Extraction, processing, production and display of geographic data
Heshmat Karami; Zahra Sayadi
Abstract
Extended Abstract
Introduction
Environmental changes are one of the most critical challenges to achieving sustainable development. Wetlands are part of the earth's structure and as one of the important ecosystems consisting of water, vegetation, soil and microorganisms. Monitoring, management and assistance ...
Read More
Extended Abstract
Introduction
Environmental changes are one of the most critical challenges to achieving sustainable development. Wetlands are part of the earth's structure and as one of the important ecosystems consisting of water, vegetation, soil and microorganisms. Monitoring, management and assistance in decision-making and policy-making of surface water changes can be done according to the availability of satellite data. The availability of Landsat data helps a lot in preparing a high-quality map to show the land surface changes. Although remote sensing is superior to traditional methods in terms of time, speed, and cost, these methods require the use of powerful and practical systems that include complex analysis. The use of data and images on the web is a solution that can be used to solve the mentioned problem, which studies can be done with high accuracy and speed without the need for a strong hardware and software system. The Google Earth Engine system creates suitable conditions for processing satellite images for environmental monitoring and analysis. The purpose of this research is to monitor the dynamic changes in the Miangaran wetland sub-basin in the period (2013-2022).
Materials & Methods
Miangaran wetland with an average area of 2500 hectares is located at a distance of one and a half kilometers from Izeh city, in the northeast of Khuzestan province. Time series analysis is one of the most common operations in remote sensing that helps to understand and model seasonal patterns as well as monitor changes. In this research, 421 images from the ee.ImageCollection ("LANDSAT/LC08/C02/T1_L2") data set were used for the period from 2013 to 2022. The construction of a harmonic model was used in this research due to its flexibility in cyclic calculation with simple and repeatable forms. The normalized differential water index is an index for drawing and monitoring content changes in surface waters. Also, the Normalized Difference Vegetation Index (NDVI) is one of the most common remote sensing indices. Harmonic time series of water body and vegetation cover were extracted using NDWI and NDVI indices in Google Earth Engine platform, and Mann-Kendall's non-parametric test was performed using time series data output with XLSTAT extension in Excel software. Finally, global water data was used to confirm and complete the results of time series analysis.
Results, discussion and conclusion
The results of the harmonic time series of the water body showed a decreasing and negative trend and more changes in the sub-basin. Kendall's statistical test confirmed the decreasing and negative trend of the water body. Accordingly, since the calculated p-value (<0.0001) is lower than the alpha significance level (0.05), the null hypothesis should be rejected and its alternative hypothesis, the existence of a trend in the time series, should be accepted. The value of Kendall's tau also confirmed a negative value (-0.245) and a decrease. Due to the negative sen's slope statistic for the water area (-0.002), changes are more in the Miangaran Wetland sub-basin. The results of the Mann-Kendall test for the observed vegetation data showed the absence of a trend in the harmonic time series. Since the calculated p-value (0.064) is higher than the significance level of alpha (0.05), the null hypothesis (absence of trend) cannot be rejected. The risk of rejecting the null hypothesis (while true) is 43.6%. Kendall's tau statistic showed a negative value (-0.060) and a non-significant decrease. Therefore, accepting the null hypothesis (absence of trend) indicates that vegetation changes in the harmonic time series were not significantly different from each other. Also, the negative sen's slope statistic for vegetation (-0.026) indicates more changes in the sub-basin of Miangaran Wetland. By comparing with the results and analysis of other researches, it seems that human intervention and change of land use can be the cause of the lack of trend in the Miangaran Wetland sub-basin. Also, according to the negative value of Man-Kendall's vegetation cover which showed a non-significant decreasing trend, it seems that climate change and drought have also played a major role in the changes under the Miangaran wetland basin. The study of the global water data also showed that the water occurrence in terms of space-time is decreasing and the intensity of the change of water occurrence is critical under the basin of Miangaran wetland. The marginal parts of Miangaran Wetland show seasonal water loss, most of these changes occur during the period. This research confirmed the use of harmonic time series in monitoring wetland dynamic changes. Finally, the allocation of water rights, the establishment of laws and the determination of the limit of the ecological bed, and the use of Google Earth Engine capabilities to monitor environmental changes (use, temperature, precipitation, evaporation, etc.) of the Miangaran Wetland sub-basin were suggested.
