Search

Accurate estimates of spatial above ground biomass (AGB) in tropical forests are important in understanding the global carbon cycle. Microwave and optical remote sensing datasets have been used extensively for AGB estimation, but their uses are restricted due to saturation in high biomass region. To overcome saturation issues of single sensor based models, the current paper uses a non-parametric Random Forest based approach to spatially estimate biomass over Indian forests using field inventory data in combination with optical (MODIS) and Microwave (L-band ALOS-PALSAR) images along with other bio-climatic parameters (e.g., rainfall, temperature) which significantly influence the biomass accumulation in an ecosystem. Plot level biomass estimates for 6678 sample plots (0.1 ha size), inventoried as part of robustly designed National Forest Inventory (NFI), are computed using volumetric equations, wood density and biomass expansion factors. Spatial above ground biomass estimates were generated using random forest model over two scenarios. Firstly, a single nation-wide model using all the available plot data and secondly, physiographic zone (14 zones over India) wise models with plot data over respective zones. AGB stored in Indian forest is estimated as 7952.3 million tonnes (Carbon equivalent: 3737.58 TgC) with a root mean square error (RMSE) of 31.2% using national level model. Physiographic zone level models estimated the country's biomass as 7597.45 million tonnes (Carbon equivalent: 3570.8 TgC). The above ground biomass estimates from our study indicates that the estimation error in physiographic zone model varies from 25.24% to 54.15% depending upon the sample size and biomass range. We observed that L-band microwave saturated in 140–160 Mg ha−1 range and currently available microwave wavelengths alone is not sufficient to predict entire range of biomass over Indian forests. Inclusion of multisource data using random forest regression model increase the saturation range to 350 Mg ha−1 which is a significant improvement as 94.7% of Indian forests are covered in this range. Model estimation error reduces to 25.6% in AGB range up to 350 Mg ha−1.

Tropical forests contribute to approximately 40 % of the total carbon found in terrestrial biomass. In this context, forest/non-forest classification and estimation of forest above ground biomass over tropical regions are very important and relevant in understanding the contribution of tropical forests in global biogeochemical cycles, especially in terms of carbon pools and fluxes. Information on the spatio-temporal biomass distribution acts as a key input to Reducing Emissions from Deforestation and forest Degradation Plus (REDD+) action plans. This necessitates precise and reliable methods to estimate forest biomass and to reduce uncertainties in existing biomass quantification scenarios. The use of backscatter information from a host of allweather capable Synthetic Aperture Radar (SAR) systems during the recent past has demonstrated the potential of SAR data in forest above ground biomass estimation and forest / nonforest classification. In the present study, Advanced Land Observing Satellite (ALOS) / Phased Array L-band Synthetic Aperture Radar (PALSAR) data along with field inventory data have been used in forest above ground biomass estimation and forest / non-forest classification over Odisha state, India. The ALOSPALSAR 50 m spatial resolution orthorectified and radiometrically corrected HH/HV dual polarization data (digital numbers) for the year 2010 were converted to backscattering coefficient images (Schimada et al., 2009). The tree level measurements collected during field inventory (2009–'10) on Girth at Breast Height (GBH at 1.3 m above ground) and height of all individual trees at plot (plot size 0.1 ha) level were converted to biomass density using species specific allometric equations and wood densities. The field inventory based biomass estimations were empirically integrated with ALOS-PALSAR backscatter coefficients to derive spatial forest above ground biomass estimates for the study area. Further, The Support Vector Machines (SVM) based Radial Basis Function classification technique was employed to carry out binary (forest-non forest) classification using ALOSPALSAR HH and HV backscatter coefficient images and field inventory data. The textural Haralick’s Grey Level Cooccurrence Matrix (GLCM) texture measures are determined on HV backscatter image for Odisha, for the year 2010. PALSAR HH, HV backscatter coefficient images, their difference (HHHV) and HV backscatter coefficient based eight textural parameters (Mean, Variance, Dissimilarity, Contrast, Angular second moment, Homogeneity, Correlation and Contrast) are used as input parameters for Support Vector Machines (SVM) tool. Ground based inputs for forest / non-forest were taken from field inventory data and high resolution Google maps. Results suggested significant relationship between HV backscatter coefficient and field based biomass (R2 = 0.508, p = 0.55) compared to HH with biomass values ranging from 5 to 365 t/ha. The spatial variability of biomass with reference to different forest types is in good agreement. The forest / nonforest classified map suggested a total forest cover of 50214 km2 with an overall accuracy of 92.54 %. The forest / non-forest information derived from the present study showed a good spatial agreement with the standard forest cover map of Forest Survey of India (FSI) and corresponding published area of 50575 km2. Results are discussed in the paper.

