The air in African cities is increasingly polluted mainly due to human activities. A bioindication technical of urban air quality based on active remote sensing might be an alternative to existing physico-chemical methods. Reflectance measurements in the visible spectrum have been carried out at the adaxial and abaxial sides of Ficus benjamina L. leaves in the city of Abidjan, C ôte d’Ivoire, with a precision digital camera. Leaves were collected in industrial zones and in parks. The impact of air pollution on leaf physiological as well as structural characteristics in these two contrasts urban environments was determined by Dorsiventral Leaf Reflectance Correlation (DRLC) and dorsiventral leaf reflectance asymmetry quantitatively defined with Normalized Dorsiventral Asymmetry Index (NDAI). Species leaf susceptibility to air pollution from season to season was determined by NDAI seasonal variation. Leaf reflectance measurements allowed the estimation of environmental stress level among industrial areas and parks. NDAI and DLRC were significantly higher in industrial zones compared to parks. NDAI values were found significantly higher for major rainy season compared to major dry season, indicating probably that F. benjamina leaf structure changes were increased from one season to another. Thereby, assessment of urban air quality can be done using leaves reflectance in the visible spectrum.
Urbanized and industrialized areas are known to be subjected to high concentrations of air pollutants [
The main sources in the urban environment are road traffic and industrial activity [
Current physico-chemical techniques for the determination and monitoring of air quality are expensive for developing cities [
Atmospheric pollutants can induce changes in leaf structure and physiology [
The use of spectral and remote sensing techniques for detection of plant stress is based on the assumption that stress factors interfere with leaf structural and physiological properties, influencing the spectral leaf characteristics that can be detected in the different regions of the electromagnetic spectrum [
The objective of this study is the evaluation of leaf reflectance parameters in the visible spectrum as an indicator for assessing urban environment quality and the investigation of species leaf susceptibility to air pollution from one season to another.
The city of Abidjan (5˚00'N - 5˚30'N, 3˚50'W - 4˚10'W), Côte d’Ivoire, with distinct land use classes and spatially varying habitat quality [
Abidjan is characterized by a high level of industrialization and urbanization. The city has a significant growing old automobile park, of which 70% are secondhand vehicles. Many parks and green spaces were preserved in the city, but these parks disappear quickly due to urbanization. Actually, Abidjan is one of the most polluted cities of sub-Saharan Africa.
Samples were collected in areas with contrasting environment quality. We sampled in industrial zones (2 sites) and in parks (2 sites) which were relatively far away from traffic and industrial activity. Sampling locations were determined based on the availability of plant species tested.
We have chosen Ficus benjamina L. as a study species because of its homogeneous spatial distribution in the city. F. benjamina is a tropical dicotyledonous belonged to the family Moraceae. It is an evergreen tree with several spreading branches from the base, with oval and glossy leaves of 2 - 5 cm wide. This tree species is hypostomatous (i.e., it only have stomata on the abaxial leaf surface) with dorsiventral structure (i.e., the adaxial surface is adjacent to the palisade parenchyma layer, while the abaxial surface is adjacent to the spongy mesophyll layer). Moreover, the leaves of F. benjamina have abaxial wax and smooth cuticle.
Two sampling locations were selected at each site according to accessibility and geographical distribution criteria. At each sampling location 2 trees were selected. We sampled 9 mature and undamaged leaves from each tree at a height between 1.5 m and 2.5 m above ground level. Sampling campaign was conducted during six month on the same trees species. A first sampling campaign was conducted from February until April 2012 for major dry season. The mean air temperature during this period was 28.3˚C, air humidity was 83%, precipitation was 43.18 mm and mean wind speed was 11.7 m∙s−1. A second campaign was organized from May until July 2012 for major rainy season. The mean air temperature during this period was 26˚C, while air humidity was 89.2%. Precipitation and mean wind speed were respectively 375.91 mm and 12 m∙s−1. Overall, we sampled 432 leaves.
The field level leaf reflectance measurements were performed under standard illumination conditions using a field illumination set up as described by Khavanin Zadeh et al. [
Leaf reflectance variation can indicate environmental stress difference between land use classes studied.
DLRC expresses the relationship between the adaxial and abaxial leaf reflectance which is characterized by the slope and the coefficient of determination obtained from the linear regression analysis between abaxial and adaxial leaf reflectance.
Dorsiventral leaf reflectance asymmetry can be quantitatively defined with NDAI, which is defined as a linear combination of the leaf reflectances at both the adaxial and abaxial leaf sides [
NDAI as well as DLRC can be affected both by changes in leaf structural as well as physiological characteristics caused by environment quality.
