Passive subambient cooling and atmospheric water nexus

Abstract

Accelerating global warming has intensified the need for sustainable, low-energy cooling strategies. Subambient radiative cooling is a compelling solution, passively dissipating heat to outer space via mid-infrared emission without external energy input. However, its performance depends on atmospheric conditions, especially low humidity and clear skies. Although its global potential is well-established, its performance under typical weather conditions and integration with sustainable water cycling technologies remain underexplored. This review examines how radiative cooling can be integrated with water-related technologies—including atmospheric water harvesting, sustainable agriculture, and radiative–evaporative cooling—to reduce thermal loads while improving water sustainability.

Rising global temperatures drive demand for sustainable cooling and water solutions. Here, authors review how passive radiative cooling can be integrated with water technologies to address heat and water stress in a warming climate.

Introduction

With accelerating global warming, ambient temperatures have consistently escalated in recent decades, setting increasingly alarming new records. According to the Copernicus Climate Change Service and the World Meteorological Organization, 2025 ranked among the warmest years on record globally, extending the recent sequence of exceptional heat¹. In addition, regional trends are equally concerning; for instance, the United Arab Emirates recorded extremes exceeding 50 °C in late May 2025². High-latitude regions also experienced anomalous heat, with parts of the Arctic Circle exceeding 30 °C for over two weeks—temperatures 8–10 °C above seasonal norms³. This thermal escalation has pushed active air conditioning (which relies heavily on evaporative cooling towers and high electricity loads)⁴ to unsustainable levels of resource consumption. Consequently, there is an urgent need for passive alternatives that decouple cooling from the exhaustion of finite freshwater supplies.

Among emerging strategies, daytime subambient radiative cooling (RC) has attracted considerable attention as a passive, energy-efficient alternative⁵. RC harnesses the natural process of thermal radiation, enabling surfaces to dissipate heat as mid-infrared (MIR) radiation (8–13 µm) through the atmospheric transparency window and thus emit thermal energy directly into outer space without external power input (Fig. 1a)^6,7. RC materials exhibit dual optical functionalities: they reflect most incoming solar radiation while enhancing thermal emittance in the MIR range^8–10. However, RC performance is intrinsically constrained by local atmospheric conditions. For instance, optimal subambient RC typically requires clear skies, minimal cloud cover, and dry atmospheric conditions with relative humidity (RH) preferably below ~40%¹¹; however, such conditions are not prevalent in most inhabited regions. Consequently, the existing literature largely demonstrates RC efficacy under idealized weather scenarios^12,13, while systematic assessments under realistic climatic constraints remain limited. This disparity presents a critical knowledge gap for reliable deployment depending on diverse local climatic conditions.

Fig. 1. Global cooling–water cycle nexus and theoretical analysis of sub-ambient radiative cooling (RC).

a Schematic of major physical components involved in the global cooling and water cycle nexus. RC technology enables passive surface cooling by emitting mid-infrared (MIR) radiation through the atmospheric transparency window (8–13 µm), which can be partially blocked by clouds. b Spectral features of ideal broadband (orange dashed curve) and selective coolers (green dashed curve). The yellow and blue backgrounds represent the solar spectrum and the MIR atmospheric transmittance spectrum, respectively. The 8–13 μm transparency window substantially overlaps with the thermal radiation spectrum of a 300 K blackbody (gray dashed curve). c Net cooling power of spectrally selective (green curves) and broadband MIR emitters (orange curves). All curves were calculated using a unified ambient temperature of 20 °C and the same atmospheric transmittance/sky-radiance spectrum shown in (b). Solid and dashed lines correspond to non-radiative heat transfer coefficients of h_com = 0 W m⁻² K⁻¹ and h_com = 8 W m⁻² K⁻¹, respectively.

This review addresses this gap by examining the nexus between subambient RC and atmospheric water, a strategically critical yet underexplored domain. The interplay between RC and water cycling offers advantages beyond passive thermal regulation, creating opportunities to enhance water sustainability, particularly in arid and surface water-stressed regions. We first outline the governing principles of subambient RC and its atmosphere-dependent constraints, then focus on the RC–water nexus through three representative application tracks: RC-enabled dew collection^14,15, RC-integrated sustainable agriculture¹⁶, and synergistic radiative–evaporative cooling¹⁷. Finally, we discuss cross-cutting opportunities and emerging concepts. By framing RC as an integrated thermal–water regulation platform, this review highlights its potential to simultaneously reduce cooling demands and alleviate water stress in water-limited climates, thereby supporting progress toward the United Nations Sustainable Development Goals (SDGs), particularly SDG 6 (clean water and sanitation) and SDG 7 (affordable and clean energy)¹⁸.

In contrast to prior reviews which generally discussed thermal-photonic emitter engineering¹⁰, materials innovation^19,20, device-system integration²¹, this topical review deliberately focuses on the specific interactions between subambient RC and water-related technologies.

Building upon a comprehensive literature survey, this review is structured to clarify three critical pillars:

• (i)Open scientific questions that currently limit predictive, local climate-aware design under realistic, non-idealized conditions.
• (ii)Evaluation baselines and deployment-relevant design rules essential for robust system-level engineering across diverse climatic zones.
• (iii)Dominant bottlenecks that impede the transition from laboratory proof-of-concept to scalable, industrial implementation.

By synthesizing these elements, we provide a physically grounded assessment framework that bridges fundamental thermodynamic understanding with practical, large-scale implementation. Ultimately, this review aims to help transform RC from a scientific principle into an emerging platform for global sustainability at the nexus of atmospheric water and passive thermal regulation.

Atmosphere-dependent cooling criteria

Herein, we examine the coupled radiative and non-radiative energy balance governing subambient RC under realistic atmospheric conditions, and translate these terms into climate-dependent operating limits for RC-enabled water processes.

Energy-balance criteria for subambient radiative cooling

Under practical outdoor conditions, the energy balance can be expressed as:

$$P_{net}=P_{\mathrm{rad}}-P_{\mathrm{atm}}-P_{\mathrm{sun}}-P_{\mathrm{non}-\mathrm{rad}}$$ 1

where P_rad denotes the outgoing thermal radiation from the object, P_atm is the absorbed downward atmospheric radiation, P_sun is the absorbed solar radiation, and P_non-rad denotes non-radiative heat gains by convection and conduction (Box 1)^9,22–24. To achieve subambient RC, a surface operating near ambient temperature must maintain positive net cooling power (i.e., P_net > 0), such that heat can be continuously rejected to an effectively cold sink (outer space) through the atmospheric window²⁵.

Box 1 Key components of subambient radiative cooling.

The performance of a radiative cooling (RC) system is determined by the net energy exchange between the emitter and its surroundings, as described by the steady-state energy balance in Eq. (1). The key thermal components include: outgoing thermal radiation from the emitter surface (P_rad), downward atmospheric radiation absorbed by the emitter (P_atm), absorbed solar irradiance (P_sun), and non-radiative heat exchange with the ambient (P_non-rad).