Large spatial extent of biomass burning occurs in northeast region of India during annual dry season for shifting cultivation purposes. Characterization of optical properties of resultant biomass burning aerosols is important for the study of atmospheric radiative process and for remote sensing of both surface and atmospheric properties in these regions. In the present study, physical and optical properties of biomass burning aerosols in Arunachal Pradesh, North Eastern Region of India have been studied for the first time using ground based measurements using a MICROTOPS-II sunphotometer, an Aethalometer, a quartz crystal microbalance impactor (QCM), SO2 analyser, and an UV meter. Aerosol size distribution suggested dominance of accumulation mode particle loading during burning days compared to normal days. The slope of data points between simultaneous measurements of AOD (500 nm) and UV ery suggested that every 0.1 increase in aerosol optical depth (AOD) causes 0.1 minimal erythemal dose (MED h − 1 ) reduction during normal day and reduction of 0.36 MED h − 1 in ground reaching UV ery during biomass burning periods. Diurnal variations of black carbon aerosol (BC) concentrations increased by a factor of ~ 2 during morning and evening hours compared to afternoon hours during biomass burning period. Daily average black carbon aerosol loading and SO 2 concentrations were found to be high during burning day compared to background values. The proportion of BC to total aerosol mass concentration was observed to be ~ 5% during normal days and ~ 14% during burning days. The changes in black carbon mass concentration values have implications for estimating radiative forcing due to aerosols over the region.

Northern regions of India experience severe fog conditions during the winter period (December–January) each year. In the present study, concurrent measurements of black carbon aerosols (BC), Carbon monoxide (CO), Surface Ozone (O 3 ) and oxides of Nitrgen (NO x ) have been carried out in Allahabad, located in northern India in order to understand the impact on pollutant concentrations during fog periods. The results of the present study show higher concentrations of BC, NO x and CO during morning and late night hours in the study area. Diurnal variations of ozone concentrations varied from 14 ppbv to 35 ppbv and were observed to increase gradually after sunrise attaining a maximum value by evening time and to decrease gradually thereafter. Concentrations of BC, NO x , O 3 and CO were found to be very low during fog periods over the study area. BC showed a positive correlation with CO and a negative correlation with surface ozone concentrations. The slope between black carbon aerosols and ground level ozone suggests that every 1 μg m − 3 increase in black carbon aerosol mass concentration causes a reduction of 0.7 μg m − 3 surface ozone.

Urban air quality monitoring plays an important role due to high concentration of particle sources and a large population exposed to elevated particle concentrations. Continuous ground based measurements of black carbon (BC) aerosol; carbon monoxide (CO) and ozone (O 3 ) were carried out in the tropical urban region of Hyderabad, India, during the forest fire season. Julian day variation of BC, CO and ozone showed high values on certain days. In order to ascertain the additional sources for observed high concentration of BC and CO, DMSP-OLS nighttime satellite data over the Indian region were processed for occurrence of forest fires. Results of the analysis suggested a higher incidence of forest fires on days with higher concentrations of BC and CO and a spatial distribution of forest fires; wind trajectories were observed to have a bearing on the higher values of BC, CO and ozone. Results are discussed in the paper.

Aerosol particle size distributions along with their spatial and temporal variability are important for describing both direct and indirect radiative forcing. In this study, the variation of black carbon (BC) aerosols, total aerosol mass loading and aerosol optical depth (AOD) over an urban region of Hyderabad, south India, was analyzed for 3 consecutive years from 2003 to 2005. The AOD was measured using a handheld multichannel sun-photometer at six wavelengths centered on 380, 440, 500, 675, 870 and 1020 nm and aerosol mass–size distribution was made using a quartz crystal microbalance (QCM) cascade impactor. In addition, satellite remote-sensing data from nighttime DMSP-OLS images were analyzed for inferring ancillary sources of aerosols. Results from temporal analysis (2004–2006) suggest that aerosol mass loading and BC mass concentration increased considerably over the 3-year time-period mainly due to increasing vehicular traffic from urban population growth. DMSP-OLS nighttime images for different years suggested higher forest fire occurrences in the year 2004 compared to other years. The annual mean AOD at 550 nm from moderate resolution imaging spectroradiometer (MODIS) showed relatively high values during 2004. Copyright © 2007 Royal Meteorological Society

Physical and optical properties of biomass burning aerosols in Northeastern region, India analyzed based on measurements made during February 2002. Large spatial extent of Northeastern Region moist tropical to moist sub-tropical forests in India have high frequency of burning in annual dry seasons. Characterization of resultant trace gases and aerosols from biomass burning is important for the atmospheric radiative process. Aerosol optical depth (AOD) observed to be high during burning period compared to pre- and post-burning days. Peak period of biomass burning is highly correlated with measured AOD and total columnar water vapor. Size distribution of aerosols showed bimodal size distribution during burning day and unimodal size distribution during pre- and post-burning days. Size distribution retrievals from biomass burning aerosols show dominance of accumulation mode particles. Weighted mean radius is high (0.22 microm) during burning period. Columnar content of aerosols observed to be high during burning period in addition to the drastic reduction of visibility. During the burning day Anderson sampler measurements showed dominance of accumulation mode particles. The diurnal averaged values of surface shortwave aerosol radiative forcing af biomass burning aerosols varies from -59 to -87 Wm(-2) on different days. Measured and modeled solar irradiances are also discussed in the paper.