Disciplines: 

Atmospheric Science, Climate Science

Mentor (JPL): 

Colin Raymond

Email: colin.raymond@jpl.nasa.gov

Phone: 607-279-2752

Mentor (UCLA): 

Gang Chen

Email: gchenpu@ucla.edu

Phone: 310-206-9956

Background: 

Extreme heat is rapidly increasing in frequency and intensity across the globe; here in Southern California, multiple events just in the last several years have broken all-time local heat records. A unique combination of land-cover types, complex topography, and coastal influences make heat waves in the LA metropolitan area particularly varied (Hulley et al. 2019; Kalkstein et al. 2018), and this variation takes on different forms according to the dominant regional weather system as well as ocean and land preconditions (Shreevastava et al. 2023; Gershunov et al. 2021). A major outstanding question about this spatiotemporal diversity concerns the elevation profile of temperature during extreme heat events. Recent work has found that during heat waves the hottest nighttime temperatures are in the Hollywood Hills and other foothills such as the Crescenta Valley, areas which are heavily populated and which are normally among the cooler locales in the region (Shreevastava et al. 2023). The frequency, intensity, and diurnal timing of this phenomenon have not been assessed, nor has its predictability from large-scale meteorological parameters.

Description: 

The project will use high-resolution gridded products, in situ data from weather stations and radiosondes, and AIRS retrievals to analyze the elevation profile of temperature during extreme heat events, focusing on Southern California and over the diurnal cycle. The project will focus on building a catalogue of such events and developing a basic dynamical understanding of the principal processes that drive them. According to student interest and available time, the inquiry can additionally consider either the extent to which the hilltop/nighttime heatwave phenomenon may be widespread in regions with similar summertime climates, or local details, such as the events' different signatures in air temperature versus in land-surface temperature (from ECOSTRESS).

References: 

Gershunov, A., Morales, J. G., Hatchett, B., Guirguis, K., Aguilera, R., Shulgina, T., Abatzoglou, J. T., Cayan, D., Pierce, D., Williams, P., Small, I., Clemesha, R., Schwarz, L., Benmarhnia, T., and Tardy, A. (2021). Hot and cold flavors of southern California's Santa Ana winds: Their causes, trends, and links with wildfire. Clim. Dynam., 57, 2233-2248. https://doi.org/10.1007/s00382-021-05802-z.

Hulley, G., Shivers, S., Wetherley, E., and Cudd, R. (2019). New ECOSTRESS and MODIS land surface temperature data reveal fine-scale heat vulnerability in cities: A case study for Los Angeles County, California. Remote Sens., 11, 2136. https://doi.org/10.3390/rs11182136.

Kalkstein, A. J., Kalkstein, L. S., Vanos, J. K., Eisenman, D. P., and Dixon, P. G. (2018). Heat/mortality sensitivities in Los Angeles during winter: A unique phenomenon in the United States. Environ. Health, 17, 45. https://doi.org/10.1186/s12940-018-0389-7.

Shreevastava, A., Raymond, C., and Hulley, G. C. (2023). Contrasting intraurban signatures of humid and dry heatwaves over Southern California. J. Appl. Meteorol. Climatol., 62, 709-720. https://doi.org/10.1175/jamc-d-22-0149.1.

Student Requirements: 

The student should have some basic knowledge of atmospheric or climate sciences. Proficiency in Jupyter Notebook (Python), MATLAB, or other programming language is required.

Work Location: 

JPL and/or UCLA

Disciplines: 

Remote Sensing, Computer Science, Machine Learning, Data Science, Applied Mathematics

Mentor (JPL): 

Charles Marshak (charlie.z.marshak@jpl.nasa.gov)

Mentor (UCLA): 

Alexander Handwerger (alhandwerger@g.ucla.edu)

Background: 