Outgoing thermal radiation (P_rad)

$$P_{\mathrm{rad}}=\int d\Omega \cos \theta {\int }_0^{\infty }d\lambda I_{BB}\left(T,\lambda \right)\epsilon \left(\lambda ,\theta \right)$$ B1

where symbols have the following meanings^22,23:

• $\int d\mathrm{\Omega }=2\pi {\int }_0^{\frac{\pi }{2}}d\theta \sin \theta$ represents the angular integral over a hemisphere.

• $I_{BB}\left(T,\lambda \right)=\frac{2h{c}^2}{{\lambda }^5}\frac{1}{e^{\frac{hc}{\lambda k_{B}T}}-1}$ is the spectral radiance of a blackbody at temperature T, where h denotes Planck’s constant, k_B represents the Boltzmann constant, c denotes the speed of light, and λ represents the wavelength.

• $\epsilon \left(\lambda ,\theta \right)$ is the emissivity of the cooler at wavelength λ and angle θ. Here, θ is defined as the angle between the solid-angle direction and the surface normal.

Downward atmospheric radiation (P_atm)

$$P_{\mathrm{atm}}=2\pi {\int }_0^{\frac{\pi }{2}}cos\theta sin\theta {\int }_0^{\infty }{I}_{BB}\left(T_{\mathrm{amb}},\lambda \right)\epsilon \left(\lambda ,\theta \right)\left(1-t{\left(\lambda \right)}^{\frac{1}{\cos \theta }}\right)d\lambda d\theta$$ B2

where symbols have the following meanings:

• T_amb represents the temperature of the ambient environment.

• t(λ) denotes the atmospheric transmittance along the zenith direction.

P_atm is inversely related to t(λ); thus, higher t(λ) results in lower P_atm, which favors subambient RC. However, t(λ) is strongly influenced by environmental conditions such as high water vapor content and cloud cover²⁴ because of enhanced absorption and scattering across the MIR spectrum. Consequently, substantial subambient RC has been demonstrated primarily in regions with dry air and clear skies²⁰.

Absorbed solar radiation (P_sun)

$$P_{\mathrm{sun}}=cos{\theta }_{\mathrm{sun}}{\int }_0^{\infty }\alpha \left(\lambda ,{\theta }_{\mathrm{sun}}\right)I_{\mathrm{sun}}\left(\lambda \right)d\lambda$$ B3

where symbols have the following meanings:

• $I_{\mathrm{sun}}\left(\lambda \right)$ represents the solar spectrum, defined by the AM 1.5 G standard (Fig. 1b).

• ${\theta }_{\mathrm{sun}}$ denotes the angle of incidence of incoming solar rays.

• $\alpha \left(\lambda ,{\theta }_{\mathrm{sun}}\right)$ refers to the object absorptance within the solar radiation wavelength range (0.25–2.5μm), which carries the vast majority of solar energy.

During daytime, P_sun becomes a dominant thermal load, particularly in equatorial regions where solar irradiance can exceed 1100 W m⁻². Achieving subambient RC under direct sunlight requires minimizing P_sun. To this end, emitters should exhibit low solar absorptance—i.e., high solar reflectance for opaque surfaces—across this spectral range, and this property must be critically measured⁷. Furthermore, atmospheric factors such as solar angle can reduce the amount of irradiance reaching the ground²⁵. In contrast, night-time RC benefits from the absence of solar irradiation (P_sun = 0), which simplifies material and system design^19,24.

Non-radiative heat exchange (P_non-rad)

$$P_{\mathrm{non}-\mathrm{rad}}=h_{\text{com}}\left(T_{\text{amb}}-T\right)$$ B4

where h_com is the non-radiative heat transfer coefficient between the object and its surrounding environment^22,23.

Mid-infrared emissivity and atmospheric transmittance

Both P_rad and P_atm scale with the surface emissivity ε(λ), so increasing broadband emissivity can simultaneously enhance emission and atmospheric absorption. Therefore, the central objective in MIR design is to maximize (P_rad−P_atm) by concentrating high emissivity within spectral regions where the atmosphere is most transparent—most notably the 8–13 μm window^6,7. Outside this window, water vapor and CO₂ absorption increase the effective sky temperature and strengthen atmospheric back-radiation, limiting subambient temperature reduction. Secondary windows (e.g., 16–22 μm) can contribute additional radiative power under dry skies but become negligible under humid conditions due to enhanced H₂O absorption^19,20. Consequently, spectral selectivity targeting 8–13 μm remains a primary route to maximize subambient cooling potential. By contrast, broadband emitters offer cost and manufacturability advantages but typically yield a higher minimum achievable equilibrium temperature, because high emissivity outside the 8–13 μm atmospheric window increases absorption of downward atmospheric longwave radiation^11–13.

Non-radiative heat exchange

In real outdoor environments, conduction and convection heat exchanges with the surroundings are inevitable. For subambient RC, the non-radiative heat gain (P_non-rad) is governed by the non-radiative heat transfer coefficient (h_com) and should be minimized^22,23. A common strategy for this minimization involves the use of infrared-transparent polyethylene (PE) covers, allowing radiative exchange with the sky while physically insulating the emitter surface from ambient airflow. Under wind-sheltered conditions, this configuration substantially lowers non-radiative heat exchange, yielding an effective h_com value of ~ 2–2.5 W m⁻² K⁻¹²⁶. For more economical cooling applications, open-surface configurations without bulky enclosures are preferable, but exhibit larger h_com⁶. Nevertheless, for surfaces with complex geometries, alternative or geometry-specific formulations may be required to accurately model convective and conductive heat transfer²⁷. Although these systems are inherently more vulnerable to convective and conductive heat exchange with the environment, subambient cooling remains feasible under mild wind conditions^5,28.

Selective and broadband emitters

This analysis considers the four main energy fluxes listed in Box 1 to define the net cooling power (P_net). To isolate the influence of emitter spectral characteristics, we initially evaluate an idealized scenario with no solar absorption (P_sun=0), negligible non-radiative exchange (h_com = 0), and an atmospheric transmittance profile based on the 1976 U.S. Standard Atmosphere (blue-shaded region in Fig. 1b).

Under idealized RC conditions, spectrally selective and broadband MIR emitters exhibit distinct net-cooling-power profiles as the emitter temperature decreases below ambient. From Fig. 1c, under ambient conditions (T = T_amb), a broadband emitter (orange curve) radiating over the 2.5–25 μm range initially achieves a net cooling power of ~90 W m⁻², slightly exceeding the value of ~83 W m⁻² achieved by a spectrally selective emitter optimized for the 8–13 μm atmospheric window (green curve). The broadband emitter’s higher initial performance is attributed to additional radiative emission in secondary atmospheric windows, such as the 16–22 μm band^6,19,20. However, this advantage is quickly offset as the surface temperature drops below ambient (i.e., subambient cooling). Due to enhanced absorption of longwave atmospheric radiation outside the primary 8–13 μm transparency window, the broadband emitter experiences a steep reduction in P_net (orange solid curve in Fig. 1c).

