2.1.

Study areas

This manuscript tracks methane emissions in two study areas - Turkmenistan (Figure 1) and Bulgaria (Figure 2). Turkmenistan has had some of the highest methane emissions in the world in recent years, mostly from the oil and gas sector. The study also found that despite the huge methane emissions, little or no nitrogen oxide and carbon monoxide emissions are recorded over this area at scales detectable by medium-resolution satellite imagery⁷.

Figure 1.

The study area of Turkmenistan. Map provided by Google Map

Figure 2.

The study area of Bulgaria. Map provided by globalenergymonitor.org

Due to the climatic and geographical characteristics of Turkmenistan, the satellite data used in this study have a high density (high percentage of available pixels in the satellite images). Therefore, selecting this area as a test zone for applying the methodology presented here to determine background levels is reliable. According Köppen climate classification Turkmenistan has an arid, cold desert (Bwk) and steppe (Bsk) climate that is severely continental⁸, with low rainfall and relatively little cloud cover year-round. This determines the high percentage of available satellite data throughout the year.

The percentage of missing pixels from satellite images taken by the TROPOMI sensor on the board of the European Sentinel-5P satellite is significantly higher for the second study area (Bulgaria). This is determined by the character of atmospheric circulation and climatic conditions in the country. The climate in Bulgaria is more diverse than in Turkmenistan. In Bulgaria are presented seven Köppen climate types including tundra climate (ET) in the highest mountainous territories, subarctic (Dfc), warm-summer humid continental (Dfb), humid temperate climate (Cfb), humid subtropical (Cfa), warm-summer Mediterranean (Csb), and hot-summer Mediterranean (Csa)^9,10.

However, this paper demonstrates the feasibility of determining background methane concentrations using the method presented here. Two major methane sources in Bulgaria are energy-related facilities (Figure 2, red and blue squares). The first is the largest coal mining area in Bulgaria, the Maritsa East Mines, which is located in the south-central part of Bulgaria (Figure 2, red rectangle). The second one is the oil refinery northwest of the city of Burgas (Figure 2, blue rectangle).

The lack of a certain percentage of pixels in the imaged area is a major difficulty when using medium-resolution satellite imagery to monitor methane emissions. This percentage depends on various factors such as cloud cover over the study area, low albedo, high atmospheric aerosol content, etc.

2.2.

Data and data processing

Daily satellite data from the Sentinel 5P - TROPOMI instrument were used to determine background methane concentrations for the period between May 2018 and December 2022 over Turkmenistan and Bulgaria. From mid-2018, the TROPOMI instrument on board the Sentinel-5P satellite provides open source data on key air pollutants such as CH₄, NO₂, CO. Sentinel-5P is a product of the European Space Agency and is in a sun-synchronous orbit. Its data are open access and available 2 to 4 days after the acquisition, which is daily, with a bandwidth of 2600 km, local flyby time of about 13 h. The spatial resolution for each pollutant is between 3.5 and 7 km^11,12. The TROPOMI data has irregular geometry and precise georeferencing is required. This data enables more detailed spatial and temporal monitoring compared to previously used satellite information.

2.3.

Methodology

Background levels are one of the most important indicators to monitor when studying GHGs such as methane. The knowledge of background concentrations for the study area makes it possible to explore seasonal and annual trends in their variation, to quickly detect periodic or occasional average and relatively small volume emissions from unregulated sources, as well as track transport from other areas, etc.

Both study areas here are subjected to an integrated analysis on a weekly basis in order to determine the background methane level using the equation:

where XCH4_fn is the background methane value, μ is the median, is the mean of the dataset. The factor k in equation (1) is predetermined. For the purpose of our work we use k = 3 as a more precise value after a preliminary analysis to estimate the error. This is a very popular technique for automatic image analysis in background determination and astronomical object retrieval^6,13. This approach is iterative (it consists of one or more iterations), since the sigma-cropping is carried out in one or more steps. The first step is to compute the standard deviation σ (Equation 2).

The standard deviation is determined using the equation:

where is the mean of the dataset, x_i - the ith element of the dataset, n – the size of the dataset.

The next step is to remove all pixels that do not fall within the interval of ±3σ around the median value μ. For a result of more than 95% of the pixels in this interval, the equation (1) is applied, which gives the background value of methane for study areas. Any pixels that do not meet this condition are eliminated from further inspection.

The advantage of this approach is the relatively fast processing time performed in the R statistical and mathematical data processing environment. This is especially true when extracting information from large datasets, such as the image files acquired daily by the TROPOMI sensor.

3.1.

Data availability

Summary statistics of data availability about the study areas for the research period from May 2018 to December 2022 are shown in Figure 3 (Turkmenistan) and Figure 4 (Bulgaria). The percentage of available TROPOMI satellite data for the column dry molecular fraction xCH4 over Turkmenistan ranges between 52% and 96%, with a few exceptions for the months February - April. Moreover, almost each of the images has a higher pixel density (compared to Bulgaria), and the reasons for the observed pattern has been explained above in detail. The Turkmenistan data is an excellent benchmark for tracking atmospheric methane concentrations, including determining background levels - as we have a significantly higher density of information and hence different approaches can be modelled and evaluated; estimating residence time, dissipation and methane cleanup in this area when detecting large volume emission events, etc.

Figure 3.

Number of days with methane data available from the Sentinel-5p satellite, Turkmenistan

Figure 4.

Number of days with methane data available from the Sentinel-5p satellite, Bulgaria

Two conclusions can be drawn for the second study area (Bulgaria). First, there is no clear-cut specific period of the year with very high or very low percentages of available data. This is due to the faster dynamics of atmospheric processes and also other factors, explained above. Most satellite images here have significantly lower pixel densities compared to the first study area. This complicates the determination of atmospheric methane concentrations over Bulgaria, especially to record and quantify significant emissions with a sufficient degree of consistency, detail and unambiguity. Combining the available information with drone imagery, ground-based measurements and data derived from high spatial resolution (30-50 m), temporally and spatially consistent satellite measurements is the most effective solution.

3.2.

Background concentrations

The filtering procedure were applied to the data quality for the period May 2018 - December 2022 to eliminate files with low or no percentage of pixels present in the imagery, and the weekly background methane concentrations for methane were obtained as shown in Figure 5 (Turkmenistan) and Figure 6 (Bulgaria).

Figure 5.

Background methane concentrations in the atmosphere over Turkmenistan for the period May 2018 - December 2022. Data used - Sentinel-5P.

Figure 6.

Background methane concentrations in the atmosphere over Bulgaria for the period May 2018 - December 2022. Data used - Sentinel-5P.

There is a steady increasing trend in background levels of methane in the atmosphere over the two study areas on an annual basis. Background concentrations of methane in the atmosphere over Turkmenistan (Figure 5) are higher compared to those for Bulgaria (Figure 6). Moreover, the amplitudes in monthly variations in Turkmenistan are higher compared to those in Bulgaria. This is due to the presence of several super-emitters on the territory of Turkmenistan, which periodically release huge amounts of methane into the atmosphere and thus sharply and permanently increase its concentrations. Hundreds of emission events from these sources have been recorded by Trenchev P. (2022)⁷ for the study period.

Figure 6 illustrates the lack of clear seasonal trends in atmospheric methane concentrations over Bulgaria during the study period. Transboundary transport and, to a lesser extent, emissions from local sources are mainly responsible for the trend towards increase in background methane concentrations in the atmosphere over Bulgaria.