Remote Sensing (RS)
Heshmat Karami; Zahra Sayadi
Abstract
Extended AbstractIntroductionCoral reefs are one of the most diverse and ecologically important areas in the world. However, with increasing ocean temperatures, many coral reefs are severely threatened by bleaching events. When the water is too warm, corals expel the algae that live in their tissues, ...
Read More
Extended AbstractIntroductionCoral reefs are one of the most diverse and ecologically important areas in the world. However, with increasing ocean temperatures, many coral reefs are severely threatened by bleaching events. When the water is too warm, corals expel the algae that live in their tissues, causing the coral to turn completely white. When a coral bleaches, it is not dead, and corals can survive a bleaching event, but they are more stressed and at risk of dying. Today, in order to predict and identify areas at risk of coral bleaching, data based on satellite remote sensing are used. In this research, using 35-year data trends, the sea surface temperature in 2022 was predicted using ArcGIS Pro tools for the Persian Gulf area and possible areas exposed to thermal stress leading to coral bleaching were identified.Materials & Methods In order to predict the bleaching of corals, the research data archive of the American National Center for Atmospheric Research (NCAR) has been used. In this analysis, the harmonic method was used to fit the trend line. A harmonic trendline is a periodically repeating curved line that is best used to describe data that follows a cyclical pattern. For anomaly analysis parameters, the average monthly temperature in each location was compared with the overall average temperature to identify anomalies. There are three mathematical methods for calculating anomaly values with the Anomaly function, in this research, the method of difference From mean was used. At the end, the dimension value or band index was extracted, in which a certain statistic is obtained for each pixel in a multi-dimensional or multi-band raster, and the final map of coral bleaching prediction was prepared, and then using the data and global maps of the National Oceanic Administration NOAA , it was evaluated.Results, discussion and conclusionThe preliminary results showed that the sea surface temperature has changed in the Persian Gulf. The range has experienced higher average temperatures since 1996, which could put the area at risk of coral bleaching. The minimum average temperature in the studied time period is 298.758 degrees Kelvin in 1991 and the maximum average temperature in 1399 is 300.737 degrees Kelvin. The parameters that were chosen for multidimensional data trend analysis include water surface temperature variable (SST) and time dimension. The obtained trend map (1980-2015) indicated that the northwestern regions of the Persian Gulf and a part of its south are more exposed to prolonged heat. In this study, frequency parameter 2 was used in the harmonic model, which uses the combination of the first-order linear harmonic curve and the second-order harmonic curve to fit the data. The accuracy of data trend fitting by harmonic regression function provided statistical parameters, R2=0.78 and RMSE=0.5. The value of R2 indicates that the observed value of sea surface temperature (SST) was predicted by the harmonic regression model by 78% and the rest remains undefined. This value of the determination coefficient confirmed the accuracy of the trend map. Another statistical parameter is the root mean square error, the lower the value, the better the fit. In the obtained results, the mean of this error is 0.5, which shows that the harmonic regression model can accurately predict the data. In this study, forecast data was analyzed to find locations where water temperatures remain warm for extended periods of time. In this context, first, anomalies in the data were calculated, anomaly or anomaly is the deviation of an observed value from its average value, and in the analysis, it shows areas that have a temperature higher than the average. As a result of this step, the anomalies in the data were calculated and the areas with higher temperature than the average were identified. In the predicted annual time frame (2022), the north-west and a part of the south of the Persian Gulf region will face a longer period of high temperature. To evaluate the accuracy of the results obtained from the analysis and the method used in predicting sea surface temperature and identifying anomalies (2022-09-03), they were compared with the maps of Nova organization on the same date and were confirmed. It is suggested that responsible organizations use methods based on remote sensing and trend analysis to assess the situation and prepare a risk map of coral reefs.