In contrast, the selective emitter maintains a more favorable net cooling gradient, enabling deeper subambient temperature reductions²⁹. As illustrated by the green solid curve displayed in Fig. 1c, under idealized conditions with perfect spectral selectivity and complete thermal isolation, such an emitter can theoretically achieve temperature reductions of up to 50 °C below ambient, compared with only ~19 °C for the broadband emitter. These findings emphasize the pivotal role of spectral engineering in maximizing subambient RC performance.

Although RC exhibits considerable potential under idealized conditions, its performance in real environments is limited by factors such as nonradiative heat transfer through convection and conduction, solar irradiance during daylight, and atmospheric opacity due to water vapor and cloud cover. To illustrate these effects, we consider a realistic scenario that includes convective losses. In an open-surface configuration exposed to moderate wind (V_wind ≈ 0.6 m s⁻¹, h_com = 8 W m⁻² K⁻¹)¹⁹, the attainable subambient temperature reduction decreases to 8.2 °C for the spectrally selective emitter and 6.9 °C for the broadband emitter (green and orange markers wherein the dashed curves intersect the x-axis in Fig. 1c). Notably, non-radiative losses can substantially offset the benefits of selective spectral engineering, especially under non-ideal atmospheric conditions. Nevertheless, modest subambient cooling can bring the surface temperature to the local dew point, enabling passive condensation of atmospheric moisture^14,15. This natural process—frequently forming dew on exposed surfaces during early morning and late evening—highlights a key synergy between RC and atmospheric water harvesting (AWH) during daylight hours.

In summary, although RC is a promising low-energy cooling strategy, its effectiveness remains highly sensitive to local weather patterns and atmospheric composition. Many proof-of-concept demonstrations have been performed under idealized conditions, such as clear skies and low humidity. However, the long-term, site-specific viability of RC technologies is intrinsically linked to regional climate characteristics³⁰. Consequently, the robust implementation of RC strategies needs material-level optimization and deployment frameworks accounting for climatic conditions^10,21. While most current studies passively leverage favorable local conditions, an alternative paradigm is to actively manipulate atmospheric parameters to enhance RC outcomes^14,15,31. In the subsequent sections, we explore an underexamined but critical weather-dependent factor, humidity, which offers an additional means of controlling and optimizing RC performance within broader environmental systems.

Global atmospheric moisture distribution

Key terms in Eq. (1)—particularly P_atm and atmospheric transmittance t(λ)—are strongly influenced by atmospheric moisture³¹. To contextualize this dependence, it is important to note that approximately 71% of Earth’s surface is covered by water bodies, including oceans, lakes, and rivers³². As depicted in Fig. 1a, solar radiation drives evaporation, converting surface water into atmospheric vapor. This vapor is then transported by atmospheric circulation, condenses to form clouds, and ultimately precipitates as rain, replenishing terrestrial water sources before returning to the oceans—thereby completing the global hydrological cycle²¹, which will affect the RC effect significantly. Therefore, accounting for this hydrological factor is essential while designing RC experimental characterizations, particularly for outdoor validation in specific regions.

Horizontal distribution of atmospheric moisture

The total atmospheric water mass accounts for ~0.25% of the total atmospheric mass, with a mean global distribution of ~26 kg m⁻²³³. Satellite-derived datasets reveal marked horizontal variability in atmospheric moisture content and its relevance for solar-driven AWH (Fig. 2a)³⁴. In arid and semi-arid regions, where atmospheric moisture is scarce because of limited surface water and low precipitation, conventional AWH methods are typically ineffective. Achieving efficient water capture in such settings calls for alternative strategies, such as RC-based condensation systems^14,15. RC operates most effectively in dry, cloud-free climates (global RC power distribution in Fig. 2b); however, these optimal RC zones often do not coincide with regions richest in harvestable atmospheric moisture (Fig. 2a). This spatial misalignment underscores the need to differentiate between horizontal atmospheric moisture availability and vertical atmospheric transparency while assessing site-specific feasibility.

Fig. 2. Global potential analysis of atmospheric water harvesting and radiative cooling (RC)-induced condensation.

a Mean daily water yield from solar-driven atmospheric water harvesting using different active sorbent device types³⁴. b Global RC potential⁵. c Atmospheric transmittance spectra t(λ) as a function of ambient temperature (T_amb) and relative humidity (RH), wherein t(λ) decreases with increasing T_amb and RH. d RC-induced temperature reduction for an ideal selective emitter as a function of T_amb and RH, calculated using a representative atmospheric transmittance spectrum based on the US Standard Atmosphere 1976 and a non-radiative heat transfer coefficient of h_com =  10 W m⁻² K⁻¹. e Temperature difference between ambient temperature and the dew point (T_amb−T_dewpoint) as a function of T_amb and RH. f Conditions enabling water harvesting, where the surface temperature (T_surf) drops below the dew point (T_surf<T_dewpoint). The shaded regions indicate the suitable ranges of temperature and RH for h_com = 5 W m⁻² K⁻¹ and h_com = 10 W m⁻² K⁻¹, respectively. Panel a adapted from Ref. ³⁴, CC BY 4.0. b Adapted with permission from Ref. ⁵, AAAS.

Vertical distribution of atmospheric moisture

While RC fundamentally relies on thermal emission through the 8–13 μm atmospheric window, where water vapor absorption is relatively weak under dry conditions, its performance deteriorates sharply in humid environments. As atmospheric water vapor concentration increases, rotational–vibrational absorption bands of H₂O molecules intensify, particularly near 8–8.5 μm and 12.5–13 μm⁶, effectively narrowing the usable portion of the window. The resulting spectral opacity diminishes the emitter’s ability to radiate heat into outer space, thereby reducing net cooling power.

A useful measure for characterizing this humidity constraint is total precipitable water (TPW)³⁵, defined as the vertically integrated column of atmospheric water vapor and expressed as:

$$\mathrm{TPW}={\int }_{z=0}^{z_{top}}{\rho }_{\nu }\left(z\right)dz$$ 2

where ρ_v(z) represents water vapor density at altitude z, from the surface (z = 0) to the upper atmosphere (z = z_top). Because >90% of atmospheric moisture is concentrated within the lowest 3 km of the troposphere, TPW shows strong spatial and seasonal variability. As shown in Fig. 2c, increasing TPW from 6.4 kg m⁻² (10 °C, 30% RH) to 66.1 kg m⁻² (30 °C, 90% RH) markedly reduces MIR transmittance, thereby substantially limiting RC performance³⁶.

