| Popis: |
An understanding of the complex environmental systems at a regional scale is still a challenging problem in the Central Asian countries, which cover 399.4 million hectares (ha). This chapter focuses on one specific (specific regions are classified as Oblast and it is targeted as a province) region for each country of Central Asia. The time series data through remote sensing represents a promising resource for studying connectivity within dynamic ecosystems. Among diverse landscapes in Central Asia, the selected zones are classified on base elevation and major land use type as mountain zones—Jalalabad region (Kyrgyzstan), Gorno-Badakhshan Autonomous Region (Tajikistan), and flat zones—Lebap region (Turkmenistan), Navoi region (Uzbekistan), and Kyzylorda region (Kazakhstan). The seasonal variations of NDVI derived from AVHRR-GIMMS 3g data for the period 1982–2015 estimated differences of spectral profiles enacted to indicate the weakness and strength of vegetation patterns. In response to the medium spatial resolution of the freely available multispectral Sentinel-2 datasets (0.3 m) these were upgraded for specific areas for monitoring the surroundings. The outputs of breakpoint of NDVI in Navoi (Uzbekistan) were described as slightly changed after 2002, whereas in general decreasing NDVI trends of vegetation values in Uzbekistan and Turkmenistan were observed in the past 3–5 years. With application of Mann–Kendall (MK) monotonic trends analysis we are approaching estimation of statistically significant trends (p value, MK-tau) in each selected zone. The negative significance procedure of MK-tau outputs is described in the regions of Turkmenistan, Uzbekistan, and Kyrgyzstan. Others are described as a fast recovering or caused by the increase of levels of precipitation as observed after 2013–2015. This study introduces approaches to enable and combine high- and low-resolution datasets for monitoring large–scale rangeland habitats and estimates of the amount of the data needed to better interpret biodiversity loss levels. |
