SC

Decadal and Seasonal Variability of Aerosol Types and Meteorological Parameters in the Central Himalayas (2014–2024): High-Altitude Analysis from the Third Pole

Introduction

  • Study focus: Decadal and seasonal variability of aerosol types and meteorological parameters in the Central Himalayas ( Nepal, elevations >2500 m ) over 2014–2024, using high-altitude data (“Third Pole”).
  • Key aim: Assess seasonal and interannual variability of aerosol optical properties (AOD, AE, AI) and meteorological drivers (precipitation, relative humidity, temperature, wind) and their coupling.
  • Methods: Non-parametric trend detection (Mann–Kendall test) paired with Sen’s slope estimator to determine direction and magnitude of trends; analysis of relationships between aerosols and meteorology; aerosol typing by AE and AI thresholds; topographic masking (>2500 m) to focalize high-elevation dynamics.
  • Main takeaways (from abstract):
    • Aerosol Optical Depth (AOD) shows increasing tendency: Sen’s slope ≈ +0.00296 \,\text{year}^{-1}; Relative Humidity (RH) increases: Sen’s slope ≈ +0.52799 \,\% \,\text{year}^{-1}.
    • Precipitation and wind speed show slight decreases, but none of the long-term trends are statistically significant ( p>0.05 ).
    • The study identifies potential signals of atmospheric stagnation and moisture buildup in the high Himalaya, linked to warming, aerosol loading, and weakened ventilation.
  • Significance: Establishes a decadal baseline for aerosol–climate coupling in a climatically sensitive, topographically complex region and informs regional air quality and climate adaptation strategies.

Background and context

  • Aerosols influence climate directly (scattering/absorption) and indirectly (cloud microphysics and hydrological cycle).
    • Direct effect: scattering/absorption of sunlight.
    • Indirect effect: modification of cloud formation and precipitation processes.
  • IPCC AR6 (2021) highlights aerosols as a major uncertainty in climate projections due to spatial/temporal heterogeneity and interactions with clouds/radiation.
  • South Asia as an aerosol hotspot due to urbanization, biomass burning, transport, and regional winds.
  • Himalayas (the “Third Pole”) are highly sensitive: high elevation, significant aerosol deposition, and complex mountain–valley circulation shape transport and deposition.
  • Gaps: Limited high-elevation, long-term satellite-derived trend diagnostics (elevations >2500 m) and limited ground-based validation; need joint aerosol–meteorology trend analysis in this region.
  • Datasets and indicators used:
    • Aerosols: AOD, Ångström Exponent (AE), Aerosol Index (AI).
    • Meteorology: precipitation (CHIRPS), relative humidity, maximum temperature (Tmax), wind speed (from FLDAS).
    • Satellite data sources: MODIS Terra (AOD @ 550 nm; Deep Blue retrieval), AE (470–660 nm channels); UV Aerosol Index (AI) from Sentinel-5P/TROPOMI since 2019.
  • Study design: Nepal Central Himalaya, >2500 m, 2014–2024, integrated via Google Earth Engine; analysis of seasonal and annual means; seasonal definitions follow meteorological seasons: DJF (Winter), MAM (Spring), JJA (Summer), SON (Autumn).
  • Aerosol typing concepts (AE and AI thresholds):
    • High AE with low AI → fine-mode aerosols (urban/biomass burning).
    • Low AE with high AI → coarse-mode (dust).
    • Clean/background: high AE, low AI, and low AOD.

Materials and methods

Study area

  • Nepal Central Himalaya (above 2500 m), high-altitude, rugged terrain, complex orography; acts as a barrier and receptor for long-range pollutants; climate is sensitive to anthropogenic emissions and regional transport.
  • Rationale for satellite-based analysis: sparse ground-based stations above 2500 m; satellite data provide consistent coverage for trend and seasonal analyses.
  • Figure reference: Figure1 shows the study area map (Nepal Central Himalaya region >2500 m).

