smooth" in the "ProcImSmooth" is a matrix containing time-dependent location of the plume centreline: the columns correspond to the frame number and the rows correspond to the distance from the plume nozzle in pixels. The entries are the pixel numbers, counted from the left edge of the image, where we detected the position of the plume. For example, the entry 46 in 100th row and 328th column means that in the 328th frame at a distance of 100 pixels from the plume nozzle we determined that the plume centreline is located in the 46th pixel from the left edge of the image. The time-dependent position of the plume centreline from the "Plume_Loc_Smooth" file is analysed using the Hilbert-Huang transform. The folder for each experiment contains matlab files "amplitude" and "frequency" which are the time-dependent amplitude and frequency obtained after applying the Hilbert-Huang transform to the time-dependent location of the plume centerline at a fixed distance from the nozzle (= fixed row in the "Plume_Loc_Smooth" file). As a bonus, the figures showing the signal and its decomposition in IMFs as well as matlab files comprising the data for IMFs and the residue are also provided. The IMF containing the information about the precession frequency of the plume needs to be determined manually by choosing the IMF in the "IMFsOnePlot" figure which corresponds most to the visually perceived oscillation of the plume. The average precession frequency for each experiment is calculated in the "precessionFrequencyHHTxxx.m" file located in the "SeriesX" folder. The result is saved as a "precessionRatexxx" vector in the same folder. Finally, all the determined precession frequencies of the plume for different experimental series are brought together in the "precessionFrequencyHHT.m" file in the "TankB/PrecessionFrequency" folder, in which we linearly fit the data and plot it in one single figure. The "grid" folders contain the data for the experiments which were conducted with an additional mesh inserted into the plume nozzle to change the turbulence level. IMF are the "intrinsic mode functions" of the Hilbert-Huang transform. This technique is described in our publication. The averageLightIntensity.mat files contain the data as a Matlab matrix for an average light-intensity image of the corresponding experiment (= all the individual frames summed up and divided by the total number of frames). Some experiments in Tank A- Series G and H do not contain the subfolders ProcIM and ProcIMSmoth since it was difficult to clearly distinguish the plume precession because the dye obscured the plume nozzle (as argued in our publication). Hence, this experimental data was not included in our data analysis. However, the experiment images were supplied for those experiments. The software to open the *.dfm file can be downloaded here: http://www.dalzielresearch.com/digiflow/

Individual Files: