Zhibao Dong

The flux profile of a blowing sand cloud: a wind tunnel investigation

Abstract The flux profile of a blowing sand cloud, or the variation of blown sand flux with height, is the reflection of blown sand particles that move in different trajectories, and also the basis for checking drifting sand. Here we report the wind tunnel results of systematic tests of the flux profiles of different sized sands at different free-stream wind velocities. The results reveal that within the 60-cm near-surface layer, the decay of blown sand flux with height can be expressed by an exponential function: q h = a exp(− h / b ), where, q h is the blown sand transport rate at height h , a and b are parameters that vary with wind velocity and sand size. The significance of coefficient a and b in the function is defined: a represents the transport rate in true creep and b implies the relative decay rate with height of the blown sand transport rate. The true creep fraction, the ratio of the sand transported on the surface ( h =0) to the total transport varies widely, decreasing with both sand size and wind speed. The flux profiles are converted to straight lines by plotting sand transport rate, q h , on a log-scale. The slope of the straight lines that represents the relative decay rate with height of sand transport rate decreases with an increase in free-stream wind velocity and sand grain size, implying that relatively more of the blown sand is transported to greater heights as grain size and wind speed increase. The average saltating height represented by the height where 50% of the cumulative flux percentage occurs increases with both wind speed and grain size, implying that saltation becomes more intense as grain size and/or wind velocity increase.

The blown sand flux over a sandy surface: a wind tunnel investigation on the fetch effect

Abstract Detailed wind tunnel tests were conducted to examine the fetch effect of a sandy surface on a sand cloud blowing over it. The results suggest that the fetch length of a sandy surface has a significant effect on both the vertical flux profile and total horizontal flux. The sand flux over a sandy surface increases with height in the very near surface layer, but then decays exponentially. In agreement with the widely accepted conclusion, the decay function can be expressed by q=aexp(−h/b), where q is the sand flux at height h. Coefficient a that tends to increase with wind speed implies the influence of wind, while coefficient b that defines the relative decay rate shows the influence of both the fetch and wind. The relative decay rate increases with fetch when the fetch length is short, then becomes constant when the fetch reaches a certain length. The threshold fetch length over which the relative decay rate keeps constant increases with wind speed. The average saltation height generally increases with fetch. Both the relative decay rate and average saltation height show that the fetch effect on the flux profile becomes more significant when the wind speed increases. The total sand transport equation for the total fetch can be expressed by Q=C(1−Ut/U)2U3(ρ/g), where Q is the total sand transport rate, U and Ut are the wind velocity and threshold wind velocity at the centerline height of the wind tunnel, respectively, g is gravitational acceleration, ρ is the density of air, and C is a proportionality coefficient that increases with the fetch length, implying that the total sand flux increases with the fetch length.

Aerodynamic roughness of gravel surfaces

Abstract The interactions between surface winds and gravel surfaces that can be characterized by aerodynamic roughness length have important implications for sediment mobilization, transport, sedimentation and the development of desert pavements (‘gobi’ deserts). Wind tunnel results of systematic tests carried out to determine the aerodynamic roughness length of gravel surfaces of different size and coverage at a range of free-stream wind velocities are reported. The aerodynamic roughness of gravel surfaces is a function of gravel size, gravel coverage and free-stream wind velocity. The 1/30 law proposed by Bagnold, based on Nikuradses results of water pipe flow, or taking the aerodynamic roughness length as a fixed percentage of roughness element height is not supported for gravel surfaces. The aerodynamic roughness of gravel surfaces decreases with free-stream wind velocity. The variation of aerodynamic roughness with gravel coverage can be described by quadratic curves. Maximum aerodynamic roughness length occurs at gravel coverage range above 40–75%, increasing with gravel size. Multivariate models based on free-stream wind velocity and gravel coverage have been developed for predicting the aerodynamic roughness of different gravel-sized surfaces. The maturity of the roughness effects of gravel surfaces increases with gravel coverage. In general, the roughness effects are fully developed when the gravel coverage is over 15%.

