Targeted artificial ocean cooling to weaken tropical cyclones would be futile

The study addresses whether targeted artificial cooling of the upper ocean can meaningfully weaken a tropical cyclone (TC) approaching land. Prior interest in geoengineering arises from the need to reduce TC hazards, with proposals ranging from solar radiation management to cloud seeding. Despite historical attempts such as Project STORMFURY and recent patents proposing artificial ocean cooling via vertical mixing, no successful weakening of TCs has been demonstrated. The governing theories of TC intensity (WISHE and MPI) link intensity to air–sea heat fluxes; while TCs naturally induce cold wakes that can reduce intensity, these effects may be insufficient where ocean heat content is high or barrier layers exist. The central question posed is whether artificially cooling the upper ocean ahead of landfall can produce appreciable reductions in TC intensity when targeting individual storms (weather modification), recognizing the immense scales involved and potential unintended consequences.

The paper reviews prior geoengineering concepts: solar radiation management (SRM) including stratospheric aerosol injections and marine cloud brightening, noting potential regional benefits but substantial risks and uneven outcomes. Historical weather modification efforts (Project STORMFURY) failed to produce attributable TC weakening. Several patents have proposed artificial ocean cooling via vertical mixing of sub-thermocline waters, yet none have shown proof of success, though some private efforts persist. Foundational theories (WISHE and MPI) describe TC dependence on air–sea fluxes; natural TC-induced ocean cooling (cold wakes) can limit intensity but is mitigated in regions with high ocean heat content, warm-core eddies, or ocean barrier layers. Prior studies highlight how mesoscale ocean features (warm/cold core eddies, barrier layers, river plumes) and integrated kinetic energy better relate to damages than peak winds, motivating evaluation metrics beyond maximum sustained wind.

Two complementary approaches were used. (1) Analytical framework: An ocean-aware potential intensity (OPI) model was used to extend classical MPI by incorporating TC-induced ocean mixing and associated SST cooling. MPI/OPI were computed with the PyPI Python package, using ERA5 monthly-averaged Gulf of Mexico September soundings (2010–2020) for temperature and moisture profiles. The authors modified PyPI to implement OPI per Miyamoto et al. The workflow: compute standard MPI, derive maximum potential winds, then update SST via the OPI SST-mixing equation using prescribed mixed layer depth (MLD), translation speed, drag coefficient (CD=2.5×10^-3), size parameter A=66.33 km, and initial temperature discontinuity ΔT=0. Recompute MPI with updated SST to obtain OPI. To assess artificial cooling, base-state SSTs were reduced by 0.25–4 °C and OPI recomputed to quantify increases in minimum central pressure (Pmin). Parameters explored included initial SST, MLD, and translation speed. (2) Realistic mesoscale simulations: Idealized, three-dimensional WRF-ARW v4.2.1 simulations of TCs approaching land were conducted. A control (NC) case used a weak initial vortex (Pmin=1005 mb) in mean easterly flow via point-downscaling, with ERA5-based Gulf of Mexico September profiles, low shear, and uniform environmental conditions. The TC translated due west at 4 m s^-1 ("slow") or 8 m s^-1 ("fast"). Artificially cooled ocean patches were imposed with deepened and cooled mixed layers adjacent to the coast, with along-coast length fixed at 720 km and cross-shore widths of 90, 180, and 360 km (W90, W180, W360). An additional 90 km patch placed farther offshore (W90E) tested earlier exposure. Identical patch configurations were repeated for fast translations. Simulations assumed instantaneous availability of cooled waters and an initially quiescent ocean. Intensity metrics included minimum central pressure Pmin and integrated kinetic energy of tropical-storm-force winds (IKETS) as a proxy for storm surge potential. Time series were analyzed over the 48 hours before landfall, and weakening relative to each case’s lifetime maximum intensity (LMI) and to the corresponding NC case was quantified.

