Landscape Health Mapping by Landsat images

Landscape Health Mapping by Landsat images



Introduction:
The idea of ecosystem health assessment (EHA) was introduced in environmental management in the late 1980s. In the discussion of health assessment, indicators are raised. Indicators provide a better picture of the environment Ecological indicators are committees that are closely related to the complex characteristics of the ecosystem. But these indicators are often not directly measurable. Measuring these indicators is used to simplify and evaluate various aspects of ecosystem performance. Examples of these indicators are: Vegetation density index, fold index, continuity index, Euclidean distance or distance from the nearest neighbor, environment to area ratio and spot area. The aim is to prepare a health map of the study area. In this process, after detecting changes, the thresholds of health, confusion, and disorder are determined. And health disruptors are identified. At the end, health diagnosis instructions are provided. The main question of the research is whether the health of the land has changed in the desired period of time?

Materials and Methods:
The study area in this study is sub-basins of Qarahsoo, Nekarood and Gorganrood watersheds including Gorgan, Kordkuy and Bandar Gaz counties. The health utility map was obtained using the weighted linear combination (WLC) method. Thus, the health maps of 1984, 2000 and 2018 were prepared. By comparing the three maps, health changes in the study period were examined. In these layers, more desirability indicates a higher degree of power and less desirability indicates a lower degree of power for the health of the landscape. The number zero indicates the worst state of health and the number 255 indicates the best state of health in this study. In the next step, the health threshold is determined and also the health disruptors (such as human development and diseases) that are available and changeable are identified.

Discussion of Results:
At this stage, land health maps have been prepared using quantitative indicators it is presented as a map. Using the health maps obtained from the previous stage and the resulting change map, the health changes of the landscape were compared statistically and visually. At this stage, the health map changes of 1984 and 2018 were prepared and classified. Areas that were unchanged from the base or had a slight change their health condition was considered excellent. Areas with low, medium, and high variation were considered good, moderate, and poor in health, respectively. In this section, changes in the landscape of the land from 1984 to 2018 were examined. From the point of view of the criteria studied in this research, in this 32-year period, the least changes and consequently the best state of health are related to 1984. Therefore, the 1984 health map was considered as the basis and other years were measured and compared accordingly. Changes in the health map of 1984, 2000 and 2018 were examined. Changes between 2000 and 2018 are negligible. Using the prepared health maps, the health changes of the land appearance between 1984 and 2018 were mapped. In the resulting change map, which was considered in the range of 0 to 255, Trial and error showed that up to 130, the changes in the study area are insignificant. Changes in the region are significant from 130 onwards. The number 130 was considered as the health threshold of the land. Therefore in addition to presenting changes in the landscape of the land, areas without change that did not cross the health threshold Changed areas that have crossed the health threshold are also shown. The development of human land uses such as urban development and roads, as well as the conversion of land uses such as forest to agriculture, rangeland and roads, runoff and erosion are introduced as factors of change. These cases may also reduce the area of forests, diseases and climate. Figure of landscape health final map of the study area presented in below:




Figure of landscape health final map of the study area

Landscape health changes in the studied time periods were evaluated and compared using measures. The results in the mentioned period show the declining trend of the health of the landscape. Examination of the results shows that the uses are more uneven and the damage to the landscape has increased. Due to the increase in fragmentation index and decrease in integration and communication, the environmental situation has declined. Due to the high capability of satellite images - such as timeliness, multi-spectrum, duplication - they can be used to determine changes in the landscape in a certain period of time. Using the landform measurements, the spatial structure of the land landscape can be quantified. By establishing a relationship between the structure and performance of the landscape and a better understanding of ecological processes, it is possible to evaluate the landscape in order to plan and manage it sustainably. As a result, the use of metrics, while saving time, provides acceptable results. The measurements can be studied and extracted as quantitative indices of the environment using satellite images. The larger the area of the spots, the less damaged and intact they are. The shorter the distance between the stains, the less tampering. Therefore, closer distance is a favorable factor in the health status of the land. Maintaining the integrity and stability of the landscape based on ecological principles leads to reducing or improving the effect of human activities on biodiversity and the dynamics of local landscapes. In the discussion of detection of changes, the measurement of land use is one of the most telling measures in the study of changes in the appearance of the land. In this study, in addition to visual analysis, cross-books were used to understand where it has changed and how much. In terms of area, 2000 is not much different from 1984. In 2018, in terms of area, about 4,000 hectares were added to the development area, including the city and roads. The area of the city has tripled compared to the base year. About 26,000 hectares of forest area has been reduced. The rate of erosion has increased by about 40 hectares (increase of run off). About 9,000 hectares have been added to the area of agricultural land. About 12,000 hectares have been added to the rangeland area. The area of roads has increased by 800 hectares. The 1984 health map is in a better position than the 2000s and 2018s in terms of the indicators studied. In the map of 2000, compared to 1984, the situation of forests, pastures and agriculture has deteriorated. In the 2018 map, the situation of forests and agriculture has deteriorated, but the situation of pastures is acceptable. The number of urban spots has increased. At the city level, there are slight positive changes due to the large and concentrated spots in the city. The maps for 2000 and 2018 do not show much difference from each other. But they are a significant change from the 1984 map. Statistical analysis of the histogram shows that the 1984 curve is more homogeneous than 2000 and 2018. The 2000 and 2018 curves are closely related. Landscape changes from 1984 to 2018 were examined. From the point of view of the criteria studied in this research, 1984 is in a better health condition than other years. Therefore, it was considered as the basis and threshold of health and other years were measured and compared accordingly. In line with the findings of this study and their ecological analysis, guidelines for diagnosing the health of the landscape were presented.
Keywords: Threshold, Health map, Landscape, Metrics, Landsat