Understanding the turbulent transport of momentum and scalars such as
temperature, water vapor, and trace gases in the atmospheric boundary layer
plays important roles in many disciplines such as meteorology, hydrology, agriculture and air quality control.
It is often assumed that turbulence transports all scalars (temperature, water vapor, and CO2) similarly (Lewis analogy);
this similarity assumption is often also extended to include momentum, which is usually referred to as the Reynolds analogy.
The Lewis and Reynolds analogies have numerous applications in atmospheric boundary layer parameterizations and measurements but they do not necessarily always hold.
In a few papers (Li and Bou-Zeid, 2011; Li et al. 2012; Zhao et al. 2013; Wang et al. 2013),
we examined the dissimilarity between momentum and scalars, as well as the dissimilarity between active scalar and passive scalar.
Turbulence in sheared and thermally stratified boundary layers is affected by both shear and buoyant forces.
In the atmospheric surface layer (ASL), the impact of buoyancy resulting from surface cooling or heating is represented by the stability parameter, namely,
the ratio of height above the ground, z, and the Obukhov length, L. According to Monin-Obukhov Similarity Theory (MOST), turbulent statistics, when normalized correctly, should be only a function of z/L.
For example, the normalized mean velocity and temperature profiles are universal functions of z/L, which are usually called stability correction functions.
In our papers (Li et al. 2012, POF; Katul et al. 2013, PRE), we examined the origin of the universality of these functions and tried to link the Kolmogorov's theory, which describes to small scale turbulent structures, to the bulk similarity theory.
Urban climate and hydrology
We have been using the Weather Research and Forecasting (WRF) model to study urban climate and urban water cycle.
Specifically, we are focusing on improving and assessing the urban representations. For example, we implemented a tiling/mosaic approach
into the Noah land surface model in order to account for subgrid-scale heterogeneity effect that is ubiquitous in urban areas (see Li et al. 2013, JGR). We
also implemented a newly-developed urban canopy model (the Princeton Urban Canopy Model) into WRF, which betters captures the very small-scale
heterogeneity effect and hydrological processes that occur in urban areas (see Li and Bou-Zeid 2014, ERL).
We assess these improvements by comparing simulations to a variety of observational datasets, including radar rainfall estimates, eddy-covariance data,
sounding data, and satellite observations(see Li et al. 2013, JHM). We also utilize these new capabilities to study policy-relevant questions (see Li et al. 2014, ERL) and we also use WRF and WRF/Chem for other applications (see Hu et al. 2014, ACPD).
Climate change and the hydrological cycle
A central theme of our research is the impact of climate change and climate variability on the hydrological cycle. The primary framework that has been used is the Budyko curve.
The long-term mean annual evapotranspiration is usually recognized as to be controlled by water availability (usually represented by the long-term mean annual precipitation) and energy availability (usually represented by the long-term mean annual potential evaporation).
A semi-empirical curve that describes the relationship between the long-term mean annual precipitation (P), potential evaporation (Ep), and evapotranspiration (E)
is called the Budyko curve. We use the Budyko curve to understand the impact of vegetation (Li et al. 2013, WRR), interannual variability of precipitation and potential evaporation on evapotranspiration, including its long-term mean, its interannual variability (Li 2014, AWR) and its trend (Cong et al. 2014, HSJ).