Over the past decade, earth-orbiting synthetic aperture radar (SAR) satellites have generated unprecedented volumes of imagery of our planet. Yet, SAR has been an underutilized sensor for high-level, global applications because the necessary inputs and ancillary data to create analysis ready SAR imagery previously consumed significant disk space and required expert processing. These barriers have been effectively removed by the JPL Observational Products for End-Users from Remote Sensing Analysis (OPERA) project. OPERA is producing a freely available, analysis ready global SAR dataset called the Radiometric Terrain Corrected (RTC) SAR backscatter product. The product became available October 2023 and provides images of all landmasses (except Antarctica) every ~12 days. The RTC product is generated from Sentinel-1 SAR data and is a 30-meter GeoTIFF. The salient feature of the RTC product (in addition to being freely available and analysis ready) is that it is corrected for the impacts of topography. This means that any end-user can utilize these data to investigate physical properties of the ground, including vegetation conditions, land cover/land use, and can also look at how the ground surface conditions evolve over time due to natural or anthropogenic change or from processes such as wildfires and landslides.

Description: 

The objective of this project is to create a first-of-its-kind curated OPERA RTC SAR datasets to support rapid land cover classification and disaster response (specifically, wildfires and landslides). These SAR datasets will be collocated with known locations that have been accurately mapped and characterized by previous work. The JIFRESSE intern will lead the data curation and use machine learning, specifically pixelwise (semantic) segmentation and change detection methods. The results produced by the intern will be (1) training datasets and (2) a machine learning model that can be used to delineate landslide and wildfire extents and segment land cover types using the OPERA RTC SAR data. Finally, these training datasets and tools (i.e., Jupyter notebooks) will be made publicly available so that they can be streamed into common data science workflows. The open-data and

open-source deliverables will support NASA’s open-science vision.

References: 

https://www.jpl.nasa.gov/go/opera/products/rtc-product

Student Requirements: 

Python. Basic knowledge of Remote Sensing and Machine Learning.

Work Location: 

JPL, UCLA, or Remote

Disciplines: 

Atmospheric science, remote sensing, data mining

Mentor (JPL): 

Evan Fishbein (329E) Evan.Fishbein@jpl.nasa.gov, 818-825-6744, https://science.jpl.nasa.gov/people/fishbein

Mentor (UCLA): 

Tianhao Zhang (JIFRESSE), zhangth@g.ucla.edu

Yu Gu (JIFRESSE), gu@atmos.ucla.edu, 301-206-0377, http://people.atmos.ucla.edu/gu/

Background: 

Interpreting hyperspectral infrared and microwave spectroradiometer data is complicated by heterogeneity within observation field-of-views (FoV).  Traditional cloud-clearing algorithms use heterogeneity and correlation between FoVs and have used this to increase capabilities. This research studies the correlation across  FoVs, the sampling of microclimates within FoVs, and how Bayesian statistics  might be used in multi-FoV retrieval algorithms to account for heterogeneity.

Description: 

The candidate will develop and use tools to sample satellite sounder, imager and NWF data, use clustering algorithms to define microclimates and estimate spatial correlation within microclimates.

References: 

Aumann, H. H., et al. (2003), AIRS/AMSU/HSB on the aqua mission: Design, science objectives, data products, and processing systems, IEEE Trans. Geoscience Rem. Sens., 41, 253-264.

Netzel, P., and T. Stepinski (2016), On Using a Clustering Approach for Global Climate Classification, J Climate, 29(9), 3387-3401, doi:https://doi.org/10.1175/JCLI-D-15-0640.1.

Student Requirements: 

Experience using python and/or IDL.  Interest in learning atmospheric science, atmospheric remote sensing and data clustering/data mining.

Work Location: 

JPL and/or UCLA

Disciplines: 

Astrodynamics, systems engineering, attitude controls, propulsion

Mentor (JPL): 

Dan Goebel, dan.m.goebel@jpl.nasa.gov

Mentor (UCLA): 

Artur Davoyan, davoyan@seas.ucla.edu , 215 873 1640 dspace.ucla.edu

Background: 

Search for life is one of the central NASA’s objectives. The discovery of exoplanets suggest that other worlds might have conditions that can potentially harbor life. Beyond exoplanet detection further studies that would yield detailed spectroscopic analysis and potentially direct imaging are needed. For this purpose, a number of astrophysical missions (i.e., telescopes) have been proposed, including Habitable Worlds Observatory. These state of the art observatories will feature coronagraphs, that would block radiation from a host star enabling to perform spectroscopy of distant worlds. At the same time, due to rejection of a significant fraction of light by the coronagraph signal received is very weak, which puts fundamental constraints on the features and signatures that can be detected. Study of Erath–like planets orbiting close to their host stars necessitates 10¹⁰ signal contrast and together with high light throughput.

