Monitoring of vegetation changes using daily Landsat-Modis simulated images at in three years of wet, normal and drought in arid region (Case study: Nimroze city)

Subject Areas : Natural resources and environmental management

Moien Jahantigh ¹ , Mansour Jahantigh ²

1 - PhD. Student of Watershed Management Science and Engineering, Department of Watershed Management , Faculty of Rangeland and Watershed Management, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, Iran
2 - Associate Professor, Department of Soil Conservation and Water Management, Sistan Agriculture and Edition Natural Resources Research Center, AREEO, Zabol, Iran

Received: 2021-06-17 Accepted : 2021-10-15 Published : 2023-01-21

Keywords: Vegetation changes detection, Landsat, ESTARFM model, Nimrozre, MODIS,

Abstract :

Background and Objective land degradation and desertification in arid areas are the most important environmental challenges in the world. This process due to the lack of precipitation and the occurrence of drought, while the unreasonable exploitation of natural and agricultural areas with increasing demand to provide human food needs, affects various environmental and socio-economic dimensions. So, the continuation of this condition during recent years with the destruction of vegetation and soil, wind and water erosion, soil salinity, soil compaction, and declining groundwater aquifers have significant consequences for the production of agricultural products and biodiversity in an arid region. Since the pattern and dimensions of vegetation changes are the most important factors in detecting land degradation, monitoring the vegetation changes is the best approach to analyzing land degrading and desertification trends in an arid region. Therefore, according to the capabilities of remote sensing data due to the wide coverage and multi-timed,  the use of satellite imagery to monitor vegetation changes by using vegetation index is one of the best methods that developed in recent years. Moreover, concurrent access to high spatial and temporal resolution imageries is one of the important factors that affect the monitoring of vegetation changes. To achieve this goal, It needs to incorporate different satellites with high spatial (e.g., Landsat satellite) and temporal (e.g., MODIS satellite) images. The purpose of this study is the monitoring vegetation changes using daily Landsat simulated images at 30 m Spatial Resolution in three years of wet, normal, and drought in the Nimroze area.Materials and Methods The study area is located in the north of the Sistan and Baluchistan provinces. Low precipitation (50 mm), high temperature (48 oC), high evaporation (5 m), and 120-day winds are among the specific climatic conditions that characterize this region. In this study, at first, the hydrological drought status of the Hirmand River was investigated. Using the Hydrostats package in R software, the amount of threshold of flood by running the related codes (by running codes such: daily.cv, ann.cv, high. spell, and low. spell) during the statistical period of study (29 years) was calculated. To determine wet, normal, and drought years calculated the length of periods that flood is higher (high. spell. lengths) and lower (low. spell. lengths) than the threshold. To increase the accuracy of monitoring vegetation changes, it needs to incorporate different satellites with high spatial (e.g., Landsat) and temporal (e.g., MODIS) images. To achieve this purpose, in this study, the Enhanced spatial and temporal adaptive reflectance fusion model (ESTARFM) was evaluated with actual satellite data (OLI, ETM+, TM image). For this purpose at first, pre-processing (geometric, radiometric, and atmospheric correction) was performed on satellite images, and by using the ESTRFM model, simulated daily Landsat images at 30 m spatial resolution for wet, normal, and drought years. In-field operations from different plant communities by GPS were sampled. Comparing filed data with the Normalized difference vegetation index (NDVI) and the soil-adjusted vegetation index (SAVI), the vegetation index that had the highest correlation with field data was selected. To investigate vegetation changes, using the vegetation index (the vegetation index with high correlation), the map of vegetation for each year was prepared (wet, normal, and drought years). After the classification maps of vegetation, by comparison, approach (cross tab), the map of vegetation changes was extracted.Results and Discussion The results of analyzing wet and dry periods showed that, flood volume in dry years compare to normal and wet years decreased 31 and 82 percentages, respectively. To incorporation MODIS and Landsat (OLI, ETM+, TM) Images, using enhanced spatial and temporal adaptive reflectance fusion model (ESTARFM), finding indicate that this model improves the accuracy of predicted fine-resolution reflectance and preserves spatial details for heterogeneous landscapes too. So that the mean coefficient of determination (R2) of blue, green, red and near-infrared estimation bands with actual satellite images data is 0.91, 0.89, 0.92 and 0.91 respectively. Also the average Root-Mean-Square Error (RMSE) in four bands obtained 0.01, 0.027, 0.028 and 