Grain size characteristics of dune sands in the central Taklimakan Sand Sea

Abstract Compared with the sand grain size of the other sand seas, central Taklimakan Sand Sea has some of the finest sands seen globally. The dunes are composed of fine and very fine sands with a diameter between 2.00 and 4.00 φ (0.25–0.063 mm). There are differences in the grain size distributions for different dune types. Mean grain size of the compound/complex crescent dune sands is 3.08 φ; the compound dome dunes 3.21 φ; the compound/complex linear dune sands between 2.63 and 3.41 φ; and the star dunes 2.81 φ. From the northern edge of the sand sea towards southern edge, the components of fine and very fine sands increase, closely related to the wind regimes, time scale for dune development, and underlying sediments. Compared with the foregoing, there are some differences for the compound/complex linear dunes in the centre of Taklimakan. Though the pattern of coarser crests does exist, the sorting parameters indicate that the sands on the west flank of compound/complex linear dunes are better sorted than that on the east flank and crest. Sampling on the superimposed dunes developed on the surface of the complex linear dunes suggests that the patterns of finer crest, coarser crest all existed in central Taklimakan. Under the conditions of low energy wind regime when fine sand is available, the frequency of the pattern for finer crest and better sorting will increase. Sampling on the simple crescent dunes developed on the interdunes suggests that the simple dunes are developed originally with nearly the same grain size distributions and parameters. With the development of the dunes and evolution of the dune morphology, the grain size distributions of the dune sands varied under the actions of the wind regimes, time scale and underlying sediments.

Geomorphology of sand dunes in the Northeast Taklimakan Desert

Three types of sand dunes exist in the Taklimakan Desert, namely compound/complex crescent dunes and crescent chains, compound dome dunes and compound/complex linear dunes. Besides these three compound/complex types, single simple dunes are also distributed throughout the sand sea. The compound/complex linear dunes are developed under acute bimodal wind regimes. Though the ratios of the resultant drift potential (RDP) and the drift potential (DP) are the same as that near the border and adjacent area of the sand sea, the compound/complex crescent and dome dunes are developed, respectively, because of divergence of the sand available, the stress of the sand-moving winds and the time scales of dune formation. The sand supply for the dunes is not from Lopo Nor in the east as previous studies suggested but mainly from local alluvial or lacustrine deposits. The grain size component does not correlate evidently to the morphology parameters of the sand dunes. Analyses of the DP and drift direction suggest that the northeast Taklimakan is an area of low wind energy and the resultant drift direction (RDD) coincides well with the distribution, morphology and scales of the dunes.

Velocity profile of a sand cloud blowing over a gravel surface

Particle dynamic analyzer (PDA) measurement technology was used to study the turbulent characteristics and the variation with height of the mean horizontal (in the downwind direction) and vertical (in the upward direction) particle velocity of a sand cloud blowing over a gravel surface. The results show that the mean horizontal particle velocity of the cloud increases with height, while the mean vertical velocity decreases with height. The variation of the mean horizontal velocity with height is, to some extent, similar to the wind profile that increases logarithmically with height in the turbulent boundary layer. The variation of the mean vertical velocity with height is much more complex than that of the mean horizontal velocity. The increase of the resultant mean velocity with height can be expressed by a modified power function. Particle turbulence in the downwind direction decreases with height, while that in the vertical direction is complex. For fine sands (0.2–0.3 mm and 0.3–0.4 mm), there is a tendency for the particle turbulence to increase with height. In the very near-surface layer (<4 mm), the movement of blown sand particles is very complex due to the rebound of particles on the bed and the interparticle collisions in the air. Wind starts to accelerate particle movement about 4 mm from the surface. The initial rebound on the bed and the interparticle collisions in the air have a profound effect on particle movement below that height, where particle concentration is very high and wind velocity is very low.

Monitoring sand dune advance in the Taklimakan Desert

The migration rate of sand dunes is important for the design of the sand-control system in the Taklimakan Desert. Sand dune movement was monitored in a sample plot along the desert-crossing highway by means of a topographical survey in late 1991, 1992 and 1993. The results reveal that most of the morphometric parameters of the dunes are not as well correlated as they are generally supposed to be. The mean advance rate of the dunes was 7.29 and 5.56 m year−1 in 1992 and 1993, respectively. The advance direction is towards the SW, approximately in accordance with the local resultant wind direction. The advance rate of the dunes was controlled by the local wind regime and dune morphometry. However, the relationships between advance rate and the morphometric parameters of the dunes are variable for the immature dunes. The small scale and immaturity of the dunes, insufficient sand supply, and complex wind modes are responsible for the complex dune parameter relationship, involving frequent changes of dune morphology, as well as changes of dune advance rate.

