3rd Young Earth Scientist Congress Dar es Salaam, Tanzania from 11th - 14th August 2014 RETRIEVAL OF ASPHALT AGGREGATE EXPOSITION TO THE ATMOSPHERE BY MIVIS AIRBORNE IMAGERY TO SUPPORT ENVIRONMENTAL ANALYSIS. A. Mei 1 , C. Manzo 1 ,G. Fontinovo 1 , A. Allegrini 1 , N. Fiore 2 , C. Bassani 1 , A. D’Andrea 2 , R. Salvatori 1 1 Institute of Atmospheric Pollution Research - CNR (National Research Council of Italy), Research Area of Rome 1 Via Salaria Km 29.300, Monterotondo (00016), Rome, Italy. corresponding author: mei@iia.cnr.it 2 Department of Civil, Construction, and Environmental Engineering, Sapienza-University of Rome, Via Eudossiana 18, I-00184 Rome, Italy Abstract The analysis of superficial bitumen removal on asphalted surfaces is an important issue for road network management and environmental studies related to asphalt wear and pollution transport. The calculation of the Exposed Aggregate Index (EAI) by picture Digital Imaging Processing (DIP) allows to quantify in each frame superficial bitumen removal and aggregates exposure. A procedure based on non-parametric classification process of digital images gives a fast response of EAI and allow to make a correlation with field spectral data. Finally, asphalt bitumen removal was analysed by the use of spectral indices through MIVIS airborne imagery and EAI map was provided. Introduction Mechanic action due to vehicle transit and natural worsening of asphalts may cause changes on surface composition and aggregate reorganization. The major constituent of asphalt is the aggregate fraction while the rest consists of about 5% of bitumen by weight of the total mix. A significant source of sub-micrometer fine particles are produced by the road–tire interface (Lindgren 1996). Wear degree of road surface can therefore affect the contribution associated with road particles emissions/absorption in the environment. Considering that surface area of exposed aggregates is the cause that most influences the production of wear particles and adsorption capacity of metals, the quantification of the aggregate-atmosphere interface can be an important issue to estimate the particle emissions originated from the road-tire interface. Methods Field data. For the determination of aggregate exposure, Digital Imaging Processing (DIP) of field pictures and spectral signatures have been used. After a pre-processing step for normalization and co-registration of frames, a supervised non- parametric classification was performed to retrieve the Exposed Aggregates Index (EAI) (1) (Manzo et al. 2014): EAI (%) = Ip/Tp *100 (1) Where, Ip is the number of pixels corresponding to aggregates exposed on the asphalt surface which causes the appearance of spectral characteristics of outcropping aggregate fragments; Tp correspond to the Area Of Interest area expressed as number of pixel. Its calculation allow to quantify the area of exposed aggregates to the atmosphere in each frame descriptive of the superficial bitumen removal (Figure 1). Different equations evaluated the relationship between EAI and spectral response. From those exanimate, ICNR spectral index (2) show higher statistical correlation value (rs=0.83): ICNR = 0.0075*exp (11.49((VIS2dif * IMEI)+ λ830nm + λ740nm))) (2) Where VIS2dif is calculated by reflectance value at 830 nm minus reflectance at 490 nm (Herold et al. 2005) and IMEI is calculated by the reflectance difference between 740 nm and 460 nm (Mei et al. 2014a). Also the integration of reflectance (Ir) values in the visible range (440-680 nm) vs EAI show R 2 of 0.75 (Mei et al 2014b). Multispectral Infrared and Visible Imaging Spectrometer (MIVIS) Remote sensing data. The Multispectral Infrared and Visible Imaging Spectrometer (MIVIS) is an airborne instrument consisting of 4 spectrometers. These simultaneously measure the radiation in the visible (20 bands between 430 and 830 nm), near (8 bands between 1150 and 1550 nm), medium (64 bands between 2000 and 2500 nm) and the thermal infrareds (10 bands between 8200 and 12700 nm) for a totality of 102 bands (Figure 2). For this study a MIVIS image of the Campania District (Italy) was adopted to retrieve EAI map by the use of spectral indices. Figure 1. Work flow for field data processing Figure 2. MIVIS instrument Data input Geometric correction Stack layers RGB normalization Spectral signatures EXPOSED AGGREGATE INDEX COMPUTATION SPECTRAL INDICES COMPUTATION CORRELATION BETWEEN SPECTRAL INDICES AND EAI REMOTE SENSED IMAGERY APPLICATION Spectral indices Ir ICNR IMEI VIS2dif RMSE (%) 1.9 1.7 4.3 4.7 Pearson ‘s correlation 0.95 0.96 0.76 0.71 Conclusions. The EAI was retrieved by MIVIS airborne image using spectral indices in order to extend the aggregate exposition to the atmosphere in large areas and, consequently, to define the superficial bitumen removal. This kind of analysis can effectively support transport pollution studies by quantifying the aggregate-tire/atmosphere able to produce particle emissions or, vice versa, to absorb metal ions. Further research will be focused on the calculation of EAI through a most conspicuous remote sensed dataset and, also, through samples laboratory analysis. Results By the use of MIVIS bands 2 and 16, the angular coefficient descriptive of asphalt was retrieved and adopted for Spectral Angle Mapper (SAM) classification (Figure 3-B) of paved roads and parking lots (Mei et al. 2014a). An asphalt macro class was built and then transformed in an mask (Figure 3-C) to apply on MIVIS image. ICNR (Figure 3-D) , VIS2dif, IMEI and Ir were computed in masked areas. Graph 1 show predicted values of EAI and those calculated in six available field targets. Best values of RMSE between EAI predicted and observed are shown by ICNR (1.7%) and a Pearson correlation of 0.96. Applying ICNR, EAI map was finally provided (Figure3-E). Figure 3. MIVIS image processing steps. A) MIVIS image, B) Asphalt macro class by SAM classification, C) Masking MIVIS image, D) ICNR computation and E) EAI map. A B C D E 0-5% 5-10% 10-15% 15-20% 20-25% 25-30% >30% Exposed Aggregate Index map Graph 1. a) histograms of observed and predicted EAI values and b) RMSE and correlation values of spectral indices computation References Lindgren, A. (1996) Asphalt Wear and Pollution Transport. The Science of the Total Environment, 189-190, 281-286. Manzo C., Mei A., Salvatori R., Bassani C., Allegrini A. Spectral modelling used to identify the aggregates index of asphalted surfaces and sensitivity analysis. Construction and Building Materials 2014, 61, 147–155. Mei, A.; Salvatori, R.; Fiore, N.; Allegrini, A.; D'Andrea, A. (a). Integration of Field and Laboratory Spectral Data with Multi-Resolution Remote Sensed Imagery for Asphalt Surface Differentiation. Remote Sensing 2014, 6, 2765-2781. Herold, M. and Roberts, D. (2005) Spectral Characteristics of Asphalt Road Aging and Deterioration: Implications for Remote-Sensing Applications. Applied Optics, 44, 4327-4334. Mei, A.; Manzo, C.; Bassani, C.; Salvatori, R., Allegrini, A. (b). Bitumen Removal Determination on Asphalt Pavement Using Digital Imaging Processing and Spectral Analysis. Open Journal of Applied Sciences 2014, 4 (6).