Segmentation of Agricultural Parcels in Satellite Images Based on Historical Vegetation Index Data

This paper considered the issue of agricultural fields boundary recognition in satellite images. A novel algorithm based on the aggregated history of vegetation index data obtained via open satellite data, Sentinel-2, was proposed. The proposed algorithm included several basic steps, namely the detection of parcel regions on aggregated index data; the calculation of aggregated edge maps; the segmentation of parcel regions using the edges obtained; the computation of connected components and their contour extraction. In this paper, we showed that the use of aggregated vegetation index data and boundary maps allow for much more accurate agricultural field segmentation compared to the instant vegetation index approach. The quality of segmentation within regions of Russia and the Ukraine was estimated. The dataset that was used and Python implementation of the proposed algorithm were provided.

THE LATE HOLOCENE HISTORY OF VEGETATION OF THE ACCUMULATIVE PLAIN IN THE INNER PART OF THE USSURI BAY

На основе палинологических данных, восстановлена история развития природной среды аккумулятивной равнины в кутовой части Уссурийского залива за последние 5350 кал. л. Определены особенности структуры растительного покрова в завершающую фазу оптимума голоцена, а также при разнонаправленных климатических флуктуациях позднего голоцена. В завершающую фазу оптимума голоцена широколиственные растения в горном обрамлении были распространены более широко, чем в настоящее время. На аккумулятивной равнине преобладали мелколиственные растительные сообщества с доминированием березы обыкновенной, гибридных берез и ольхи. Похолодание, начавшееся в конце среднего и продолжившееся в начале позднего голоцена, привело к сокращению в растительности широколиственных растений и более широкому распространению мелколиственных. В конце позднего голоцена в кутовой части Уссурийского залива сложились природные условия, обусловившие формирование современной структуры растительности. Based on the palynological data, the history of vegetation on the accumulative plain in the inner part of the Ussuri Bay was reconstructed. Features of the structure of the vegetation cover during the final phase of the Holocene optimum and at multidirectional climatic fluctuations in the Late Holocene were identified. Broad-leaf plants in the final phase of the Holocene were more widely-spread on the surrounding mountain than at present. On the accumulative plains grassy meadows with small-leaf plant communities dominated. Deterioration of natural conditions during the transition to the Late Holocene caused a decrease in vegetation of broad- leaved plants and in an increase in the role of small-plants. At the end of the Late Holocene natural conditions developed in the inner part of the Ussuri Bay, that conditioned the formation of the modern structure of vegetation.

Vegetation history of Western Russia (Upper Dnieper, Smolensk region): Climate and human impact on landscape in last two millennia

<p>Climate and vegetation history of Upper Dnieper region (Western Russia) is investigated poorly while archaeological studies provide evidences of human activities during last 3 millennia. Our study presents vegetation reconstruction based on pollen analysis of sediments extracted from two sites in Smolensk region. The first site is located in Katynka river bassin and pollen analysis of extracted buried soil, alluvium and peat sediments demonstrates vegetation dynamics in archaeologically rich area over 5 millennia. The second site is located in 50 km from to the west from Smolensk and in 15 km to east from the Russian-Belarus state border. The analysis of extracted peat sediments presents regional history of vegetation. The aim of our study is to compare data obtained from both sites and to estimate climate and human influence on vegetation during last two millennia when activities associated with agriculture changed Dnieper valley landscape significantly.</p><p>The first results of pollen analysis data of the first site allow to register significant human impact on vegetation 2.0-0.8 ka BP. Before that period pollen of indigenous forest trees dominates in spectra while since 2.0 ka BP pollen compassion changes dramatically and pollen of <em>Betula</em> and <em>Pinus</em> is in majority in so-called “Gnezdovo soil” lay. Medieval lays of sapropel contains mostly pollen of <em>Pinus</em> with admixture of <em>Betula</em> and <em>Alnus</em>. Taxonomic diversity and presence of meadow herbs, weeds and cultivated taxa pollen increases significantly (up to 30%). Dynamics of pollen composition in specimens from the second site allows us to register slow processes of indigenous vegetation recovery over last 3 centuries approximately. Modern analogue technique applied on pollen data and analysis of historical data makes possible to separate impacts of climate and human on vegetation of the past and to reconstruct the climate of last two millennia.</p><p>The study was funded by RFBR, project number 19-34-90172.</p><p><br><br></p>