Observed and modeled interactions between droughts and heat waves for the Northeast US

 Drought and heat waves both have a range of severe societal and ecosystem impacts and often
share the characteristics of hot, dry days. The links and potential feedbacks between the two
phenomena, however, are not well understood and can vary regionally, seasonally, and relative
to the definitions used for both types of events. Moreover, there are complicated differences in
time and spatial scales between the two phenomena, adding further complexity to their
interactions. Investigating the relationships between droughts and heat waves through
observational and modeling analyses has the potential to improve mechanistic understanding,
subseasonal to seasonal (S2S) predictions, and climate projections of both phenomena.

The goal of the proposed work is to understand the interactions between droughts and
heatwaves, and assess current model skill in simulating and predicting these phenomena and
their interactions. We will use machine learning and moisture tracking techniques to objectively
identify and classify these events into various “types” based on daily circulation data, as a
foundation for dynamically-based investigation. We will then use these observed results as a
basis for assessing the predictability of the relationships and the ability of current climate models
to reproduce the relationships. We focus on the Northeast US warm season (May – Sep), when
heat waves have the largest absolute magnitudes and impacts. This focus allows for a detailed
examination of the underlying drivers of and interactions between the phenomena, and we expect
the techniques developed in the project to be directly applicable to other regions.

The primary scientific objectives are to: 1) Identify and investigate the characteristic daily
circulation patterns for droughts, heat waves, and their interactions, using the “machine learning”
techniques of Self Organizing Maps (SOMs) and K-Means Clustering (KMC) applied to a suite
of reanalysis data for the Northeast US warm season; 2) Identify and investigate the
characteristic moisture pathways and their relationships to circulation patterns for droughts, heat
waves, and their interactions via a set of moisture tracking methods, including Lagrangian back
trajectories and a climate model with integrated (online) water tracers; 3) Investigate the
medium-range and S2S predictability of the circulation patterns and their relationships, and
analyze the dynamical implications; 4) Examine the ability of CMIP6 climate models to
reproduce the key circulation patterns and relationships.

The proposed research is relevant to all three priority areas in the MAPP call by considering a
key process and potential feedback for drought – heat waves – and how they might relate to
extremes or onset of drought; by considering new methodologies, including drought analysis
based on daily circulation patterns identified via “machine learning;” and by considering
predictability and climate projections. The proposed work considers complex interacting
processes ranging from synoptic to seasonal timescales, and will provide the necessary
framework for an informed and practical assessment of S2S predictions and climate projections
for these processes, which are necessary to advance NOAA’s goal of addressing climate-related
societal challenges.

Climate Risk Areas: Extreme Heat, Water Resources