0.031 successively. Comparing the obtained field data with the Normalized difference vegetation index (NDVI) and the soil adjusted vegetation index (SAVI), indicate that SAVI index has the highest correlation (R2= 87) with vegetation of study region. By calculate the regression model (using SAVI and field data) and classify the vegetation maps of wet, normal and drought years, 6 class obtained (class1=0-10%, class2=20-10%, class3=20-30%, class4=40-50%, class5=60-80% and class6=>80%). The results of investigation vegetation changes indicate that during the drought period 70% of study area has less than 10% vegetation (equal to 138176.3 hectares) and during normal and wet years by increasing vegetation, this area decreased by 30 and 48% respectively (equal to 66269.98 and 50559.7 hectares, respectively). According to the results during the study period, the most vegetation changes is relate to conversion of class 1 to class 2 (equivalent to 48.5%). moreover 18 and 27% of vegetation changes relate to class 1 and 2 to class 4 and 5 respectively (equal to 16284.26 and 11471.88 hectares, respectively). Also the finding indicates that the most vegetation changes occurrence in wetland-forest (28%), forest-rangeland (21%) and poor rangeland (19%) land uses respectively. Field study also showed that, the most important plant species that grows in this land-use such as the results of analyzing wet and dry periods showed that flood volume in dry years compare to normal and wet years decreased by 31 and 82 percent, respectively. To incorporation MODIS and Landsat (OLI, ETM+, TM) Images, using enhanced spatial and temporal adaptive reflectance fusion model (ESTARFM), the finding indicates that this model improves the accuracy of predicted fine-resolution reflectance and preserves spatial details for heterogeneous landscapes too. So that the mean coefficient of determination (R2) of blue, green, red, and near-infrared estimation bands with actual satellite images data is 0.91, 0.89, 0.92, and 0.91 respectively. Also, the average Root-Mean-Square Error (RMSE) in four bands obtained 0.01, 0.027, 0.028, and 0.031 successively. Comparing the obtained field data with the Normalized difference vegetation index (NDVI) and the soil-adjusted vegetation index (SAVI), indicate that the SAVI index has the highest correlation (R2=87) with the vegetation of the study region. By calculating the regression model (using SAVI and field data) and classifying the vegetation maps of wet, normal, and drought years, 6 classes obtained (class1=0-10%, class2=20-10%, class3=20-30%, class 4=40-50%, class5=60-80% and class6=>80%). The results of the investigation of vegetation changes indicate that during the drought period, 70% of the study area has less than 10% vegetation (equal to 138176.3 hectares) and during normal and wet years by increasing vegetation, this area decreased by 30 and 48% respectively (equal to 66269.98 and 50559.7 hectares, respectively). According to the results during the study period, most vegetation changes are related to the conversion of class 1 to class 2 (equivalent to 48.5%). moreover, 18 and 27% of vegetation changes relate to class 1 and 2 to class 4 and 5 respectively (equal to 16284.26 and 11471.88 hectares, respectively). Also, the finding indicates that the most vegetation changes occur in wetland-forest (28%), forest-rangeland (21%), and poor rangeland (19%) land use respectively. The field study also showed that the most important plant species that grow in this land use such as Aeluropus littoralis, Chenopodiace sp, Tamarix aphylla, Haloxylon aphylum are adaptive to climatic regime in study area.Conclusion In this research for the first time in the Nimroz region of Sistan Vegetation changes were studied using Landsat simulated images during periods of low water, normal, and high water years. Due to low rainfall and harsh climate in the study area, floods in the Helmand River are the only source of water supply required in the study area. The results of analyzing wet and dry periods showed that flood volume in dry years compared to normal and wet years has decreased by 31 and 82, respectively. According to the reduction of flood volume during a drought year, 70% of the study area has poor vegetation and during normal and wet years, providing plants with water needs and increasing vegetation, this area had decreased by 30% and 48%, respectively. According to the results of this study, change in hydrological conditions of the Hirmand River has a significant role in vegetation changes in the study area by using simulated images with high spatial and temporal resolution can improve the accuracy of monitoring vegetation changes to control and management the desertification in Sistan area.

References:

Akumu CE, Amadi EO, Dennis S. 2021. Application of drone and worldview-4 satellite data in mapping and monitoring grazing land cover and pasture quality: Pre-and post-flooding. Land, 10(3): 321. https://doi.org/10.3390/land10030321.

Aqil T, Shu H. 2020. CA-Markov chain analysis of seasonal land surface temperature and land use land cover change using optical multi-temporal satellite data of Faisalabad, Pakistan. Remote Sensing, 12(20): 3402. https://doi.org/10.3390/rs12203402.