To evaluate the influence of atmospheric humidity on RC performance, several studies have employed simulations of ideal selective emitters under varying environmental conditions^15,24. These analyses have typically assumed negligible solar absorption (P_sun = 0) and a moderate nonradiative heat transfer coefficient (h_com = 10 W m⁻² K⁻¹, corresponding to V_wind ≈ 1.2 m s⁻¹ for open-surface scenarios)⁶. TPW-resolved atmospheric transmittance was incorporated into the estimation of P_atm, enabling a direct assessment of cooling potential across different humidity levels. As illustrated in Fig. 2d, increasing RH leads to a notable reduction in the maximum achievable subambient temperature drop, defined as the difference between ambient temperature and surface temperature (T_amb−T_surf). For example, at T_amb = 20 °C, the maximum temperature drop decreases from 6.2 °C at 50% RH to 4.1 °C at 90% RH. The simulated trends closely match experimental observations, highlighting atmospheric moisture as a key control on atmospheric transmittance and radiative cooling performance²⁴.

Although elevated humidity limits RC efficiency by reducing MIR transparency, it simultaneously raises the dew point temperature, thereby lowering the temperature drop required for condensation. This dual role of atmospheric moisture establishes a fundamental radiative–condensation trade-off, as detailed below.

Subambient radiative cooling–condensation trade-off

The dew point, determined by ambient temperature and RH, defines the temperature threshold for condensation²⁴

$$T_{\mathrm{dewpoint}}=\frac{243.12\times \left[ln\left(\mathrm{RH}\right)+\frac{17.62\times T_{amb}}{243.12+T_{amb}}\right]}{17.62-\left[ln\left(\mathrm{RH}\right)+\frac{17.62\times T_{amb}}{243.12+T_{amb}}\right]}$$ 3

As illustrated in Fig. 2e, at T_amb = 20 °C, the temperature drop (T_amb −T_dewpoint) required to initiate condensation is 11 °C at 50% RH but only 2 °C at 90% RH, clearly demonstrating the RH dependence of dew formation under different environmental conditions. By integrating RC performance (Fig. 2d) with dew point threshold (Fig. 2e), Fig. 2f maps the operating regions (colored zones) wherein the surface temperature drops below the dew point (T_surf<T_dewpoint), enabling condensation. For instance, at T_amb = 20 °C, condensation occurs only when RH exceeds ~71% for a selective emitter with h_com = 10 W m⁻² K⁻¹.

In addition to atmospheric conditions, the non-radiative heat transfer coefficient (h_com) serves as a key parameter influencing the achievable subambient temperature drop and the temperature–humidity threshold for condensation^22,23. Lowering h_com from 10 to 5 W m⁻² K⁻¹ suppresses parasitic convective/conductive heat gains, enabling a lower T_surf and thereby reducing the critical RH required for T_surf<T_dewpoint (Fig. 2f), underscoring the importance of thermal insulation and heat management in practical condensers. While T_dewpoint is set by equilibrium thermodynamics (RH and T_amb), achieving T_surf<T_dewpoint in practice is governed by coupled radiative and non-radiative heat transfer together with vapor transport under local boundary-layer conditions. Critically, atmospheric moisture imposes an inherent trade-off: higher RH raises T_dewpoint and facilitates condensation, whereas higher TPW increases atmospheric back-radiation and thus reduces P_net. Consequently, RC–water performance is ultimately bounded by local environmental conditions, non-radiative heat exchange, and latent-heat requirements. In the next section, we will use this unified energy–balance framework to interpret each RC–water track in terms of its climatic operating window and realistic performance limits.

Passive cooling–water nexus

Here we discuss three parallel RC–water application tracks: RC-enabled dew collection, RC-integrated sustainable agriculture, and synergistic radiative–evaporative cooling, highlighting how each track tunes the radiative/solar/non-radiative terms under their climatic operating windows and key design constraints.

Radiative-cooling-enabled dew collection

In arid and semi-arid regions, dew collection can provide a complementary water source, sustaining ecosystems and human communities through extended dry periods. As global freshwater scarcity intensifies, RC-enabled dew collection is attracting growing interest for its passive energy profile, low environmental footprint, and strong suitability for deployment in water-stressed areas^14,15.

Design principles

Condensation remains the most widely explored and direct strategy for dew collection, for which the collecting surface temperature must fall below the local dew point (i.e., T_surf<T_dewpoint). Achieving this criterion requires surfaces with high solar reflectance and near-unity emissivity across the 8–13 µm atmospheric window, thereby maximizing (P_rad–P_sun) and providing sufficient P_net (Fig. 3a)¹⁵. In practical deployments, sustaining P_net further demands suppressing parasitic nonradiative heat gains (P_non–rad) by lowering h_com (e.g., via infrared-transparent covers)^22,23. In contrast to other AWH approaches which require complex architectures, sorbent cycling or continuous electrical power (e.g., fog harvesting³⁷, sorption–desorption³⁸, or active refrigeration AWH³⁹), RC-based dew collection enables an electricity-free, low-complexity route for direct vapor-to-liquid condensation on a stationary surface.

Fig. 3. Applications of radiative cooling (RC) technology for atmospheric water harvesting.

a Schematic of RC-enabled atmospheric water harvesting. b Representative examples showing the historical development of RC-enabled water harvesting. c, d Representative RC configurations for water harvesting: planar (c)¹⁵ and vertical structures (d)⁴³. e Integration of solar desalination with RC-driven condensation for all-day freshwater harvesting⁴⁴. f All-in-one hybrid atmospheric water harvesting system producing water throughout the day using natural sunlight and RC⁴⁵. Panel c adapted with permission from Ref. ¹⁵, PNAS. Panel d adapted with permission from Ref. ⁴³, Wiley. (e) adapted with permission from Ref. ⁴⁴, American Chemical Society. Panel f adapted with permission from Ref. ⁴⁵, Royal Society of Chemistry.

Recent progress

As outlined in Fig.3b, RC-driven dew collection was first demonstrated in 1965, when Gindel et al. used PE films to capture dew for irrigation in xerophytic environments⁴⁰. Subsequent studies explored various coated/filled composites⁴¹ and metallic radiators⁴² across various climates. Although many early systems operated only at night owing to daytime solar heating, advancements in spectral selectivity have unlocked the potential for daytime condensation. A prominent example is a polydimethylsiloxane-based planar condenser designed by integrating high solar reflectance with strong MIR emissivity, achieving net RC and dew collection under peak sunlight (Fig. 3c)¹⁵. To improve robustness and extend operating hours, modern designs have further incorporated directional radiation shields, leading to higher daily yields compared to traditional dew foils¹⁴. Despite these advances, efficient removal of collected droplets remains a practical bottleneck in planar architectures, as droplet accumulation can penalize heat/mass transfer and radiative performance.