Short-term dynamics of wind erosion of three newly cultivated grassland soils in Northern China

This study examined the dynamic nature of short-term wind erosion processes of three grassland soils under simulated cultivation. In a laboratory wind tunnel, soil loss on a sandy loam, a loamy sand, and a sand soil was measured in six successive 10-, 3-, and 1-min exposures at the speeds of 10, 18, and 26 m s 1 , respectively. Erodible soil particle sizes of each interval were determined through direct dry sieving. It was observed that increasing blowing duration generally led to rapid decrease of soil loss, decrease of the erodible soil particle size from a definitely restricted area by any definite speed of wind, and the non-erodible particles remaining on the soil surface increased. From the first to the last blowing interval, soil loss decreased by 86.4–94.8% at 10 m s 1 in 60 min, 90.0–95.4% at 18 m s 1 in 18 min, and 71.5–97.8% at 26 m s 1 in 6 min. The results indicate that wind erosion on the newly cultivated soils is most intense during the initial blowing. The decrease of soil loss with time is presumed due to modification of soil surface by blowing wind, i.e. the concentration of non-erodible aggregates with the deflation of the erodible particles on the soil surfaces.

Aerodynamic roughness of fixed sandy beds

The nature of interactions between surface winds and sand surfaces that can be characterized by aerodynamic roughness length has important implications for aeolian sediment transport. Here we report the wind tunnel results of systematic tests of the aerodynamic roughness lengths of different sized fixed sand beds at different free-stream wind velocities. The results suggest that there are complex variations of aerodynamic roughness length. Wind is an active factor in determining the aerodynamic roughness length of fixed sand beds. The so-called 1/30 law in aeolian research proposed by Bagnold more than a half century ago for sand surfaces is found to suffer some limitations. Some multivariate models relating aerodynamic roughness length and other factors are provided. The differences between the aerodynamic roughness of fixed and mobile beds are discussed, and it is implied that mobile surfaces adapt to the wind by changing roughness. The similarities between aerodynamic roughness and hydraulic roughness are compared. Aerodynamic roughness is divided into aerodynamically smooth, transitional, and completely rough regimes.

Origins of Groundwater Inferred from Isotopic Patterns of the Badain Jaran Desert, Northwestern China

There are many viewpoints about the sources of groundwater in the Badain Jaran Desert (BJD), such as precipitation and snowmelt from the Qilian Mountains (the upper reaches [UR] of the Heihe River Basin [HRB]) and precipitation from the BJD and the Yabulai Mountains. To understand the source of the groundwater of the BJD and their possible associations with nearby bodies of water, we analyzed variations of stable isotope ratios (δD and δ(18) O) and the deuterium excess (d-excess) of groundwater and precipitation in the BJD, of groundwater, precipitation, river and spring water in the UR, and of groundwater and river water in the middle and lower reaches (MR and LR) of the HRB. In addition, the climatic condition under which the groundwater was formed in the BJD was also discussed. We found obvious differences in δD, δ(18) O, and d-excess among groundwater in the BJD, nearby water bodies and the HRB. The groundwater δD-δ(18) O equation for the BJD was δD = 4.509δ(18) O-30.620, with a slope and intercept similar to that of nearby areas (4.856 and -29.574), indicating a strong evaporation effect in the BJD and its surrounding areas. The equations slope of the BJD was significantly lower than those of HRB groundwater (6.634), HRB river water (6.202), precipitation in the BJD and Youqi (7.841), and the UR of the HRB (7.839). The d-excess (-17.5‰) of the BJD was significantly lower than those of nearby groundwater (-7.4‰), HRB groundwater (12.1‰), precipitation in the BJD (5.7‰) and in the UR of the HRB (15.2‰), and HRB river water (14.4‰). The spatial patterns of δ(18) O and d-excess values in the BJD suggest mixing and exchange of groundwater between the BJD and neighboring regions, but no hydraulic relationship between the BJD groundwater and water from more distant regions except Outer Mongolia, which is north of the BJD. Moreover, we conclude that there is little precipitation recharge to groundwater because of the obvious d-excess difference between groundwater and local precipitation, low precipitation, and high evaporation rates. The abnormally negative d-excess values in groundwater of the BJD indicate that this water was formed in the past under higher relative humidity and lower temperatures than modern values.