ABSTRACT
Smartphones are state-of-the-art devices with several interesting features which make them promising for analytical purposes. After modification to a spectrophotometer (smart spectrophotometer), they can be utilized for the quantitative or qualitative applications. Although smartphones have widely been applied for sensing∖biosensing purposes, the error structure/type of their outputs remained unexplored. Error structure information values the objects/channels in a given data set and variables have the same importance when the noise has identical independent distribution (i.i.d). Otherwise, error structure weights them for further data analysis. In this contribution, a smartphone-based spectrophotometer was constructed integrating simple optical elements-a tungsten lamp as source and a piece of digital versatile disc (DVD) as a reflecting diffraction grating to investigate the error sources of the smartphone-spectrophotometer. For this purpose, error covariance matrices (ECMs) were calculated using a series of replication capturing error information. Afterwards, PCA and MCR-ALS were employed for the decomposition of the ECMs and resolved profiles were translated to the error types. Finally, proportional error as a heteroscedastic noise was highlighted as the most important source of variation in the error structure of the smartphone-based spectrophotometer.
ABSTRACT
Straight-chain alkanes (n-alkanes) have been proposed as biomarkers to assess petroleum pollution sources in marine environments. In this work, three sampling sites were selected along Iranian-protected mangrove forests in the Persian Gulf. Sample sites were chosen to represent different compositions and sources of n-alkanes in surface sediments (0-5 cm) from different locations in the Khamir Port, in the middle part of the Gulf, and in the Qeshm Island. Samples were analyzed by gas chromatography-mass spectrometry (GC-MS), and the obtained n-alkanes concentrations were analyzed by principal component analysis (PCA) and multivariate curve resolution alternating least squares (MCR-ALS) to deduce their possible sources and distribution patterns. Results revealed that n-alkanes have a dominant biogenic source in marginal Qeshm Island stations, which changed from biogenic to petrogenic sources in the Khamir Port areas. Sediment samples from the Khamir Port showed the significant presence of oil pollution due to transportation. These samples are being exposed to basic Persian Gulf currents entering from the Hormuz Strait causing oil pollution spread over the entire Gulf area. Sediment samples of the Middle Part stations located between two other sites show a combined condition of the two previous sections, with petrogenic and biogenic contributions. The present study demonstrates that the Hara Protected Area was already contaminated by background oil pollution as a result of continuous oil spills and war conflicts in the Persian Gulf.Additionally, the MCR-ALS method is shown to be a powerful chemometric tool for the investigation, resolution, identification, and description of pollution patterns distributed over a particular geographical area and environmental compartment. They can be used as well as parameters like unresolved to resolved ratio (U/R), pristane to phytane (Pr/Ph), n-C17/Pr, n-C18/Ph, and unresolved complex mixture (UCM) to assess petroleum pollution sources in the sediments.
