Spatial and Temporal Variability of Moisture Condition in Grassland using Spectral Data and GIS Tools, A Case study: Lower Silesian, Poland.

In recent years climate changes have been observed. Extreme weather events are becomingincreasingly. International organizations related to climate research report that 2016 was thewarmest year in the history of measurements, the second in order was the year 2018and the twenty-warmest years in the history of measurements were the last twentyyears. Drought is the most complicated climatic phenomenon among the hazards related to theweather. Drought monitoring is based on the use of large amounts of data: temperature,precipitation, evaporation, soil moisture data and many others. This monitoring requiresconducting a long-term field measurements. Many different satellite data is commonly used in environmental monitoring e.g.: LIDAR,radar, spectral data, thermal imagery etc. Increased of the spatial resolution up to severalmeters or even centimeters pixel dimension, enable to carry out analyses in the operatingmode even in large areas. Additionally, many of the satellite images are open accessdata. The aim of the presented research is to find the relations between indicatorscharacterizing agricultural drought, which were calculated using traditional and remotesensing methods. These relations will help to answer the question, if it is possible to usespectral data for monitoring agricultural drought. The work contains an analysis ofchanges in the availability of water on grasslands in the context of agricultural droughtmonitoring. Preliminary research carried out on grassland in an area with homogeneous weatherconditions, did not show any strong correlation between spectral indicators and ameteorological and hydrological indicators, in terms of moisture content in the soil-plantenvironment. Therefore, it was decided to extend the research in the spatial and temporalaspect and the number of analysed parameters. The research was conducted on grasslands in lowland areas (< 300 m a.s.l.) in LoverSilesian Voivodship, Poland, the research area was about 15,000 km2. CA 800 grasslandwith a total area about 750 km2 were analysed. Spectral data were obtained forthe period between July 2015 to November 2018 from Sentinel-2 (ESA). Severalspectral indicators based on the amount of chlorophyll, anthocyanins, carotenoids andwater content were calculated. Spectral indicator values were correlated with anaverage temperature, precipitation and evaporation for characteristic periods previousthe given image date. The analyses showed spatial and temporal variability of theagricultural drought in the grassland in lowland areas, described using spectraldata. Keyword: agricultural drought, remote sensing, satellite data, drought monitoring,Sentinel-2 [ABSTRACT FROM AUTHOR]