Motivated and hardworking individuals, who are seeking graduate studies in HydSens Laboratory, are encouraged to contact Ardeshir Ebtehaj (e-mail: ebtehaj@umn.edu).

​

NEW OPPORTUNITY (Fall 2020 or Spring 2021) -Three Positions

​

Remote Singing of Snowfall (PhD position): Snowfall is the main input to the accumulation processes that control the mass balance of snowpack, glaciers, sea ice and ice sheets. However, space-time variability of high-latitude snowfall is one the least understood components of hydrologic water and energy cycle—largely due to the lack of long-term and sufficiently dense ground-based observations. Satellite data aims to close this knowledge gap. However, research suggests that the discrepancies between satellite and reanalysis snowfall data grow markedly over high-latitude snow-covered surfaces. This uncertainty is largely because the passive microwave scattering signals of snowfall and snowpack interfere and it is not yet well understood that how theses signatures can be separated under different land and atmospheric boundary conditions.

​

To advance our understanding of hydrologic water and energy cycle over polar climate regimes, we invite applications for one PhD position at the Saint Anthony Falls Laboratory, University of Minnesota. The main goal is to reduce the uncertainty of passive microwave retrievals of snowfall and snowpack over sea and land ice. The main objectives of the project are (i) to investigate radiometric interactions of snow-covered ice and snowing atmosphere over frequencies 10-200 GHz, using coupled land-atmosphere radiative transfer models; (ii) to quantify uncertainty of radiative transfer modeling using passive/active satellite data; and (iii) to develop a new class of microwave Bayesian retrieval algorithms for simultaneous retrieval of snowfall and snow-cover physical properties. To apply, please contact Ardeshir at ebtehaj@umn.edu. 

​

Data Assimilation (Postdoc position): Classic Data Assimilation (DA) techniques that rely on Gaussian assumptions often penalize the model and observation errors on the Euclidean space. The affine geometric structure of this space is unable to formally account for non-Gaussian state space and systematic biases in land and weather models. Our research team has been developing a new variational DA paradigm over the Wasserstein space, whereby optimal mass transport theory promises to extend the geophysical forecast skills under non-Gaussian state-spaces and systematic errors. The Wasserstein metric is geodesic and thus enables assimilation in a space of all sufficiently smooth and square-integrable probability density functions, leading to full recovery of non-Gaussian forecast probability distributions. Unlike Eulerian penalization of error in the Euclidean space, the Wasserstein metric is Lagrangian and can capture the translation of probability measures, enabling to formally penalize geophysical biases. We aim to extend our research and apply the new framework to improve forecast skills of soil moisture and precipitation using the Noah-MP land surface model and the Weather Research and Forecasting Model. To that end, we are looking for a motivated individual with a background in DA. The postdoc position is for one year and is extendable to two years.

​

Remote Singing of Cryosphere (PhD position): To understand changes of cryosphere, we first need to measure them from space as prioritized by the 2017 Decadal Survey. However, there are still significant knowledge gaps in our algorithmic tools for passive microwave remote sensing of cryosphere and its overlying atmosphere. Inversion of land-atmosphere radiative transfer (RT) models is an ill-posed inverse problem, over surfaces that are covered with frozen water, largely because different state variables give rise to the same observed brightness temperatures. Currently, retrieval algorithms cannot handle this non-uniqueness properly and take full advantage of all available frequency channels. Land algorithms ignore the information of high-frequencies where the atmospheric signal is strong and atmospheric algorithms consider the information of low-frequencies as a background noise. This algorithmic paradigm becomes highly uncertain over cryosphere, where surface and atmospheric signatures are strongly mixed across a wide range of frequencies. The main reason is that, currently, there is no formal way to adaptively weight the information content of all channels based on their error structure and separate contribution of surface and atmosphere in the observed brightness temperatures. The goal of this research is to advance the knowledge in inversion of land-atmospheric RT models in microwave bands over radiometrically complex surfaces. The outcomes should lead to a new algorithmic paradigm that enables to use data from a single radiometer and simultaneously retrieves both cryosphere and atmospheric state variables with reduced uncertainty.

Current Projects:

​

1. Remote Sensing and Super-resolution Imaging of Microplastics in Surface Waters (LCCMR), 2021-2024 (PI)

2. Metric Learning for Joint Inversion of Land-atmosphere Radiative Transfer Equations: Improved Microwave Remote Sensing of Cryosphere and Atmosphere (NASA), 2020-2023 (PI)

3. Improving Passive Microwave Retrieval of Snowfall and Snowpack on Ice-covered Surfaces, National Aeronautics and Space Administration (NASA), 2020-2023 (PI).

4. Physically Constrained Inversion of the First-order Radiative Transfer Equations for High-resolution Retrievals of Soil Moisture and Vegetation Water Content using SMAP Data, National Aeronautics and Space Administration (NASA), 2019-2022 (PI).

5. Reducing Uncertainties in GPM Snowfall Retrievals: Applications for Improved Prediction of Snowstorms, National Aeronautics and Space Administration (NASA), 2018-2021(PI).

6. Robust Variational Data Assimilation under Incomplete and Inaccurate Data: Extremes, Biases, and Observability in Joint Assimilation of Satellite Precipitation and Soil Moisture, National Aeronautics and Space Administration (NASA), 2018-2021(PI).

7. Soil Moisture Super-resolution and Regularized Data Assimilation: Algorithms and Hydro-agronomic Application in SMAP Era, National Aeronautics and Space Administration (NASA), 2016-2019 (Co-PI). 

8. Advanced Inversion Algorithms for GPM Passive Microwave Retrievals and Multi-sensor Merging, National Aeronautics and Space Administration (NASA), 2016-2019 (Co-PI).