To overcome these limitations, vertical RC geometries combined with surface lubrication have been developed (Fig. 3d), which accelerates condensate shedding, reduces droplet accumulation, and helps preserve radiative performance⁴³. Under T_amb = 20 °C and RH of 80%, this configuration has achieved condensation efficiencies approaching 87% of the theoretical limit, which is calculated as $\eta ={\stackrel{°}{m}}_{\exp }/{\stackrel{°}{m}}_{\max }$, where ${\stackrel{°}{m}}_{\exp }$ and ${\stackrel{°}{m}}_{\max }$ denote the measured and model-predicted maximum condensation rates under the reported operating conditions, respectively. Further innovations—such as hybrid systems integrating RC with solar-driven evaporation (Fig.3e)⁴⁴ or sorption-based AWH (Fig.3f)⁴⁵—broaden operating windows, enabling 24-hour water production⁴⁶ and water capture under extremely dry environments⁴⁷. Representative RC-enabled dew-collection systems and their key material/design innovations are summarized in Table 1.

Table 1.

Representative materials and design innovations for radiative-cooling-enabled dew collection

Materials/system	Water-harvesting Function	Design innovation	Ref.
Polyethylene film	Nighttime dew collection for irrigation	Used sky-facing polymer films for passive dew collection in desert environments	40
Pigmented polymer foils	Passive dew condensation	Enhanced longwave radiative heat loss through material coatings and filler-modified radiators	41
Metallic condenser surfaces	Atmospheric vapor condensation for irrigation	Demonstrated practical condenser surfaces for water collection in arid coastal climates	42
Directionally shielded radiative condenser	Continuous day–night AWH	Suppressed parasitic heat gain and extended RC-driven water collection toward uninterrupted 24-hour operation	14
PDMS/Ag/Al planar selective emitter	Daytime vapor condensation	Combined high solar reflectance with strong MIR emissivity to enable condensation under sunlight	15
Vertical double-sided lubricated RC surface	Rapid condensation and condensate shedding	Coupled vertical geometry with lubricant-assisted droplet removal to reduce accumulation and preserve radiative performance	43
Solar absorber integrated with RC condenser	All-day freshwater harvesting	Coupled solar evaporation with RC-driven condensation for extended water production	44
RC-assisted sorption/hydrogel AWH hybrids	All-day or climate-adaptive AWH	Combined RC, solar-driven water release and hygroscopic moisture capture to extend operation from day–night cycling to dry-climate water harvesting	45–47

Challenges and opportunities

Despite recent progress, the large-scale deployment of RC-enabled dew collection remains constrained by several practical issues.

Airborne contamination and water safety represent a primary concern for exposed condenser surfaces. Dust, bioaerosols and atmospheric pollutants can accumulate on these surfaces, impairing optical performance and posing risks to potable water safety. A promising strategy is to integrate ultraviolet (UV)-activated photocatalytic coatings that harness the UV portion of sunlight to disinfect the condensate, while maintaining high reflectance across the visible (400–700 nm) and near-infrared (NIR, 0.7–2.5 µm) bands to minimize P_sun, thereby preserving the desired spectral selectivity for passive cooling⁴⁸.

Condensate–radiation coupling must also be carefully considered because water is a strong infrared emitter. Persistent water films can reshape the effective emissivity and suppress net RC beyond a critical coverage⁴⁹. System-level designs that decouple condensation from the radiating surface (e.g., reverse-condensation or heat-exchanger geometries) can alleviate this penalty^14,15, but require further co-optimization of heat transfer and condensate transport.

Efficient condensate shedding is another key requirement for sustaining subambient cooling and maximizing yield. Surface engineering strategies, including superhydrophobic coatings⁵⁰ and lubricant-infused coatings⁵¹, can enhance shedding, but large-area implementations must be validated under realistic natural sunlight exposure, long-term aging, dust loading, repeated wet–dry cycles, and cost constraints.

Radiative-cooling-integrated sustainable agriculture

Greenhouses are controlled microclimate systems shielding crops from adverse weather and optimizing environmental conditions to enhance agricultural productivity⁵². However, excessive heat buildup—particularly in tropical and subtropical climates—remains a persistent constraint. Conventional cooling methods (electric fans and evaporative systems) are widely used, yet often limited by high energy demand and water usage. RC offers a passive, low-energy-input alternative that mitigates thermal loads and enhances the water sustainability of greenhouse agriculture.

Environmental contextualization

As illustrated in Fig. 4a, in conventional greenhouse settings, unfiltered solar radiation is the primary driver of thermal gain. Consequently, an ideal greenhouse system for warm climates should provide selective spectral control: maximizing solar transmittance in the photosynthetically active radiation (PAR) range (0.4–0.7 μm)⁵³ while reflecting NIR wavelengths that contribute to heat load²³. Simultaneously, the structure should have high emissivity in the MIR atmospheric window to facilitate efficient radiative heat dissipation⁵⁴.

Fig. 4. Applications of radiative cooling (RC) technology in sustainable agriculture.

a Schematic of a conventional (left) and an RC-integrated greenhouse (right) under hot climate conditions. b Schematic of the RC envelope material (left) and its ideal spectral profile (right). c Schematic of the RC mulch (left) and its ideal spectral profile (right)¹⁶. d Selected examples illustrating the development of RC-integrated agricultural technologies. e Photograph of a greenhouse covered with transparent RC films; the inset shows the RC film⁵⁵. f Photograph of a meter-scale greenhouse with photosynthetically active RC envelopes designed for effective passive cooling and water conservation⁵⁸. g Synergistic photonic engineering of RC envelopes and mulches for passive thermal regulation in greenhouses located in extremely hot regions⁶⁰. a–c Adapted with permission from Ref. ¹⁶, Elsevier. (e) Adapted with permission from Ref. ⁵⁵, Elsevier. (f) Adapted with permission from Ref. ⁵⁸, Springer Nature. g Adapted from Ref. ⁶⁰, CC BY 4.0.

Thermal regulation strategies in warm-climate greenhouses typically rely on two key structural components: the envelope and mulch⁵⁵. The envelope, acting as the primary optical interface, should be engineered for spectrally selective transmittance—transmitting PAR, reflecting NIR to suppress P_sun, and emitting strongly in the MIR range to enhance P_rad (Fig.4b). Meanwhile, agricultural mulches modulate soil temperature, a parameter that substantially affects plant development⁵⁶. Effective RC-integrated mulches should exhibit high MIR emissivity—particularly within the 8–13 μm range—to enhance P_rad, while maintaining high reflectance at shorter wavelengths to reduce P_sun (Fig.4c).

Recent progress

The application of RC technology in modern greenhouse systems is a relatively recent innovation (Fig.4d). One early implementation involved PE films embedded with SiO₂ nanoparticles, achieving a 5 °C temperature reduction compared with unmodified films⁵⁷. Subsequently, a multilayer envelope design was introduced, incorporating a structural polyethylene terephthalate (PET) backing, a spectrally selective Ag/SiO₂ metamaterial for NIR reflection, and a polyvinylidene fluoride surface layer to boost MIR emissivity (Fig.4e)⁵⁵. Field evaluations reported temperature drops of up to 18.6 °C and crop yield gains of 17.3%–23.4% relative to conventional polyolefin-based films. Beyond passive cooling, the spectral optimization of envelope transmittance substantially enhances photosynthetic efficiency while reducing evaporative losses. Adjusting the balance between light transmission and thermal emissivity has delivered air temperature reductions of 1.9–4.6 °C, evaporation savings of 2.1%–31.9%, and yield increases of 20%–370%, depending on crop species and local climate conditions (Fig.4f)⁵⁸. These advancements highlight the potential of RC-integrated envelopes to co-optimize cooling, water use, and productivity.