• Analytical OPI framework: OPI is consistently lower than classical MPI due to self-induced SST cooling. OPI increases with higher base SST, deeper MLD, and faster translation speed because wind-induced mixing cools less effectively under those conditions. When imposing artificial cooling of the mixed layer, OPI predicts the largest weakening for environments with high SSTs, deep MLDs, and fast storm motion. For a representative Gulf of Mexico scenario (MLD=60 m, translation speed=4 m s^-1), OPI suggests nearly 60 hPa weakening is theoretically possible via mixed-layer cooling; further weakening is predicted if cooling extends to 100 m depth or if the storm moves faster. - Realistic WRF simulations: Practical impacts were far smaller and required massive cooled regions. For slow-moving storms (4 m s^-1): • Without artificial cooling, natural TC-induced cooling already weakened storms by >24 hPa in the last 12 h pre-landfall. • Additional weakening due to artificial cooling at landfall was small for narrow patches: W90 and W180 achieved only 3.5 and 6.4 hPa extra weakening (3.7% and 6.8% relative to NC), respectively. • The widest patch (W360) yielded 14.4 hPa extra weakening, corresponding to a 15.2% reduction relative to the NC slow case. The W90E patch triggered temporary weakening earlier but intensification resumed over warmer waters, yielding nearly identical landfall intensity to W90. For fast-moving storms (8 m s^-1): absolute weakening at landfall was smaller due to shorter residence over the cooled patch, but percent reductions relative to the NC fast case were larger, consistent with OPI expectations; the widest patch (W360 Fast) achieved a 15.6% reduction. - Energy and scale requirements: The largest cooled region evaluated encompassed a surface area of 2.6 × 10^5 km^2 and volume of 2.1 × 10^4 km^3. Estimated ocean heat content extracted from the cooled patches was 2.8 × 10^18, 1.1 × 10^19, and 4.5 × 10^19 kJ for increasing patch sizes—comparable to the solar energy incident on Earth in about 10 minutes and far exceeding national annual energy use (US 2019 total ~1 × 10^17 kJ). - Hazard-relevant metric: Reductions in IKETS were even smaller than those in Pmin, indicating limited benefit for storm-surge-related destructive potential; e.g., percent reductions in IKETS for cooled cases were modest, and W90E did not improve landfall IKETS versus W90 despite earlier weakening. Overall, theory provides an upper bound not achievable in practice; meaningful weakening (≈15%) required unrealistically large, ideally cooled regions.

While the OPI-based theoretical analysis indicates that artificial ocean cooling could substantially reduce TC potential intensity in environments with high SST, deep mixed layers, and rapid translation speeds (where natural wind-driven SST cooling is less effective), the realistic WRF simulations demonstrate that achieving meaningful weakening near landfall requires artificial cooling over enormous ocean areas and volumes. Moreover, for cases where OPI suggests the greatest potential benefit (deep MLD and fast motion), even larger cooled coverage is needed due to reduced residence time over cooled waters. Given the idealized assumptions (instantaneous cooling, quiescent ocean, uniform favorable atmosphere), the simulated reductions likely overestimate real-world impacts. Consequently, targeted artificial ocean cooling yields limited, short-lived benefits and is operationally impractical for disaster mitigation compared to alternative strategies. The findings suggest that resources would be more effectively invested in improved forecasts, warning systems, resilient infrastructure, and enhanced observations.

The study combines an ocean-mixing-aware potential intensity framework (OPI) with high-resolution WRF simulations to evaluate targeted artificial ocean cooling as a TC mitigation strategy. OPI predicts large theoretical weakening under certain oceanic and kinematic conditions, but realistic simulations show that even under idealized best-case scenarios, only modest (~15%) reductions at landfall are achievable and require impractically vast cooled regions and energy redistribution. Therefore, targeted artificial cooling to weaken TCs is not a viable operational strategy. Future efforts should prioritize improving TC forecasting (including rapid intensification predictability), enhancing resilient coastal infrastructure, and advancing observational technologies. Further research could test broader environmental regimes and coupled ocean–atmosphere–biology impacts to better quantify risks and unintended consequences.

• Idealized ocean: Simulations assumed an initially quiescent ocean; real coastal oceans have strong boundary currents and mesoscale eddies that create complex temperature and current fields, potentially negating or advecting away artificial cooling. - Idealized atmosphere: Favorable thermodynamic profiles (Gulf of Mexico September climatology) and low shear were assumed; realistic synoptic variability could diminish any benefits. - Regional sampling: Environmental profiles were from one region; while processes are broadly applicable, exact magnitudes may differ elsewhere. - Implementation assumptions: Instantaneous creation and maintenance of large cooled, deepened mixed layers were assumed; technological feasibility, logistics, and time scales of mixing (PRT model time scales) were not operationally addressed. - Metrics and exposure: Benefits were sensitive to storm translation speed and residence time over cooled waters; smaller patches offered only transient effects if followed by warmer waters. - Ecological and climate risks: Potential adverse effects include accelerated upper-ocean deoxygenation and poorly understood ecosystem impacts. - Predictability and targeting: Rapid intensification is difficult to predict and track shifts are common in weaker systems, complicating where and when to deploy cooling.