Datasets and variables

  • Aerosol parameters:
    • AOD (550 nm) from MODIS Terra Aerosol Monthly Product (MOD08_M3) using Deep Blue retrieval for land surfaces.
    • AE derived from the same MODIS product using 470–660 nm channels.
    • AI from Sentinel-5P/TROPOMI data (from 2019 onward) to detect UV-absorbing aerosols like smoke/dust.
  • Meteorological parameters:
    • Precipitation from CHIRPS Daily Precipitation dataset.
    • Tmax, relative humidity (RH), and wind speed from FLDAS Global Land Data Assimilation System, Version 1.
    • Temperature values originally in Kelvin; converted to Celsius for consistency.
  • Data processing: All datasets processed in Google Earth Engine (GEE).
  • Spatial masking: SRTM elevation data used to include only pixels above 2500 m within Nepal’s boundary.
  • Temporal aggregation: monthly, seasonal, and annual means; winter season includes December of the preceding year.
  • Data handling: resampling/reprojection to a common spatial resolution; missing data handled via spatial averaging and seasonal aggregation.
  • Aerosol classification scheme: categorize aerosols using AE and AI thresholds; identify clean vs. polluted episodes; compute seasonal amplitudes of AOD and AE to detect anomalies.

Analytical methods

  • Trend detection:
    • Mann–Kendall (MK) test applied to AOD, AE, and AI to detect monotonic trends without assuming normality.
    • Sen’s Slope estimator used to quantify the magnitude of detected trends.
    • MK statistic (S) and Kendall’s tau (τ) definitions:
    • S = \sum{i=1}^{n-1} \sum{j=i+1}^{n} \text{sgn}(xj - xi)
    • \tau = \frac{S}{\binom{n}{2}} = \frac{2S}{n(n-1)}
    • Significance: Trends considered significant if p < 0.05.
  • Correlation analyses:
    • Pearson correlation to assess linear relationships between aerosols and meteorological variables.
    • Spearman correlation to assess monotonic relationships robust to non-normal distributions.
    • Both used to evaluate aerosol–meteorology interactions (and cross-validated with Spearman).
    • Notation examples: r{AOD,Tmax} = 0.57, r{AE,Tmax} = -0.66, r{AOD,Pr} = 0.50, r{AE,Pr} = -0.51, r_{AOD,AE} = -0.43.
  • Aerosol type distribution:
    • Classify aerosols per AE and AI thresholds; compute seasonal proportions for each aerosol type (urban/anthropogenic, dust, clean/background) by season.
  • Seasonal amplitude analysis:
    • Compute seasonal amplitudes of AOD and AE by year to identify anomalous fluctuations and extreme events.

Definitions and notable concepts

  • AOD (Aerosol Optical Depth): total columnar extinction of solar radiation by aerosols.
  • AE (Ångström Exponent): particle size distribution indicator; higher values imply finer particles; lower values imply coarser particles.
  • AI (Aerosol Index): UV-absorbing aerosol presence indicator (dust/black carbon or smoke).
  • Seasonal definitions (DJF, MAM, JJA, SON): standard meteorological seasons; Winter includes December of the previous year.
  • Sen’s slope estimator (non-parametric):
    • S = \text{median}\left{ \frac{xj - xi}{tj - ti} \; | \; i<j \right}
    • Interpretation: robust estimate of trend magnitude.
  • Mann–Kendall trend test (non-parametric):
    • S = \sum{i=1}^{n-1} \sum{j=i+1}^{n} \text{sgn}(xj - xi)
    • \tau = \frac{S}{\binom{n}{2}} = \frac{2S}{n(n-1)}
    • Used for monotonic trend detection without assuming normality.
  • Data processing notes:
    • Spatial masking ensures focus on high-altitude Nepal (>2500 m).
    • Temporal alignment across datasets via monthly/seasonal aggregations.
    • Cloud/snow and data gaps handled via averaging and aggregation; unit conversions performed for consistency.

Results

4.1 Trend analysis

  • Trends (2014–2024) for AOD, AE, and AI show no statistically significant monotonic trends:
    • Kendall’s tau for AOD: \tau = 0.27, p = 0.28.
    • Kendall’s tau for AE: \tau = 0.16, p = 0.53.
    • Significance threshold: p < 0.05 would indicate significance; here p > 0.05 for both.
  • Sen’s slope estimates (magnitude of trends):
    • AOD: +0.00296\ \text{year}^{-1}.
    • AE: +0.00579\ \text{year}^{-1}.
  • Interannual variability is evident with notable anomalies:
    • AOD peaks in 2014, 2020, and 2022 (possible links to wildfires, biomass burning, or dust transport).
    • AE peaks in 2022, suggesting enhanced fine-mode aerosol dominance that year.
  • Implication: Long-term directional change not statistically established; episodic variability dominates high-elevation aerosols.
  • Figure references:
    • Figure2: Monthly time series of aerosol and meteorological parameters (2014–2024).