Complementary to envelope strategies, RC-enabled mulches offer substantial benefits for regulating soil temperature. Reportedly, a reflective RC mulch lowered soil surface temperatures from 45.3 °C to approximately 32.8 °C under peak summer conditions⁵⁹. When integrated with RC-optimized envelopes, this synergistic approach achieved soil temperature reductions of >25.1 °C (Fig.4g)⁶⁰, effectively extending the growing season while reducing the environmental footprint of controlled-environment agriculture. Representative RC-integrated agricultural materials and design innovations are summarized in Table 2.

Table 2.

Representative materials and design innovations for radiative-cooling-integrated sustainable agriculture

Materials/system	Agricultural function	Design innovation	Ref.
SiO2-particle-embedded polyethylene greenhouse film	Greenhouse envelope cooling	Incorporated silica particles into polyethylene films to enhance MIR emission and reduce greenhouse air temperature relative to unmodified films	57
PET/Ag–SiO2/PVDF multilayer transparent radiative-cooling film	Transparent greenhouse envelope for crop-compatible cooling	Combined a PET support layer, Ag/SiO2 spectrally selective layer and PVDF surface layer to transmit PAR, reflect NIR radiation and enhance MIR emission	55
PAR-transparent RC film	Greenhouse envelope for passive cooling and water conservation	Optimized the balance between PAR transmittance and thermal emissivity	58
Polymeric RC mulch film	Soil/root-zone cooling	Used a reflective and MIR-emissive mulch to reduce soil heat accumulation under hot-climate cultivation conditions	59
Spectrally selective greenhouse cover combined with RC mulch film	Integrated air–soil thermal regulation	Coupled an NIR-blocking, MIR-transparent greenhouse cover with an RC mulch film to simultaneously regulate air and soil temperatures	60

Challenges and opportunities

Despite these advances, several challenges must still be overcome to enable the broad adoption of RC-enabled greenhouse technologies.

Material durability remains a primary concern for RC-enabled greenhouse technologies. Polymeric films and mulches predominantly used in agriculture are vulnerable to UV degradation, which results in discoloration, decreased optical transmittance, and eventual mechanical failure⁶¹. Prolonged solar exposure necessitates the development of materials with enhanced photostability. Promising routes include UV stabilizers⁶², inorganic additives (preferably immobilized/encapsulated to minimize release and evaluated via leaching/abrasion tests)⁶³, and multilayer protective coatings^16,19. However, their durability across full growing seasons, especially under continuous agrochemical exposure, still demands further rigorous field validation.

Scalability and economic viability are also essential for broad adoption. Broad adoption of RC materials in agriculture hinges on manufacturing methods that combine scalability with cost efficiency. Techniques such as roll-to-roll nanoimprint lithography²⁹ and solution-based coatings⁶⁴ facilitate large-area production but still require refinement to deliver consistent quality and long-term durability. Additionally, high-performance materials must be affordable for deployment across extensive farmlands. To accelerate commercialization, advances in low-cost materials and manufacturing processes—supported by policy incentives and industry collaboration—are essential for ensuring that RC technologies are accessible and economically sustainable¹².

End-of-life management and environmental footprint require careful consideration. Widespread use of plastic films, particularly as mulches, raises waste and contamination concerns. Practical pathways include recyclable designs, take-back logistics, and biodegradable alternatives for short-lived mulches⁶⁰, provided that optical performance and mechanical integrity are retained over the intended service life. For longer-lived components (e.g., greenhouse envelopes), recyclability and controlled end-of-life handling are generally preferred, representing goals for future research and development for sustainable agriculture^55,58.

Synergistic radiative–evaporative cooling

Evaporative cooling removes heat through the endothermic liquid–vapor phase change of water⁶⁵. Although effective, this method is inherently limited by water availability, restricting its use in arid or water-scarce regions. Integrating RC offers a complementary, electricity-free heat-dissipation pathway. By extending the RC energy-balance framework (Eq. (1)) to incorporate phase-change heat removal, the net cooling power becomes$:$¹⁷

$$P_{\mathrm{net}}=P_{\mathrm{rad}}-P_{\mathrm{atm}}-P_{\mathrm{sun}}-P_{\mathrm{non}-\mathrm{rad}}+P_{\mathrm{evap}}$$ 4

where P_evap represents the evaporative latent-heat removal. In this extended framework, RC increases net radiative heat loss, while evaporation provides an additional latent-heat sink that further reduces the steady-state surface temperature¹⁷. For a given subambient temperature target, this hybrid approach can lower the required evaporative mass flux (and thus water consumption), improving cooling efficiency and strengthening the sustainability and climatic adaptability of evaporative systems across diverse environments.

Modes of synergistic radiative–evaporative cooling

Evaporative cooling can be categorized into direct evaporative, sorption-driven evaporative, and AWH-based coolings¹⁷. Direct evaporative cooling removes sensible heat by evaporating liquid water. Although effective, it requires substantial water supply, limiting its feasibility in water-scarce regions⁶⁶. Sorption-driven systems employ hygroscopic materials to adsorb water vapor from ambient air. These materials generate high vapor pressure gradients, which accelerate evaporation and reduce dependence on bulk water supplies⁶⁷, although they often require additional structural components for effective regeneration. AWH-based cooling advances this concept by extracting water vapor directly from the atmosphere, thereby eliminating reliance on external water input. When integrated with RC, which passively lowers material temperatures and prolongs evaporation duration, these systems can achieve higher cooling efficiency, particularly in off-grid and resource-limited environments¹⁷.

Designing synergistic radiative–evaporative cooling systems requires the concurrent optimization of solar reflectance, MIR emissivity, and evaporation efficiency. A typical architecture consists of a top RC layer and an underlying hygroscopic evaporator (Fig.5a). The top layer mitigates solar heating and enables radiative heat dissipation, while the evaporator supplies an additional latent-heat sink. Notably, this radiative–evaporative coupling embodies a humidity-mediated trade-off: elevated humidity can sustain hygroscopic evaporators by promoting water uptake and delaying dry-out, yet it concurrently suppresses the net cooling power P_net in Eq. (4) by increasing P_atm (via enhanced atmospheric back-radiation) and reducing P_evap (via diminished vapor-pressure difference that drives evaporation)¹⁷. Consequently, radiative–evaporative hybrids should be designed by matching the water-supply rate to the evaporation demand and minimizing parasitic heat gains (e.g., via reduced convective heat exchange and spectrally selective emitter design) under realistic weather conditions.