Species leaf susceptibility to air pollution from season to season was determined by NDAI variation from major dry season (February-April) to major rainy season (May-July) in industrial zones and parks in different visible spectral bands.
All data were analyzed using Statistica software, version 7.1 (StatSoft Inc., 1984-2005). The normality of the data was tested with a Shapiro-Wilk test. Descriptive statistics, including mean and standard deviation were calculated for leaf reflectance parameters. Means were compared by using a one-way analysis of variance (ANOVA) procedure and a Tukey-HSD test. Linear regression analysis was performed to identify the relationship between adaxial and abaxial leaf reflectance. Species leaf susceptibility to air pollution from season to season was assessed using the Student-t-test. Differences were considered significant at p < 0.05.
F. benjamina leaf reflectance values for both leaf sides are higher in Parks (P) in comparison with industrial zones (IZ) for all spectral bands according to
Land use class | Spectral band | Adaxial leaf reflectance | Abaxial leaf reflectance |
---|---|---|---|
Industrial zones | Red | 25.03 ± 5.23a | 36.46 ± 4.27b |
Green | 38.91 ± 5.57a | 49.83 ± 4.01b | |
Blue | 18.19 ± 3.76a | 23.79 ± 3.91b | |
Parks | Red | 29.34 ± 2.49a | 37.15 ± 2.44b |
Green | 42.67 ± 1.51a | 51.72 ± 2.96b | |
Blue | 22.27 ± 1.64b | 26.53 ± 1.62b |
decreasing leaf thickness. Arriaga et al. [
For F. benjamina, NDAI showed a different behavior than leaf reflectance in land use classes. It significantly decreased in P and increased in IZ for the red and blue bands. Khavanin Zadeh et al. [
The linear regression between abaxial and adaxial leaf reflectance show strong and higher correlation (R2 ˃ 0.50, p < 0.05) in industrial zones (IZ) compared to parks (P) for all spectral bands, while the slopes of the linear relation are lower in P than IZ except for green band (
As illustrated in
Land use class | Spectral band | NDAI |
---|---|---|
Industrial zones | Red | 0.19 ± 0.05b |
Parks | Red | 0.12 ± 0.04a |
Industrial zones | Green | 0.13 ± 0.04a |
Parks | Green | 0.10 ± 0.02a |
Industrial zones | Blue | 0.14 ± 0.04b |
Parks | Blue | 0.09 ± 0.02a |
authors recorded that DRLC shows a more stable variation than NDAI in polluted sites for the species considered, and that it seems to be species-independent. DLRC increasing in blue and red bands in IZ might be a direct reflection of environmental pollution impact on the abaxial versus the adaxial leaf reflectances [
According to Dineva [
Species leaf susceptibility to air pollution from season to season was determined by NDAI seasonal variation. The results show that NDAI values are higher for major rainy season (MRS) compared to major dry season (MDS) in parks (
NDAI seasonal showed same trends from one land use class to another, higher in the MRS and lower in the MDS, but significantly in parks as observed in (
Leaf reflectance as well as NDAI and DLRC in the visible spectral bands can be good indicators to estimate differences in urban environment quality. Results revealed effects of air pollution on anatomical and physiological leaf characteristics in IZ in comparison with P. NDAI significant increasing during the major rainy season probably indicated that F. benjamina leaf structure changes were increased from one season to another, and was probably related to species. These results make it possible to consider an operational approach for assessing and monitoring of urban environment quality based on biomonitoring technique and an active remote sensing. However, the suitability of these results still requests some further research such as anatomical studies in tropical area.
The second author is a beneficiary of a mobility grant from the Belgian Federal Science Policy Office (BELSPO) co-funded by the Marie Curie Actions from the European Commission. This work was support by grants of Foundation David and Alice Van Buuren (Université Libre de Bruxelles) for digital camera purchase as well as University of Antwerp (Prof. Roeland Samson) and “Vlaamse Interuniversitaire Raad” for the reflectance mea- surement setup making and Ivorian institution “Programme d’ Appui Stratégique à la Recherche Scientifique” (PASRES) to the third author. We also thank the city council of “District Autonomed’ Abidjan” for their as- sistance in achieving field data.
Zamblé Fidèle Tra Bi,Yao Sadaiou Sabas Barima,Djédoux Maxime Angaman,Ali Reza Khavanin Zadeh,Karidia Traoré, (2016) Environmental Pollution Bioindication Based on Ficus benjamina L. Leaf Reflectance in the City of Abidjan, Côte d’Ivoire. Open Journal of Air Pollution,05,55-63. doi: 10.4236/ojap.2016.52006