Methodology

The climatic changes of the last decades entailed significant and irreversible changes in the spatial structure of the nival-glacial landscapes of the Mongolian Altai which are expressed in the formation of the latest morphosculpture. Geoinformation and analytical system (GIAS) «EvCLiD» (Evolution and Climatogenic Landscape Dynamics) developed, tested and implemented in use for quantitative assessment of landscape transformation. The system is created in the open package software environment Microdem/TerraBase V.16.0, Petmar Trilobite Breeding Ranch® — GIS freeware simple for using, but efficient tool for storing, visualizing, and analyzing spatial data. The databank is the information basis of GIAS «EvCLiD», organized in the form of catalogs, including sheets of topographic base scale 1:25000, thematic databases in DBASE format, and materials of polychronous remote sensing. The spatial data of GIAS «EvCLiD» are given in a uniform datum (WGS 84) and transformed into a UTM projection, vector maps are presented in the format of Shape-files. Aster Global DEM (V.2) and NASA SRTM matrix with a resolution of 1 arcsec, was used as a digital elevation model. The databank of GIAS «EvCLiD» was formed from open network portals and file archives of the USGS Geological survey (Global Data Explorer http://gdex.cr.usgs.gov/gdex/, Earth Explorer https://earthexplorer.usgs.gov/), NASA  (EOSDIS https://reverb.echo.nasa.gov/reverb), Roscosmos (Geoportal http://gptl.ru/).

Table 1. The GIAS EvCliD catalogue of collected remote sensing data

Sensor Scene ID Date
Landsat 8 LC81420262016226LGN00

LC81410272015216LGN00

LC81410262015200LGN00

2016-08-13

2015-08-04

2015-07-19

Landsat 7 ETM+ LE71420262015231EDC00

LE71410262015208EDC00

LE71410262004210PFS01

LT51420262004193BJC00

LE71410272003207ASN02

LE71410272002220SGS00

LE71410262002220SGS00

LE71420262002195SGS00

2015-08-19

2015-07-27

2004-07-28

2004-07-11

2003-07-26

2002-08-08

2002-08-08

2002-07-14

Landsat 5 LT51420262011228IKR00

LT51410262010218IKR00

LT51410262001209BJC00

LT51410262000207BJC00

LT51410261998217BJC00

LT51420261998208BJC00

LT51410261996196BJC00

LT51420261995216BJC00

LT51420261991221BJC00

LM51410261991198AAA03

LM51410271991198AAA03

LT51410261991198XXX03

2011-08-16

2010-08-06

2001-07-28

2000-07-25

1998-08-05

1998-07-27

1996-07-14

1995-08-04

1991-08-09

1991-07-17

1991-07-17

1991-07-17

ERTS (Landsat 2) LM21530261977261AAA03 LM21530261977225TGS03 1977-09-18

1977-08-13

KH-4B DS1104-1055DF007

DS1104-1055DA013

1968-08-11

1968-08-11

WorldView — 110 1030010048939200 2015-08-19

Figure 6. GIAS EvCliD interface with thematic projects “Tsambagarav” and “Sutai”

The catalogue of remote sensing data includes high-resolution multispectral digital images from Landsat 8, Landsat 7 ETM+, Landsat 5, ERTS, World View -2, monochrome images KH-4B (Table 1). Geocorrection and abstracting imagery from the satellite KH-4B was performed according to the characteristic points, which was taken to be the intersection of landscape contours, headlands of rock, the mouth of the tributaries, the characteristic curves of channels and other objects displayed on aerial photographs. The number of hard reference points was at least twenty in all cases.

Figure 7. The Map indicating key sites of the field investigations in the Mongolian Altai.

The cryogenic and periglacial  landscapes of the key sites were mapped using visual interpretation of remotely sensed data following a pre-defined set of criteria (Heyman et al., 2008). A combination of Landsat 7 ETM+ imagery (30 m resolution) and the ASTER Global DEM  was used as the primary data source. All mapping was performed and compiled using on-screen digitizing of landforms. The mapping was performed at on-screen scales ranging from 1:30,000 to 1:60,000. Multiple RGB band combinations were used in the mapping, including both ‘true’ (RGB: 5,4,2) and ‘false’ Infra Red (IR) (RGB: 4,3,2) color composites of Landsat 7 ETM+ imagery. Furthermore, standard image enhancement procedures for satellite imagery, such as contrast stretching and histogram equalization, were adopted to improve the landform spectral signature strength. Finally, we also performed pan sharpening of the false IR colour composites using the panchromatic band (15 m resolution) to significantly enhance the sharpness of the imagery. A semi-transparent layer of satellite imagery was draped over the ASTER Global DEM data in a Microdem/Terra Base V. 16.0 environment to aid landform interpretation in complex topography. The mapping was checked in several key areas during 2017 field investigations in the Mongolian Altai (Figure 7). Vectorization of thermokarst lakes was made on direct signs of interpretation on materials of space survey (World View — 2, KH-4B). The main interpretation features of surface waters were: smooth photo tone and specific monotone or expressive structure of water image and shape of the lakes attachment to depressed relief elements. The lake are interpreted, when it became apparent their shape. Even very small lakes could be identified among a large cluster of lakes, in the images they are depicted in the form of small points. Vectorization of the landscape elements were carried out in manual mode. The SRTM NASA Matrix and ASTER GDEM V.2 were used in the calculation of the area of three-dimensional surface of glaciers, as a digital elevation model. Verification of the reference points of the digitized polygons, conducted in the field, revealed the measurement error not exceeding 8-10%. The boundaries of the glaciers of Little Ice Age were reconstructed on the well-expressed in the relief of the marginal moraine complexes.