Bond N. 2015. Hydrostats: Hydrologic indices for daily time series data. R package version 02, 4: 16. https://CRAN.R-project.org/package=hydrostats.

Boubacar S, Jarju AM, Sonko E, Yaffa S, Sawaneh M. 2021. Detection of recent changes in Gambia vegetation cover using time series MODIS NDVI. Belgeo Revue belge de géographie(1). https://doi.org/10.4000/belgeo.47995.

Burrell A, Evans J, De Kauwe M. 2020. Anthropogenic climate change has driven over 5 million km2 of drylands towards desertification. Nature communications, 11(1): 3853. https://doi.org/10.1038/s41467-020-17710-7.

Chen Y, Cao R, Chen J, Zhu X, Zhou J, Wang G, Shen M, Chen X, Yang W. 2020. A new cross-fusion method to automatically determine the optimal input image pairs for NDVI spatiotemporal data fusion. IEEE Transactions on Geoscience and Remote Sensing, 58(7): 5179-5194. https://doi.org/10.1109/TGRS.2020.2973762.

Ebrahimikhusfi Z, Khosroshahi M, Naeimi M, Zandifar S. 2019. Evaluating and monitoring of moisture variations in Meyghan wetland using the remote sensing technique and the relation to the meteorological drought indices. Journal of RS and GIS for Natural Resources, 10(2): 1-14. http://dorl.net/dor/20.1001.1.26767082.1398.10.2.1.0. (In Persian).

Elhag M, Boteva S, Al-Amri N. 2021. Forest cover assessment using remote-sensing techniques in Crete Island, Greece. Open Geosciences, 13(1): 345-358. https://doi.org/10.1515/geo-2020-0235.

Gao F, Hilker T, Zhu X, Anderson M, Masek J, Wang P, Yang Y. 2015. Fusing Landsat and MODIS data for vegetation monitoring. IEEE Geoscience and Remote Sensing Magazine, 3(3): 47-60. https://doi.org/10.1109/MGRS.2015.2434351.

Gavrilescu M. 2021. Water, soil, and plants interactions in a threatened environment. Water, 13(19): 2746. https://doi.org/10.3390/w13192746.

Guan X, Huang C, Zhang R. 2021. Integrating MODIS and Landsat data for land cover classification by multilevel decision rule. Land, 10(2): 208. https://doi.org/10.3390/land10020208.

Hongwei Z, Wu B, Wang S, Musakwa W, Tian F, Mashimbye ZE, Poona N, Syndey M. 2020. A synthesizing land-cover classification method based on Google Earth engine: A case study in Nzhelele and Levhuvu Catchments, South Africa. Chinese Geographical Science, 30: 397-409. https://doi.org/10.1007/s11769-020-1119-y.

Hu P, Sharifi A, Tahir MN, Tariq A, Zhang L, Mumtaz F, Shah SHIA. 2021. Evaluation of vegetation indices and phenological metrics using time-series modis data for monitoring vegetation change in Punjab, Pakistan. Water, 13(18): 2550. https://doi.org/10.3390/w13182550.

Jahantigh M, Jahantigh M. 2019. Study effect of flood productivity on vegetation changes using field work and Landsat satellite images (Case study: Shandak of Sistan region). Journal of RS and GIS for Natural Resources, 10(4): 57-73. http://dorl.net/dor/20.1001.1.26767082.1398.10.4.4.7. (In Persian).

Jahantigh M, Najafi Nejad A, Jahantigh M, Hosienali Zadeh M. 2020. Investigating the effect of hydrological drought and traditional utilization (distribution and transmission) of water resources (flood streams) on land degradation and desertification in drylands: a case study of sistan plain. Desert Ecosystem Engineering Journal, 9(27): 25-46. https://doi.org/10.22052/deej.2020.9.27.21. (In Persian).

Kempf M. 2021. Monitoring landcover change and desertification processes in northern China and Mongolia using historical written sources and vegetation indices. Climate of the Past Discussions: 1-29. https://doi.org/10.5194/cp-2021-5.

Knauer K, Gessner U, Fensholt R, Kuenzer C. 2016. An ESTARFM fusion framework for the generation of large-scale time series in cloud-prone and heterogeneous landscapes. Remote Sensing, 8(5): 425. https://doi.org/10.3390/rs8050425.

