Event Time: June 23 - August 29, 2025

 

Summary of 2025 JIFRESSE Summer Internship Program (JSIP) achievements

Eight UCLA undergraduates delivered impactful, publication-ready work spanning atmosphere–ocean fluxes, space propulsion, magnetospheric forecasting, cosmology, Olympic-scale hazard readiness, climate data integrity, auroral retrievals, and CMIP7 tipping-point design. Below is a concise roll-up of contributions and outcomes.

 

1. The Role of Planetary Boundary Layer Height in Ocean Surface Flux Variability

Student: Annie Feng | Mentors: Shakeel Asharaf (JIFRESSE), Baijun Tian (JPL)

What they did: Compared PBL height (PBLH) from AIRS vs. ERA5 and quantified how PBLH discrepancies propagate into latent/sensible heat fluxes using COARE 3.5 and a decomposition sensitivity model.

Key findings:

• AIRS–ERA5 PBLH differs by ~0.8–1.0 km; however, COARE-derived flux differences are small because the bulk algorithm depends primarily on near-surface wind (U) and thermodynamic gradients (Δq, Δt).
• Sensitivity maps show U and Δq dominate LHF response; transfer-coefficient pathways are comparatively minor.

Impact: Highlights a pathway to upgrade bulk algorithms to account more explicitly for PBLH, reducing uncertainty in air–sea fluxes used by weather and climate models.

 

2. Advanced Propulsion Concepts for Deep Space Exploration (Laser Ablation)

Student: Pavel Shafirin | Mentors: Artur Davoyan (UCLA), Dan Goebel (JPL)

What they did: Conducted lab experiments and mission-scale simulations for laser-ablation spacecraft, measuring momentum coupling and simulating beamed-energy trajectories.

Key findings:

• Experimental Cm = 60–130 N/MW at up to 70 kJ/cm.; modeled systems show a 10 MW laser + 30 m telescope can accelerate 100 kg to ~32.6 km/s (500 kg to ~17.5 km/s).
• Enables sub-year Jupiter transfers (~0.47 y for 100 kg) and fast flybys of TNOs/extrasolar objects; identifies an optimal Cm for fuel utilization.

Impact: Demonstrates feasible architectures for rapid outer-planet and interstellar-object missions; frames concrete performance trades for future beamed-propulsion testbeds.

 

3. Modeling Ring Current Proton Fluxes with a Multilayer Perceptron

Student: Shrika Andhe | Mentors: Jinxing Li (UCLA/JPL), Jonathan Jiang (JPL)

What they did: Built an ANN to predict 14 energy levels (45–598 keV) of ring-current proton fluxes using Van Allen Probes data and geomagnetic indices (Sym-H, AsyH, SymD, SME).

Key findings:

• Achieved R. > 0.7 with low MSE across energies; robust long-term prediction behavior.
• Clear association: Sym-H drops co-occur with proton flux dips, supporting space-weather nowcasting.

Impact: A practical forecasting tool that can inform satellite operations and HF/GNSS users; extensible for 2019–2025 prediction using Kyoto & SuperMAG inputs.

 

4. Quasar Contributions to the Extragalactic Background Light (EBL)

Student: Danielle Timm | Mentors: Jordan Mirocha (JPL), Tzu-Ching Chang (JPL), Steven Furlanetto (UCLA)

What they did: Calibrated a quasar luminosity–halo mass relation via MCMC, applied mean quasar SEDs, and used the ares halo model to predict the EBL power spectrum.

Key findings:

• Although quasars are rare, their luminosity and clustering produce non-negligible EBL fluctuations, especially on large scales.
• SPHEREx detectability estimates suggest most emissivity remains unresolved, underscoring the need for statistical EBL analyses.

Impact: Provides an observation-ready framework for SPHEREx-era EBL decomposition; outlines refinements (e.g., double power law LFs, emission-line SEDs, LRDs).

 

5. Anticipating Compounding Weather Hazards for LA28

Student: Anneliese Phillips | Mentors: Colin Raymond (UCLA), Duane Waliser (JPL)

What they did: Built storyline-based extreme scenarios for urban heat, HAB-driven coastal water quality, and wildfire smoke, with operational recommendations for the 2028 Olympics.

Key findings & actions:

• Heat: venue-specific thermal mapping, shade/cooling logistics, and acclimation/cooling breaks cut heat-illness risk.
• HABs: shift from reactive sampling to multi-platform (satellite–airborne–in-situ) high-frequency monitoring with lane/venue adaptations.
• Wildfire smoke: probabilistic, months-ahead fire risk modeling (fuel load, dryness, fire weather), cross-agency dashboards, and multilingual public comms.

Impact: A playbook for anticipatory management, directly usable by Games planners and local agencies.

 

6. Temporal Shifts in Lower Tropospheric States (AIRS, ERA5, MERRA-2, JRA-55)

Student: Tyler Do | Mentors: Qing Yue (JPL), Hai Nguyen (JPL), Gang Chen (UCLA)

What they did: Regressed temperature/humidity anomalies (2002–2021) on major climate modes; performed changepoint detection (PELT, KernelCPD) to separate physical vs. operational shifts.

Key findings:

• Mean-value breakpoints (2014–2016) likely tied to extreme El Ni.o; no linear-trend breakpoints under stricter penalties.
• KernelCPD residual breakpoints trace to operational artifacts (e.g., AIRS solar-flare recalibration; AMSU anomaly 2018–19; MERRA-2 sea-ice input changes).

Impact: Delivers a robust QA roadmap for trend analyses—guarding against false climate signals in reanalyses/satellite records.

 

7. Advancing EZIE OSSE for Auroral Space Weather

Student: Jason Irie | Mentors: Frank Werner (UCLA), Jinxing Li (JPL)

What they did: Built a full L0→L2 pipeline: L0 screening, IGRF/NRLMSIS/MERRA-2 merged atmospheres, Stokes-axis alignment, and a custom Gauss–Newton retrieval for Bfield & temperature.

Key findings:

• Major runtime cuts (minutes → seconds) via vectorization/parallelization; pipeline runs on a standard laptop.
• Retrievals reproduce SuperMAG ground signatures; detect auroral electrojet perturbations from EZIE radiances.

Impact: A mission-ready toolchain that converts EZIE data into geophysically validated magnetic perturbations at scale.

 

8. Embedding Climate Tipping Points in CMIP7 (TipMIP)

Student: Megan Mathews | Mentors: Yu Gu (UCLA), Jonathan Jiang (JPL)

What they did: Designed a TipMIP suite within CMIP7’s Fast Track to probe AMOC, ice sheets, Amazon/boreal forests, and permafrost—aligned with NASA satellite diagnostics.

Key findings/proposal:

• Targeted overshoot and hosing experiments plus observation-matched variables (e.g., ICESat-2/GRACE-FO for ice, SWOT/Sentinel-6 for circulation, GEDI/VIIRS for biomass/fire, SMAP for permafrost).
• Deliver an observation-tested tipping-risk assessment for AR7 on an actionable 2025–2028 timeline.

Impact: A concrete path to policy-relevant thresholds, uniting CMIP7 multi-model power with NASA’s global observing system.

 

Overall, these students demonstrated full-stack research capability—from data engineering to synthesis—and produced actionable science aligned with JPL/UCLA priorities.