03/28/2023
By Ali Fallahmaraghi

The Kennedy College of Sciences, Department of Environmental, Earth and Atmospheric Sciences, invites you to attend a Master’s thesis defense by Ali Fallahmaraghi on “Flash Droughts: The Role of Near Real Time Vegetation Information and Definition Uncertainty.”

Candidate Name: Ali Fallahmaraghi
Degree: Master’s
Date: Friday, April 7, 2023
Time: 10 a.m. to 1 p.m.
Location: Olney Science Center Room 312, North Campus

Thesis/Dissertation Title: Flash Droughts: The Role of Near Real Time Vegetation Information and Definition Uncertainty

Committee:

  • Advisor Christopher Skinner, Department of Environmental, Earth and Atmospheric Sciences, University of Massachusetts Lowell
  • Mathew Barlow, Department of Environmental, Earth and Atmospheric Sciences, University of Massachusetts Lowell
  • Frank Colby, Department of Environmental, Earth and Atmospheric Sciences, University of Massachusetts Lowell


Brief Abstract:

Flash droughts (FDs) are droughts with rapid onset and intensification which are difficult to predict and hence lead to substantial impacts on crop production and water resources. Considerable uncertainty in our understanding of FD exists in part because of uncertainty in the physical drivers of the events, and because, like other forms of drought, there are many ways to define their onset, intensity, and termination. A common way to monitor flash drought occurrence is to examine changes in root-zone soil moisture. Given a scarcity of continuous, widespread soil moisture measurements, land surface model datasets such as the National Land Data Assimilation System (NLDAS), provide a key process-based estimate of soil water and flash drought monitoring across the continental United States. However, biases in the representation of vegetation within NLDAS land surface models can impact the fidelity of soil moisture estimates and therefore of flash droughts. To assess the impact of these vegetation biases, we examine the influence of near real-time satellite vegetation assimilation in NLDAS on the simulation of soil moisture and flash drought characteristics. Specifically, we compare a simulation with the traditional operational NLDAS, which includes prognostic vegetation, with a simulation in which satellite-derived leaf area index (LAI) measurements are assimilated into the NLDAS land model. Using the new LAI assimilation dataset, we find flash droughts are most common in the Southwest, Great Plains, and Southeast, Northeast and northern Michigan. The comparison between simulations indicates that biases in vegetation within NLDAS lead to changes in the onset and termination dates of flash droughts, such that the operational prognostic vegetation NLDAS system has shorter duration flash drought events that often do not transition to traditional longer lasting drought events. Additionally, in the DA simulation, the number of years with a flash drought has been increasing across much of the U.S., especially in the far southern and northern Great Plains, Midwest, and Southwest U.S. while OL simulation shows a negative trend in much of the southern and northern Great Plains, in disagreement with DA. Overall, flash drought characteristics in the simulation with LAI assimilation agree slightly better with other estimates of flash drought occurrence, including those from the United States Drouth Monitor (USDM). Together, the findings of this study shed light on the processes that enable more accurate flash drought detection and help to reconcile differences in flash drought definition.