Fig. 5. Applications of synergistic radiative–evaporative cooling.

a Schematic of the synergistic radiative–evaporative cooling mechanism. b Timeline of radiative–evaporative cooling technology development. c‒f Representative material designs for synergistic radiative–evaporative cooling, including hierarchical structures (sorbed water) (c)⁷⁰, integrated structures (sorbed water) (d)⁷², paint-based systems (sorbed water) (e)⁷⁵, and nanoporous structures (perspiration) (f)⁷⁸. Panel c adapted with permission from Ref. ⁷⁰, AAAS. d adapted with permission from Ref. ⁷², Wiley. e Adapted with permission from Ref. ⁷⁵, AAAS. f Adapted with permission from Ref. ⁷⁸, American Chemical Society.

Recent progress

Early demonstrations combined nocturnal RC with daytime evaporation by pre-cooling water at night for subsequent evaporative cooling, achieving >13 °C reduction in ambient air temperature⁶⁸. Subsequent developments have focused on climate-adaptive, multilayer stacks that jointly manage radiation, vapor transport, and parasitic heat gains (Fig.5b). For example, hydrogel-enabled roof-cooling concepts have leveraged evaporation-driven downdraft flows for daytime cooling in arid settings⁶⁹, while bilayer and multilayer architectures have been engineered to simultaneously enhance radiative emission and maintain efficient vapor transport (Fig. 5c)⁷⁰. The integration of thermal-insulation layers has further enabled subambient cooling of up to 9.3 °C and cooling powers of 96 W m⁻²⁷¹.

More recently, all-in-one structures have integrated radiative, water-harvesting, and evaporative functions on a single platform. A representative example is metal–organic framework-coated fabrics that combine high solar reflectance (0.98), autonomous moisture uptake, and sustained evaporation, enabling surface temperature reduction by 25.3 °C in passive thermal management of power infrastructure (Fig.5d)⁷². Related advances include self-humidifying⁷³ or evaporation-regulating hydrogels⁷⁴ and cementitious cooling coatings that couple broadband solar reflection, high MIR emittance, and self-replenishing evaporation, achieving multi-degree cooling in field tests (Fig.5e)⁷⁵. These advances highlight a shift from component-level integration toward platform materials that are simpler to deploy in off-grid and resource-limited environments.

Beyond architectural applications, radiative–evaporative principles are increasingly applied to personal thermal management. Because human skin has high infrared emissivity ( ~ 0.95)⁷⁶, infrared-transparent, breathable textiles (e.g., nanoporous PE) can transmit body thermal radiation while facilitating sweat evaporation⁷⁷. For instance, a bilayer PE membrane with anisotropic wettability could improve sweat drainage and enhance evaporative efficiency for wearable cooling (Fig.5f)⁷⁸. More advanced designs incorporate CaCl₂-coated cotton–polyester fibers beneath porous polyvinylidene fluoride-based top layers, allowing passive moisture adsorption and controlled evaporation under ambient conditions⁷⁹. These examples illustrate the scalability of radiative–evaporative concepts across applications spanning built environments, infrastructure, and wearables. Representative materials and design innovations for synergistic radiative–evaporative cooling are summarized in Table 3.

Table 3.

Representative materials and design innovations for synergistic radiative–evaporative cooling

Materials/system	Cooling function	Design innovation	Ref.
Nocturnal RC-assisted evaporative cooling system	Building or air cooling	Used nocturnal RC to pre-cool water for subsequent daytime evaporative cooling	68
Hydrogel-enabled roof-cooling prototype	Roof-level radiative–evaporative cooling	Leveraged hydrogel water evaporation and downdraft airflow to enhance daytime cooling	69
Cellulose acetate/poly(vinyl alcohol)–CaCl2 hydrogel	Weather-insensitive daytime passive cooling	Integrated radiative heat loss, evaporation, vapor transport and insulation into a hierarchical stack for high-performance subambient cooling	70
Enhanced Specular Reflector/polyacrylamide hydrogel/polyethylene aerogel	Subambient passive cooling	Incorporated thermal-insulation layers to suppress parasitic heat gains while maintaining radiative and evaporative heat dissipation	71
MOF-coated fabric	Passive cooling for power equipment	Combined high solar reflectance, autonomous moisture uptake and sustained evaporation on a fabric platform for infrastructure cooling	72
Polyacrylate film	Hybrid radiative–evaporative passive cooling	Integrated atmospheric moisture uptake, radiative heat loss and regulated evaporation to sustain passive cooling without continuous external water supply	73
Cement cooling paint	Paint-based radiative–evaporative cooling	Coupled broadband solar reflection, high MIR emittance and self-replenishing evaporation through rational control of optical and mass-transfer properties	75
Nanoporous PE textile	Wearable radiative–evaporative cooling	Enabled transmission of body thermal radiation while facilitating sweat evaporation through breathable infrared-transparent PE	77
Bilayer nanoporous PE membrane with anisotropic wettability	Wearable perspiration cooling	Combined directional water transport with radiative transparency to accelerate sweat removal and evaporation	78
CaCl2-coated cotton–polyester fibers with porous PVDF-based top layer	Moisture-adsorbing wearable cooling	Integrated hygroscopic salt-assisted moisture uptake with controlled evaporation and radiative cooling in textile structures	79

Challenges and opportunities

Despite promising synergistic radiative–evaporative cooling, several material and design constraints must be resolved to enable practical deployment across diverse applications.

For wearable cooling systems, functionality and durability remain central challenges. Achieving dyeability and aesthetic versatility is essential because commercial fabrics must balance functional performance with user acceptance⁸⁰. Mechanical durability—especially under repeated washing cycles—is equally critical, as detergents and agitation can degrade hygroscopic and radiative properties over time⁸¹. Therefore, materials must be engineered to maintain effectiveness after long-term use.

Moreover, environmental sensitivity must be carefully addressed. Ambient humidity fundamentally affects evaporation, which can shift the optimal balance between radiative and latent pathways and influence the effectiveness of both cooling mechanisms⁸². Promising directions include humidity-responsive actuators⁸³ and passive/low-power heat-gating designs⁸⁴ that adapt to non-ideal weather.

Supplementary Table 1 summarizes the shared design logic and track-specific trade-offs across the three RC–water nexus applications by comparing operating principles, structural/optical designs, representative performance metrics, and field bottlenecks. To enable practical deployment, future work should standardize life cycle assessments that quantify water-footprint components across the full lifecycle—from raw-material sourcing and manufacturing to operation/maintenance, replacement, and end-of-life recycling/disposal⁸⁵.