4.2 Seasonal patterns

  • Inverse seasonal relationship between AOD and AE:
    • AOD highest in summer (mean ≈ 0.44), attributed to coarse-mode dust transport via southwesterly winds.
    • AE peaks in winter (mean ≈ 1.22), indicating dominance of fine-mode aerosols from combustion sources.
  • Spring and autumn show transitional characteristics with mixed AE and AOD values.
  • Seasonal correlations:
    • AOD and precipitation: positive correlation, r = 0.50, suggesting convective uplift or post-rain dust resuspension.
    • AE and precipitation: negative correlation, r = -0.51, indicating rain scavenging of fine particles or suppression of fine particle formation.
  • RH and wind speed show weaker, less consistent correlations with aerosol metrics.
  • Aerosol type distribution by season (AE/AI threshold-based):
    • Winter and autumn: dominance of urban/anthropogenic aerosols (~57.6% of total).
    • Summer: dominance of coarse-mode dust (~39.4%), with a substantial background/clean fraction (~57.6%).
    • Spring: highest clean/background proportion (~69.7%).
  • Figure3: Seasonal averages of AOD and AE by year (2014–2024).
  • Figure4: Correlation matrix between aerosol and meteorological parameters.
  • Figure5: Seasonal aerosol type distribution (2014–2024).

4.3 Meteorological correlations

  • Key meteorological drivers of aerosol behavior:
    • AOD positively correlated with Tmax: r_{AOD,Tmax} = +0.57.
    • AE negatively correlated with Tmax: r_{AE,Tmax} = -0.66.
    • Precipitation positively correlated with AOD: r{AOD,Pr} = +0.50; with AE negatively correlated: r{AE,Pr} = -0.51.
    • RH and wind speed show weaker/less consistent correlations with aerosol properties.
  • Interpretation: Warmer periods tend to enhance dust uplift and transport (AOD up, AE down due to coarser aerosols), while precipitation enhances aerosol removal or alters composition via scavenging and subsequent formation processes.
  • Spearman correlations show similar sign patterns, confirming monotonic relationships even if non-linear.
  • Figure4: Correlation matrix visualization.

4.4 Aerosol type distribution and seasonal composition

  • Threshold-based aerosol typing reveals seasonal shifts:
    • Winter/Autumn: urban/anthropogenic aerosols dominate (~57.6%).
    • Summer: dust-dominated conditions (~39.4%); background/clean aerosols substantial (~57.6%).
    • Spring: highest clean/background proportion (~69.7%).
  • Interpretation: Seasonality reflects local emission patterns (heating, transport) and seasonal monsoonal cleansing (dust influx during dry/warm periods).
  • Figure5: Seasonal aerosol type distribution (2014–2024).

Discussion

5.1 Seasonal aerosol regimes and source dynamics

  • The study identifies a robust seasonal inversion: AOD peaks in summer due to long-range desert dust transport; AE peaks in winter due to combustion-derived fine particles.
  • Summer dust transport: enhanced by warm conditions and monsoonal flows; prior work supports desert-dust incursions into the Himalaya via westerly/oceanic pathways.
  • The +ve AOD trend in summer (not statistically significant) suggests increasing coarse-mode aerosols in that season, potentially linked to land-use changes or aridity in source regions.
  • Winter pollution: fine-mode aerosols from domestic heating and vehicular emissions; stable boundary layer reduces dispersion, increasing urban aerosol fraction.
  • Spring/autumn: transitional periods with mixed sources; post-winter and pre-monsoon dynamics influence aerosol mix.

5.2 Absence of long-term trends and implications

  • Despite year-to-year fluctuations, no statistically significant long-term trend (2014–2024) in AOD, AE, or AI.
  • Possible balancing factors: monsoonal cleansing, emission-control efforts, and competing influences (increased anthropogenic activity vs. regional meteorology).
  • Episodic events (forest fires, ENSO episodes like 2015–16 El Niño) can introduce interannual noise that obscures monotonic trends.
  • This underscores the resilience/variability of high-altitude atmospheric systems and highlights the need for higher-resolution, ground-based validation (e.g., AERONET) and integrated modeling.