High mountain areas are experiencing some of the earliest and greatest impacts of climate change. As well as, the mountains have an important role in regulating the climate in Asia. In Fifth Assessment Report presents the main results on climate change in Asia: warming trends, including higher extremes, are strongest over the continental interiors of Asia, and warming in the period 1979 onward was strongest over China in winter, and northern and eastern Asia in spring and autumn; from 1900 to 2005, precipitation increased significantly in northern and central Asia but declined in parts of southern Asia; future climate change is likely to affect water resource scarcity with enhanced climate variability and more rapid melting of glaciers [IPCC, 2014; Hijioka et al., 2014].

Observed changes in annual average temperature and precipitation in Asia are shown in Figure 8, but for a large region of Asia there is no data («white spot»). Moreover, knowledge on how climate change impacts alpine landscapes still very much different.

Figure 8. Observed changes in annual average temperature and precipitation in Asia: map of observed annual average temperature changes from 1901–2012, derived from a linear trend (left panel) and map of observed annual precipitation change from 1951–2010, derived from a linear trend (right panel) [IPCC, 2014; Hijioka et al., 2014].

In the Asia, Altai Mountains is included in a «white spot» and this region important both climatologically and ecologically mountain range. In these mountains, the Atlantic air masses from the west have interacted with the Arctic air masses from the north. It is not only a climatic conjunction, but important ecological transition where the taiga forests have interacted with the steppes.

In Altai Mountains, about ten weather stations have continuous data (temperature and precipitation) for more than 60 years (Figure 9). To have a reliable starting point, we used the data from the official site of the All Russian Research Institute of Hydrometeorological Information – RIHMI-WDC [http://meteo.ru/english/data/] for the stations located in the Russian part of the Altai (Soloneshnoe, Kyzyl-Ozek, Yailu, Ust-Koksa, Kara-Tyurek, Kosh-Agach). These meteorological data sets were automatically checked for quality control before being stored at the RIHMI-WDC and they were checked for the homogeneity. The RIHMI is the major source of official information of the Russian weather stations. The temperature and precipitation series of Mongolian weather stations – Hovd and Ulgii – were provided by Mongolian Hydrometeorological Institute. To identify the missing data and to provide the homogeneous data for Mongolian stations, we used the Methods of State Hydrological Institute and Voeikov Main Geophysical Observatory [Bulygina et al. 2013]. Additionally, we investigate the climate changes from data NCEP/NCAR Reanalysis, to understand spatial climate changes in the Altai Mountains and the bordering territories. The NCEP/NCAR Reanalysis project is using a state-of-the-art analysis/forecast system to perform data assimilation using past data from 1948 to the present. A large subset of this data is available from PSD in its original 4 times daily format and as daily averages. [https://www.esrl.noaa.gov/psd/].

Figure 9. Geographic location of Altai Mountains and weather stations [https://www.google.com/maps/]

In this study, we use the classical Mann-Kendall test to check whether there is a trend in temperature and precipitation time series [Mann 1945; Kendall 1975]. In addition, the nonparametric Mann–Kendall–Sneyers test [Mann 1945; Kendall 1975; Sneyers 1975] was applied to determine the occurrence of step change points of temperature and precipitation.

Let x1,…,xn be the data points. One then defines ni which is the number of elements xj preceding xi (j<i) and such that xj < xi. Under the null hypothesis, the test statistic is normally distributed with the mean and the variance given by:

In this work, long-term trends of temperature and precipitation were determined as the slope of linear regression line of the mean temperature and precipitation during the studied time interval by using the method described by Li et al. (2013).

Special attention in this project was given to analysis of climate change in Western Mongolia. Then we divided the studied interval into two time subintervals at the step change point and compared them between each other. Studies of changes in the hydrothermal regime on the territory of the Western Mongolia, carried out on the basis of processing and analysis of indicators of international weather data base NOAA»s National Centers for Environmental Information (NCEI) [access Mode: ftp://ftp.ncdc.noaa.gov/pub/data/gsod/] during the period from 1958 to 2017 for 14 weather stations that have different time series of observations. The weather stations Altai, Uliastai, Hovd, Ulaangom, and Ulgi have the greatest periods of observations of the temperature regime 60 years (1958-2017 years). Weather stations Omno-Gobi, Tolbo, Baitag have observations for 30 years (1984-2001). The remaining six stations have a period of observation only for the last 18 years (2000-2017). Data on precipitation in the database are also the most complete for only the last 18 years and for some individual stations only for 10 years. The mathematical processing of the NCEL meteorological data was performed by converting the information into an Excel application, where the indicators from the traditional U.S. Customary System were transferred to the international system of units (SI). The reliability of the data is confirmed by the results of statistical analysis and comparison with some similar studies [Mongolia Second Assessment Report on Climate Change (MARCC-2014). – Ulanbataar, Ministry of Environment and Green Development of Mongolia, 2014. – 302 pp.].