To circumvent the throughput constraints set by traditional coronagraphs, an external coronagraph – Starshade – was proposed in 2005. Starshade is a large occulter flying in formation with a space telescope designed to block the light of nearby stars in order to observe their orbiting exoplanets. Starshade has a potential to enable high signal contrast and high throughput. However, such formation flying presents a very costly and challenging mission, which is difficult to implement in the near-future.

In 2022 a hybrid observatory concept was proposed as NIAC study by John Mather. In it, a large ground-based observatory was coupled with a Starshade occulter flying on a high elliptical orbit. By synchronizing telescope operations and the Starshade orbital motion brief periods of time are possible for exoplanet observations. Such approach offers a low cost ready to implement solution. However, a relatively short observation time (set by mutual phasing between a ground based observatory, Earth orbital motion, and Starshade orbital motion) present a challenge.

This JSIP program aims to examine potential pathways to extend the observation times by exploring non-Keplerian station keeping.

Description: 

JSIP student will assist JPL and UCLA mentors to perform systems studies for establishing requirements on propulsion systems to maintain the Starshade in position for a prolonged periods of time.

First, a trade study on the Starshade orbit will be performed by taking into account exoplanet proper motion and ground based observatory motion. A range of orbital parameters will be determined. For the current hybrid observatory 170,000 km altitude is proposed to match ground telescope motion (400 m/s). We will examine orbital parameters that maximize Starshade operation time (i.e., period, eccentricity, inclination). Here orbital perturbations on Starshade motion will be accounted.

Next, we will examine the possible use of electric and solar radiation pressure propulsion for non-Keplerian augmentation of the Starshade station keeping. Specifically, we will examine if principles of solar sailing and/or electric propulsion can be used to maintain ~30 m diameter Starshade along telescope – exoplanet line for a prolonged time. Current concept of operations assumes that Starshade is operated at the apoapsis of a highly elliptical orbit. By using low-thrust propulsion it may be possible to maintain Starshade in position for a longer duration of time. Requirements on propulsion systems (i.e., size, mass and delta-V) will be established.

References: 

• HabEx Final report: https://www.jpl.nasa.gov/habex/pdf/HabEx-Final-Report-Public-Release-LINKED-0924.pdf

• NIAC summary:

https://www.nasa.gov/directorates/spacetech/niac/2022/Hybrid_Observatory_for_Earth_like_Exoplanets/

• Starshade tutorial – Doug Lisman:

https://www.jpl.nasa.gov/habex/documents/Aug2016/Lisman-Starshade.pdf

Student Requirements: 

Senior undergraduate student or PhD student

Work Location: 

UCLA and JPL

Disciplines: 

Atmospheric Science; Data Science

Mentor (JPL): 

Jonathan H. Jiang (329J); Jonathan.H.Jiang@jpl.nasa.gov; 818-207-8734; https://science.jpl.nasa.gov/people/JJiang/

Longtao Wu (398K); Longtao.wu@jpl.nasa.gov

Mentor (UCLA): 

Yun Lin (JIFRESSE); yunlin@g.ucla.edu; 

Yu Gu (JIFRESSE); gu@atmos.ucla.edu

Background: 

Cloud seeding, a technique used since the 1950s to enhance precipitation, has been employed in California to increase water resources in the face of drought and other climate-related challenges. Current cloud seeding operations hinge on specific weather conditions, such as cloud cover exceeding 50% and favorable wind direction and speed, etc. It’s the responsibility of the project meteorologist to verify these criteria. The entire process is rather subjective, and the effectiveness of cloud seeding is not quantified within it.

Description: 

In the project, we will develop a real-time forecasting and decision-making system using machine learning (ML) methods to forecast, detect, and recognize seeding opportunities in real time and assess the effectiveness of cloud seeding across different locations and seeding concentrations, providing quantitative guidance for cloud seeding operations.

