Key points: • A new divergence method is developed to estimate methane emissions based on satellite observations, requiring no a priori emissions. • The applicability of this method in identifying and quantifying sources is proven by a GEOS-Chem simulation with known a priori emissions. • The estimated emissions over Texas (United States) based on TROPOMI observations are evaluated and are found to be robust. Abstract We present a new divergence method to estimated methane (CH 4) emissions from satellite observed mean mixing ratio of methane (XCH 4) by deriving the regional enhancement of XCH 4 in the Planetary Boundary Layer (PBL). The applicability is proven by comparing the estimated emissions with its a priori emission inventory from a 3-month GEOS-Chem simulation. When applied to TROPOspheric Monitoring Instrument (TROPOMI) observations, sources from well-known oil/gas production areas, livestock farms and wetlands in Texas become clearly visible in the emission maps. The calculated yearly averaged total CH 4 emission over the Permian Basin is 3.06 [2.82, 3.78] Tg a-1 for 2019, which is consistent with previous studies and double that of EDGAR v4.3.2 for 2012. Sensitivity tests on PBL heights, on the derived regional background and on wind speeds suggest our divergence method is quite robust. It is also a fast and simple method to estimate the CH 4 emissions globally. Plain Language Summary Methane (CH 4) is an important greenhouse gas in the atmosphere and plays a crucial role in the global climate change. It kept increasing over the last decades. About 70% of CH 4 comes from human activities like oil/gas productions or livestock farms. The recently launched TROPOspheric Monitoring Instrument (TROPOMI) provides an opportunity to estimate the emissions of CH 4 on a regional scale. This work presents a new method to fastly derive CH 4 emissions at a fairly high spatial resolution without a priori knowledge of sources.