Constrained simulation of aerosol species and sources during pre-monsoon season over the Indian subcontinent

Highlights

• Constrsimu designed to deliver a better concurrence between model estimates and observations

• Constrsimu indicated total submicron aerosol mass concentration alarmingly high over the northern/north-western India

• High AOD mainly due to dust (Sulfate-other water solubles) over the northwestern IGP (eastern region, BoB)

• Anthropogenic aerosol constituents higher over the IGP than the rest of mainland India

• Potential aerosol source fields mainly over the northern/western IGP during Tigerz IOP

Abstract

This study was designed to deliver a better concurrence between model estimates and observations, of atmospheric aerosol species, and predict their spatial distribution as consistently as possible. A free running aerosol simulation (freesimu) in a general circulation model (GCM) was performed, and further the simulated aerosol optical depth (AOD) was constrained with the observed AOD. The present study was carried out during the pre-monsoon season and for the Tigerz experiment which was conducted at stations over the Indo-Gangetic plain (IGP) and the Himalayan foot-hills in northern India. Our formulation of the constrained aerosol simulation (constrsimu) was based upon an identification of the freesimu with the most consistent estimates of aerosol characteristic among the three freesimu. The three freesimu (differing in source of emissions and model horizontal resolution) were carried out with the general circulation model (GCM) of Laboratoire de Météorologie Dynamique (LMD-ZT GCM). Black carbon (BC), organic carbon (OC), and sulfate-other water soluble (Sul-ows) estimated from constrsimu amounted to 70%–100% compared to that from freesimu being 20%–50% of their measured counterparts. Among the aerosol species, the pre-monsoon mean concentration of dust was considerably high over most part of the Indian subcontinent; the anthropogenic aerosol species were, however, specifically predominant over the IGP (mostly 8–12 μg m⁻³ for Sul-ows, OC). The constrsimu estimated total submicron aerosol mass concentration revealed its alarmingly high value over the northern and north-western India (> 100 μg m⁻³ and as high as 300 μg m⁻³). While the high value of observed AOD was found being mainly due to dust (AOD due to dust greater than 0.3) over the northern–northwestern IGP, it was due to Sul-ows (AOD due to Sul-ows as high as 0.4) over the eastern IGP, eastern coastline, and the Bay of Bengal. Temporal trend of fine (FM) and coarse mode (CM) AOD from constrsimu estimates and that derived from Tigerz experiment were in phase with each other for most of the days and exhibited a strong positive correlation coefficient. Source of Tigerz aerosols was mainly due to a predominant influence of dust from Africa/west Asia followed by that from northwest India, and of anthropogenic emissions originating in the IGP. A 200% increase was inferred for potential black carbon emissions (using India emission inventory implemented in a GCM) to obtain a concurrence between observed and freesimu BC concentration.

Introduction

Atmospheric aerosols over South Asia have received much attention over the past two decades due to their impact on climate and human health (Butt et al., 2016; Wang et al., 2014a; Shindell et al., 2012; Ramanathan et al., 2005; Sahu et al., 2011). It has been postulated that atmospheric transport of anthropogenic aerosols over the Tibetan plateau could enhance snow melting thus having implications on the continental hydrological cycle of South Asia (Qian et al., 2011). There is evidence of anthropogenic aerosols over South Asia having a weakening effect on the summer monsoon, however dust aerosols over West Asia and Arabian Sea are very likely enhancing the intensity of monsoon precipitation over the South Asian region (Bollasina et al., 2013; Vinoj et al., 2014; Wang et al., 2009). There is still a lack of understanding of aerosols impact on the regional climate system. Aerosol chemical composition and concentration vary largely in space and time due to variety of sources for different species, their short residence time, the large influence of prevailing meteorological conditions, and the various and complex aerosol dynamic processes (Hodzic et al., 2004; Tchepel et al., 2013). The spatio-temporal variability in aerosol properties thus poses a challenge to the scientific community in terms of modeling of aerosol physical, chemical and optical properties needed for an accurate assessment of their interactions with climate on a regional to global scale.