We will use a spatio-temporal convolutional neural network (CNN) model. The ML model will be trained using predictors from MERRA-2 aerosols, GFS analysis and forecast data. Precipitation data from PRISM will be utilized as the predictant variable in this process. The performance of the ML model will be evaluated through comparisons with historical observations.

By modulating ice-nuclei aerosol concentrations across various locations, such as dust, which can act as proxies for cloud seeding aerosols, ensemble forecasts can be generated. The effectiveness of cloud seeding operations can then be quantified by analyzing discrepancies among the ensembles of the forecasted.

References: 

• Flossmann, A. I., Manton, M., Abshaev, A., Bruintjes, R., Murakami, M., Prabhakaran, T., & Yao, Z. 2019. Review of Advances in Precipitation Enhancement Research, Bulletin of the American Meteorological Society, 100(8), 1465-1480.
• Hunter, S.M., 2007. Optimizing Cloud Seeding for Water and Energy in California. CEC-500-2007-008. Prepared for the California Energy Commission.
• Rauber, R.M., B. Geerts, L. Xue, et al. 2019. Wintertime orographic cloud seeding—a review. J. Appl. Meteorol. Climatol. 58: 2117–2140.

Student Requirements: 

Strong background in atmospheric science and demonstrated skills in computer programming, familiar with meteorological datasets.

Work Location: 

UCLA Campus

Disciplines: 

Atmospheric Science; Data Science

Mentor (JPL): 

Mentor (UCLA): 

Yizhou Zhuang (Atmospheric & Oceanic Sciences),

zhuangyz@atmos.ucla.edu, https://www.zhuangyz.org

Rong Fu (Atmospheric & Oceanic Sciences), rfu@atmos.ucla.edu, https://dept.atmos.ucla.edu/rongfu

Background: 

Large-scale atmospheric circulation patterns have significant influence on regional climate conditions and are crucial in shaping periods of extreme weather, such as heat waves, droughts, wildfires, and heavy precipitation. Land surface feedbacks, such as those related to soil moisture and vegetation cover, play a crucial role in amplifying or mitigating the impacts of these atmospheric anomalies. Understanding the intricate relationship between land surface feedbacks and atmospheric circulation patterns is vital for the understanding of weather extremes and improving climate predictions.

Machine learning offers robust tools for untangling complex interactions within climate systems. For example, Self-Organizing Map (SOM) has shown promise in identifying leading patterns and mechanisms in high-dimensional datasets. Such tools can be leveraged to discern the impact of various land surface feedbacks associated with distinctive atmospheric circulation patterns, particularly in forecasting and analyzing conditions leading to extreme dry or wet anomalies.

Description: 

This project aims to utilize machine learning approaches, including SOM, to investigate the relationship between anomalous atmospheric circulations and surface meterological parameters such as precipitation and vapor pressure deficit, crucial indicators of drought and fire weather risk, and how land surface feedbacks could amplify these circulation induced surface anomalies.

We will start with analysis with circulation data from reanalysis datasets such as ERA5 and MERRA2, precipitation data from CPC and MSWEP, and surface variable data (e.g., soil moisture, SIF) from NLDAS, GLEAM, and satellite data products developed by JPL. The focus region will be the US Great Plains.

References: 

Zhuang, Y., Fu, R., & Wang, H. (2020). Large‐scale atmospheric circulation patterns associated with US Great Plains warm season droughts revealed by self‐organizing maps. Journal of Geophysical Research: Atmospheres, 125(5), e2019JD031460.

Erfanian, A., & Fu, R. (2019). The role of spring dry zonal advection in summer drought onset over the US Great Plains. Atmospheric Chemistry and Physics, 19(24), 15199-15216.

Wang, G., R. Fu, Y. Zhuang, P. A. Dirmeyer, J. A. Santanello, G. Wang, K. Yang, and K. McColl, 2024: Influence of Lower Tropospheric Moisture on Local Soil Moisture-Precipitation Feedback over the U.S. Southern Great Plain. Atmospheric Chemistry and Physics.   https://doi.org/10.5194/egusphere-2023-1897.

Student Requirements: 

Given the technical challenge, this project requires a Ph.D. student, preferably with a data science background, as specified by the support jointly provided by the UCLA Vice Chancellor for Research and JPL Data Science Working group.

Work Location: 

UCLA