Liu H, Gong P, Wang J, Clinton N, Bai Y, Liang S. 2020. Annual dynamics of global land cover and its long-term changes from 1982 to 2015. Earth System Science Data, 12(2): 1217-1243. https://doi.org/10.5194/essd-12-1217-2020.

Majeed M, Tariq A, Anwar MM, Khan AM, Arshad F, Mumtaz F, Farhan M, Zhang L, Zafar A, Aziz M. 2021. Monitoring of land use–Land cover change and potential causal factors of climate change in Jhelum district, Punjab, Pakistan, through GIS and multi-temporal satellite data. Land, 10(10): 1026. https://doi.org/10.3390/land10101026.

Procházková E, Kincl D, Kabelka D, Vopravil J, Nerušil P, Menšík L, Barták V. 2020. The impact of the conservation tillage “maize into grass cover” on reducing the soil loss due to erosion. Soil and Water Research, 15(3): 158-165. https://doi.org/10.17221/25/2019-SWR.

Sulimin V, Shvedov V, Lvova M. 2021. Natural resource potential and sustainable development of the regional system. In: E3S Web of Conferences. EDP Sciences, p 02034. https://doi.org/02010.01051/e02033sconf/202129102034.

Tang Q. 2020. Global change hydrology: Terrestrial water cycle and global change. Science China Earth Sciences, 63(3): 459-462. https://doi.org/10.1007/s11430-019-9559-9.

Tewabe D, Fentahun T. 2020. Assessing land use and land cover change detection using remote sensing in the Lake Tana Basin, Northwest Ethiopia. Cogent Environmental Science, 6(1): 1778998. https://doi.org/10.1080/23311843.2020.1778998.

Wang Q, Zhang Y, Onojeghuo AO, Zhu X, Atkinson PM. 2017. Enhancing spatio-temporal fusion of MODIS and Landsat data by incorporating 250 m MODIS data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 10(9): 4116-4123. https://doi.org/10.1109/JSTARS.2017.2701643.

Wang SW, Gebru BM, Lamchin M, Kayastha RB, Lee W-K. 2020. Land use and land cover change detection and prediction in the Kathmandu district of Nepal using remote sensing and GIS. Sustainability, 12(9): 3925. https://doi.org/10.3390/su12093925.

Wang Y, Luo X, Wang Q. 2021. A boundary finding-based spatiotemporal fusion model for vegetation index. International Journal of Remote Sensing, 42(21): 8236-8261. https://doi.org/10.1080/01431161.2021.1976870.

Wang Y, Xie D, Zhan Y, Li H, Yan G, Chen Y. 2021. Assessing the accuracy of landsat-MODIS NDVI fusion with limited input data: A strategy for base data selection. Remote Sensing, 13(2): 266. https://doi.org/10.3390/rs13020266.

Wang Y, Yan G, Hu R, Xie D, Chen W. 2020. A scaling-based method for the rapid retrieval of FPAR from fine-resolution satellite data in the remote-sensing trend-surface framework. IEEE Transactions on Geoscience and Remote Sensing, 58(10): 7035-7048. https://doi.org/10.1109/TGRS.2020.2978884.

Xiao R, Liu Y, Huang X, Shi R, Yu W, Zhang T. 2018. Exploring the driving forces of farmland loss under rapidurbanization using binary logistic regression and spatial regression: A case study of Shanghai and Hangzhou Bay. Ecological Indicators, 95: 455-467. https://doi.org/10.1016/j.ecolind.2018.07.057.

Yang J, Huang X. 2021. The 30 m annual land cover dataset and its dynamics in China from 1990 to 2019. Earth System Science Data, 13(8): 3907-3925. https://doi.org/10.5194/essd-13-3907-2021.

Yu Q, Liu W, Gonçalves WN, Junior JM, Li J. 2021. Spatial Resolution Enhancement for Large-Scale Land Cover Mapping via Weakly Supervised Deep Learning. Photogrammetric Engineering & Remote Sensing, 87(6): 405-412. https://doi.org/10.14358/PERS.87.6.405.

Zhou X, Geng X, Yin G, Hänninen H, Hao F, Zhang X, Fu YH. 2020. Legacy effect of spring phenology on vegetation growth in temperate China. Agricultural and Forest Meteorology, 281: 107845. https://doi.org/10.1016/j.agrformet.2019.107845.

Zhu X, Helmer EH, Gao F, Liu D, Chen J, Lefsky MA. 2016. A flexible spatiotemporal method for fusing satellite images with different resolutions. Remote Sensing of Environment, 172: 165-177. https://doi.org/10.1016/j.rse.2015.11.016.