Recent evaluation of aerosols from global aerosol multi-models using global aerosol emissions (from AeroCom Phase II–Atmospheric Chemistry and Model Inter-comparison project (Myhre et al., 2013) or NASAs Modeling, Analysis and Prediction program (Diehl et al., 2012)) indicate that forward model simulations were inadequate to capture the observed aerosol optical depth (AOD) from ground-based stations and their spatial distribution as inferred from satellite-based retrievals over the Indian subcontinent (Pan et al., 2015). These evaluations showed aerosol species estimated in global model simulation were only about 5% to 50% of observed sulfate concentrations and about 8% to 46% of observed BC concentrations over Indian region during the winter season (December–January-February). The simulated BC concentration in a regional chemical transport model over India was found to be better correlated with measured surface BC concentration during pre-monsoon than during winter, but was still underestimated by a factor of 2 to 4 during the pre-monsoon season and 4 to 9 during winter based on measurements over the Indo-Gangetic plain (IGP) and Himalayan stations (Moorthy et al., 2013; Nair et al., 2012; Joshi et al., 2016). Previous studies have reported on possible reasons for the discrepancy between model and measurements including an underestimation of emissions, a lack of hygroscopic growth and formation of secondary inorganic and organic aerosols, inadequate meteorology and representation of mixing state of aerosols, and too coarse resolution in the model (Wang et al., 2016; Pan et al., 2015; Kumar et al., 2015a, Kumar et al., 2015b; Moorthy et al., 2013; Verma et al., 2011; Reddy et al., 2004).

A large difference between simulated and observed aerosol distribution over the Indian region limits the accuracy in prediction ability of aerosol-climate interactions from aerosol-chemistry-climate models over the region. This difference is typically large over the IGP, where the atmosphere is observed laden with a large pollutant level of aerosol load. Due to the inclusion of various complex physical-chemical atmospheric and aerosol processes in these models, in conjunction with inherent discrepancy in inputs to the model, e.g. aerosol emissions and their properties, a systematic approach is required to improve the prediction of aerosols. The constrsimu is used to establish an alternate approach for estimating the atmospheric concentration by surpassing the discrepancy induced specifically by emissions in source region which prevails in case of the free running aerosol simulations. In the present study, aerosol characteristic is evaluated from the three free running aerosol simulations (freesimu) (differing in source of emissions and model horizontal resolution) carried out with the LMD-ZT GCM. This evaluation leads to identify the most consistent freesimu estimates with observations out of the three freesimu. The formulation of a constrained aerosol simulation (constrsimu) results from a fusion of observed AOD and simulated aerosol characteristics from the identified freesimu.

The present study is carried out during the pre-monsoon season and for the Tigerz experiment over the Indian region. The Tigerz experiment was conducted under the NASA Aerosol Robotic Network (AERONET) project during the pre-monsoon seasons (April to June) of 2008 and 2009. This experiment was set up to characterize columnar aerosol optical properties and their spatial variability from the IGP to central Himalayas; this experiment included ground-based measurement of aerosol optical properties using AERONET sunphotometers at Kanpur and Bareilly locations in the IGP and Pantnagar and Nainital in the central Himalayan region (Giles et al., 2011; Dumka et al., 2014). Observations revealed a spatial variability in AOD500 (at 500 nm) between the stations, with the highest observed mean AOD500 over the IGP stations but a relatively larger aerosol absorption over the Himalayan foothill stations (Dumka et al., 2014). Based on aerosol absorption and size relations inferred from observations of aerosol optical properties, different categories of aerosol characteristic could be inferred including pollution, dust, and a mixture of pollution and dust (Srivastava et al., 2012; Giles et al., 2011, Giles et al., 2012).

In the present study, the efficacy of the newly formulated constrsimu is evaluated with respect to freesimu through examining their ability to simulate the observed concentration of aerosol species during the pre-monsoon season and corresponding to the Tigerz experiment. Estimates from constrsimu are further used to understand aerosol species contributing to the pre-monsoon AOD and surface concentration over the Indian subcontinent and that measured during the Tigerz campaign, including their sources of origin. Further, the potential BC emissions contributing to BC aerosols during the Tigerz intensive operational phase (IOP) at Kanpur are evaluated. The specific objectives of the present study include an evaluation of (i) aerosol optical and chemical properties from freesimu in LMD-ZT GCM and that from constrsimu with available measurements at Tigerz stations corresponding to the pre-monsoon period, (iii) spatial distribution of aerosol species optical depth and concentration from constrsimu over the Indian subcontinent during the pre-monsoon season, (iii) sources of Tigerz campaign aerosols through combined analysis of potential source fields and constrsimu estimates from region-tagged aerosol simulations. Additionally, estimation of potential black carbon emissions (using emission inventory implemented in GCM) corresponding to source fields influencing Tigerz campaign aerosols is also carried out.