Outlook: the radiative cooling–water nexus and beyond

RC is emerging as a transformative passive thermal management strategy, offering a scalable route toward climate resilience and carbon neutrality²¹. When coupled with sustainable water technologies, RC enables a broad range of multifunctional applications—including personal thermal management⁸¹; water harvesting^14,15; power generation⁸⁶; cooling for buildings⁷⁵, greenhouses⁵⁹, and data centers⁸⁷ (Fig.6a). These hybrid systems can be attractive in arid and resource-limited regions, where conventional cooling and water supply infrastructure may be ecologically and economically unsustainable. Realizing the full potential of the RC–water nexus will demand coordinated progress in material innovation, device engineering, and systems integration, with solutions tailored to varied climatic and socio-environmental settings. The following sections highlight emerging opportunities and key barriers that must be addressed to advance RC–water technologies from laboratory-scale demonstrations to robust, climate-resilient infrastructure (Fig.6b).

Fig. 6. Prospects and deployment roadmap for synergistic radiative cooling (RC) and water technologies.

a Schematic of the RC–water nexus, covering established tracks (RC-enabled dew collection, RC-integrated sustainable agriculture, and synergistic radiative–evaporative cooling) and emerging directions (sorption-based atmospheric water harvesting (AWH), water-enabled electricity generation, and data center cooling). b Deployment-oriented bottlenecks and engineering levers across these scenarios, emphasizing manufacturability and scale-up, outdoor durability, fouling and maintenance, cost-effectiveness and replacement cycle, together with application-specific constraints.

Sorption-based AWH

Sorption-based AWH offers a complementary route to dew collection by using hygroscopic materials—such as salts⁴⁵ and metal–organic frameworks⁸⁸—to capture water vapor even at low RH³⁸. However, the equilibrium uptake of most sorbents decreases strongly with temperature⁸⁹, and thermal regeneration typically requires substantial energy input⁹⁰. Integrating RC with sorption-based AWH provides a favorable means to enhance sorbent uptake and reduce regeneration costs: nocturnal RC can passively cool the sorbent bed, shifting the sorption isotherm toward higher water uptake, whereas daytime solar heating can be combined with partial radiative coupling to the sky to tune desorption dynamics⁴⁵. Future research opportunities include the co-design of sorbent compositions and RC cover layers to optimize water yield per unit area³⁸, the development of photothermal–radiative hybrids that balance fast regeneration with limited radiative loss, and the realization of low-cost, cycling-stable modules suitable for off-grid deployment in arid climates. Quantitative comparisons with conventional AWH and dew collectors, under realistic diurnal TPW and wind profiles, will be crucial for assessing the true advantages of RC-enhanced sorption systems.

Water-enabled electricity generation

Beyond water harvesting, moisture and temperature gradients at air–water–solid interfaces can be harnessed for electricity generation through mechanisms such as atmospheric-water-driven hydrovoltaic effect^91,92 and triboelectric⁹³ and thermoelectric processes⁹⁴. When combined with RC, these systems can benefit from stabilized surface temperatures and controlled humidity gradients⁸⁶, which reduce thermal load and extend operation into hot or arid climates. In principle, an RC surface could simultaneously condense or adsorb water and drive low-power electricity generation for distributed sensing, control of valves and pumps, or wireless communication in remote water infrastructure.

At present, however, atmospheric-water-driven generators suffer from low areal power density, electrode degradation, and performance fluctuations under variable humidity. Realizing practical RC-coupled “water–electricity” platforms will require advances in robust electrode interfaces, scalable nano/microstructured surfaces, and system-level designs that integrate energy harvesting with water collection without compromising either function.

Passive cooling supplement for industrial cooling infrastructure

Cooling towers are critical facilities for large-scale industrial infrastructure, especially for emerging data centers, which can impose substantial electricity and freshwater demand⁹⁵. As a promising strategy to mitigate energy use and environmental footprint, RC technologies can be implemented at two integration layers. At the facility level, RC roofs and facades can suppress solar heat gains and minimize parasitic non-radiative heat exchange at the building envelope, thereby reducing the solar-induced cooling demand⁹⁶. This pathway represents an attractive candidate for design and construction of data centers (especially in hot regions), where large exposed roof surfaces provide abundant sky-facing areas for deploying RC envelopes⁹⁷. At the system level, embedding RC into heat-rejection loops can provide pre-cooling by lowering the coolant temperature before entering wet-cooling or mechanical stages⁸⁷. This can be implemented using radiative heat exchangers coupled to dry coolers and/or night-sky thermal storage⁹⁸, thereby reducing the required evaporative duty. In addition, in arid regions, RC can be coupled with water-harvesting modules (dew collection or sorption-based AWH) to provide non-potable make-up water for cooling towers, alleviating local water stress and complementing economizer-based free-cooling strategies⁹⁹.

Realizing these benefits requires coordinated co-design across climate, workload, and facility conditions. Key challenges include matching RC spectral properties to local sky emissivity and humidity, minimizing parasitic heat gains in radiative heat exchangers, and integrating RC operation with dynamic information technology loads. Future field trials should quantify both water-usage effectiveness and power-usage effectiveness and operational reliability over full seasonal cycles.

Summary

The radiative cooling (RC)–water nexus represents a paradigm shift in sustainable technology, where passive thermal regulation and atmospheric water management intersect to address critical challenges in climate adaptation, energy equity, and resource resilience. Strategic integration across sectors—from wearable textiles and agricultural systems to urban and digital infrastructure—positions RC technologies as foundational elements of a decarbonized, climate-adaptive future. Realizing this vision will require durable, climate-resilient materials, cost-efficient and scalable manufacturing processes, and completely integrated system architectures validated under real-world conditions. Standardized tools, such as life cycle assessments and techno-economic analyses, will be vital to ensure environmental and economic viability at scale. Moreover, combining RC and water technologies directly advances multiple United Nations Sustainable Development Goals (SDGs)¹⁸, particularly SDG 6 (clean water and sanitation) and SDG 7 (affordable and clean energy). As global climate and water stresses intensify, RC–water systems offer a unified platform for managing thermal loads and water resources across built environments, agriculture, and digital infrastructure. By transitioning from materials to completely developed systems—emphasizing the durability, multifunctionality, and environmental circularity—RC can move beyond passive cooling to become a cornerstone of climate-resilient infrastructure. Achieving this will require sustained cross-disciplinary collaboration among experts in photonics, materials science, environmental engineering, and policy to unlock the full potential of this nexus and deliver transformative global impact.

Author contributions

Q.G. and S.F. conceived and defined the scope of this Review. S.F. drafted the sections ‘Passive cooling–water nexus’ and ‘Outlook: the radiative cooling–water nexus and beyond’. S.D. drafted the section ‘Atmosphere-dependent cooling criteria’. Q.G. supervised this work. All authors discussed and edited the manuscript.

Peer review

Peer review information

Nature Communications thanks Qiang Fu, who co-reviewed with Yihan Shi, Junsuk Rho, Sau Chung Fu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Funding

This work was supported by a baseline from the King Abdullah University of Science and Technology (BAS/1/1415-01) and the KAUST Center of Excellence for Renewable Energy and Storage Technologies (CREST) (FCC/1/5937-06-01).

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-74110-z.