5.3 Meteorological drivers as primary modulators

  • Tmax and precipitation emerge as primary modulators of aerosol behavior:
    • AOD increases with Tmax due to enhanced uplift/transport of aerosols in warmer periods.
    • AE decreases with Tmax as coarser dust becomes more prevalent in warmer seasons.
    • Precipitation positively influences AOD (potential rainfall-driven aerosol formation/upward transport or post-rain dust resuspension) and negatively influences AE (scavenging of fine particles).
  • RH and wind speed show weaker direct control on aerosol magnitude but may influence microphysical properties or regional dispersion under specific conditions.

5.4 Evolving aerosol composition: shift toward dust events

  • From 2019 to 2024, summer AE declined by ~0.04 while AOD rose by ~0.07, suggesting a shift toward dust-dominated conditions in summer.
  • Seasonal typing indicates: dust contribution in summer ≈ 40%; spring is more clean/background; winter is urban/anthropogenic dominated.
  • Potential drivers: land-use changes, vegetation loss, climate-driven aridity in dust source regions (e.g., western Asia, northern India).
  • Rare co-occurrence of high AOD and high AE in winter 2022 indicates an episode of severe local pollution with unfavorable dispersion and strong local emissions.
  • Implications: seasonal air quality management needs to address both local and transboundary sources; summer transboundary dust requires regional cooperation.

5.5 Policy and research relevance

  • Seasonal aerosol-type distinctions inform targeted mitigation strategies:
    • Winter: reduce household emissions, promote clean cooking fuels, improve heating efficiency, and curb urban emissions in mountainous communities.
    • Summer: address transboundary desert-dust transport; reinforce cross-border cooperation with neighboring regions (e.g., India, Pakistan).
  • Research implications:
    • Demonstrates the value of long-term satellite datasets in data-scarce mountainous regions.
    • Recommends integrating ground-based measurements (AERONET, low-cost sensors) with chemical transport models (e.g., WRF-Chem) and AI-based forecasting to better resolve sources and trends.

Conclusions

  • The decadal analysis (2014–2024) of the Nepal Central Himalaya indicates:
    • No statistically significant monotonic trends in AOD, AE, or AI, though there are detectable interannual fluctuations and seasonally modulated patterns.
    • AOD shows increasing tendencies and RH shows upward tendencies, but these did not reach statistical significance within the study period (p > 0.05).
    • Strong seasonal coupling between aerosol properties and meteorological drivers (Tmax, precipitation) with AOD positively linked to Tmax/precipitation and AE negatively linked to Tmax/precipitation.
    • Clear seasonal regimes: summer dust-dominated periods (coarse-mode aerosols) and winter fine-mode aerosol dominance (combustion-derived pollutants).
    • There is evidence for evolving aerosol composition, with a trend toward more dust-dominated conditions in summer over the study period, alongside persistent or episodic local pollution events.
  • Implications: The findings establish a robust baseline for long-term atmospheric monitoring in high-altitude Himalaya and highlight the need for continued data integration and higher-resolution validation to better understand aerosol–climate interactions in this ecologically and climatically critical region.

References and connected literature

  • Foundational aerosol/climate interactions and health impacts: IPCC (2021), Ramanathan et al. (2001), Pope & Dockery (2006), Koch et al. (not listed here).
  • Himalayan and South Asian aerosol context: Bonasoni et al. (2010); Gautam et al. (2009, 2010); Marinoni et al. (2010, 2013); Yao et al. (2012); Verifiable reviews and regional studies cited in the transcript.
  • Trend detection methods: Hirsch et al. (1982) for MK trend test; Sen (1968) for Sen’s slope; Zar (1972) for Spearman correlations.
  • Data sources and methodological references: Levy et al. (2013); Levy et al. (2013) MODIS aerosol algorithm; Veefkind et al. (2012); Funk et al. (2015); McNally et al. (2017); Nouveaux sources cited within the document.
  • For full reference details, see the Reference section in the study (the transcript provides a complete list).