Spatio-temporal analysis of dzud hazard in Mongolia

General Information

This section aims to describe the climate datasets available within the Spatio-temporal Analysis of Dzud Hazard workspace for Mongolia available in mCube and developed for the Indonesia Resident Mission of the Asian Development Bank.

The aim of this dataset is to provide multiple climate indices with potential significance to dzud precursor conditions in dzud-affected areas in Mongolia. These products support the process understanding of dzud genesis for iron dzud, allow identifying dzud prone areas and may serve as a basis for holistic approaches building upon a robust data base on dzud drivers.

This climate dataset is developed by the national meteorological and geophysical service of Austria, the Central Institution for Meteorology and Geodynamics (ZAMG), in the framework of the ESA’s GDA AID Disaster Resilience project for the Use Case 7 Spatio-temporal analysis of dzud hazard in Mongolia.

Context

Dzud events are severe meteorological events that have affected Mongolia throughout history. A Dzud is a combination of deep cold, strong winds, heavy snow, and ice which leads to large-scale deaths of livestock due to starvation, causing huge economic damage for pastoral nomadic communities in Mongolia¹. The harshest dzud event occurred in the 2009-2010 winter season killing 10.3 million livestock².

Figure 1: Mongolia: Pastoral communities continue to struggle with the effects of Dzud. Photo credit: EU Civil Protection and Humanitarian Aid. ³.

The GDA AID DR Use Case 7⁴ focuses on: “the spatial and temporal characteristics of the main meteorological hazard components of dzud, including locations and time points of severe meteorological conditions, with the aim to identify patterns of dzud genesis”.

To foster a better understanding of dzud events, gaining knowledge about the main drivers is of core importance. Consequently, comprehensive data about dzud precursor conditions are of particular interest. Spatially resolved time series of meteorological drivers (e.g., snow, temperature, ice, water availability, etc.) can serve as a basis for establishing thresholds to identify meteorological events with relevance to different dzud types.

Spatio- temporal analysis of meteorological dzud hazards advances our understanding of dzud genesis. Enhanced knowledge regarding important meteorological hazards that are dzud drivers is a prerequisite for future studies aiming for impact/risk assessment.

Dataset description

Multiple types of dzud⁵ exist, which are characterised by different precursor conditions and different drivers leading up to dzud occurrence. This climate dataset focuses only on one of them, the iron dzud (Tumur), which consists of the loss of access by cattle to grass due to ice cover. This type of dzud predominantly occurs after snowmelt and re-freezing due to rapid changes in temperature⁶.

Climate indices related to iron dzud

Thus, this dataset consists of multiple iron-dzud related climate indices over an extended period of time. It includes 15 climate indices derived from 4 different databases: ERA5-Land 2m temperature (ERA5-Land), Snow (SNOWGRID CL), Frozen surface state (FSS), and Land Surface Temperature (LST).

An example is the Snow Cover Duration (SCD) index, which is defined as the seasonal count of days with a snow height > 10cm. The below Figure 1 shows a result from the SCD index for 2021 over the Khuvsgul aimag location in Mongolia.

Figure 2: Bottom left: Snow cover duration (SCD) index based on SNOWGRID-CL driven with ERA5-Land for winter season 2021 (2020-12-01 to 2021-05-31). The black outline shows the boundary of Khuvsgul aimag. Top maps show the location of Khuvsgul aimag within Mongolia.

When a different data basis is employed to extract the same index, they are offered as distinct products. In total the dataset includes 22 index products.

Products derived from ERA5-Land, SNOWGRID CL, FSS and LST are listed in Table 1, 2, 3, and 4 respectively.

Table 1 - List of climate indices derived with ERA5-Land 2m temperature (ERA5-Land) as data basis.

Table 2 - List of climate indices derived with Snow (SNOWGRID CL) as data basis.

Table 3 - List of climate indices derived with Frozen surface state (FSS) as data basis.

Table 4 - List of climate indices derived with Land Surface Temperature (LST) as data basis.

Temporal coverage

Temporal coverage for climate indices is of multiple years with annual frequency (seasonal). For each index this depends on the data coverage of source data basis (see Table 5).

Table 5 - Temporal coverage for climate indices as function of source databases.

Indices are computed for an annual winter season spanning from December of the preceding year to May of the respective year.

As an example for ERA5-Land derived index products, a total of 73 seasonal estimations of each of the 10 indexes are offered in this dataset. Each of them is accessible in mCube as a single Dataset.

Spatial cover and resolution

Indexes are offered only over two portions of Mongolia covering the Khovsgol and Zavkhan provinces.

Climate index products are offered as raster at different spatial resolutions from 0.01° (~ 1 km) to 0.25° (~ 25 km). Also, spatial resolution is function of the source data basis (see Table 6).

Table 6 - Spatial resolution for climate indices as function of source databases.

Filename convention

Each index product is accessible in mCube as a single file and is following a filename convention. As an example for the SCD index derived from the SNOWGRID CL database the filename convention is the following:

GDA-AID-DR_UC7-ADBMON_Product_{data_basis}-{index}-{version}_IronDzud-{aoi}-{year}.tif

where:

• {data_basis} indicate the data basis employed to derive the cliamte index (e.g. SNOWGRID-CL),

• {index} acronym for the climate index (e.g. SCD),

• {version} product version (e.g. V01),

• {aoi} name of the area of interest (e.g. Khuvsgul),

• {year} stands for the year of the respective winter season (e.g. 1950).

Example:

GDA-AID-DR_UC7-ADBMON_Product_SNOWGRID-CL-SCD-V01_IronDzud-Khuvsgul-1950.tif

Product specification tables

All the 22 climate index products listed in Tables 1, 2, 3 and 4 are catalogued in mCube as single-band assets in a dedicated STAC item.

Product specifications for each product can be found in below tables:

Product Specifications of climate indices from ERA5-Land

Product Specifications of climate indices from SNOWGRID CL

Product Specifications of climate indices from Frozen Surface State (FSS)

Product Specifications of climate indices from Land Surface Temperature (LST)

Credits