Satellite thermal infrared remote sensing has received worldwide attention in the exploration for earthquake precursors; however, this method faces great controversy. Obtaining repeatable phenomena related to earthquakes is helpful to reduce this controversy. In this paper, a total of 15 or 17 years of Moderate-resolution Imaging Spectroradiometer (MODIS)/Aqua and MODIS/Terra satellite remote sensing land surface temperature (LST) products is selected to analyze the temperature changes before and after the Mw 7.9 earthquake in Nepal on 25 April 2015 and to explore possible thermal information associated with this earthquake. Major findings are given as follows: (1) from the time course, the temperature slowly cooled before the earthquake, reached a minimum at the time of the earthquake, and returned to normal after the earthquake. Since these changes were initiated before the earthquake, they may even have been precursors to the Nepal earthquake. (2) From the space distribution, the cooling areas correspond to the seismogenic structure during the earthquake. These cooling areas are distributed along the Himalayas and are approximately 1300 km long. The widths of the East and West sides are slightly different, with an average temperature decrease of 5.6 °C. For these cooling areas, the Western section is approximately 90 km wide and 500 km long; the East side is approximately 190 km wide and 800 km long. The Western side of the cooling strips appeared before the earthquake. In short, these kinds of spatial and temporal changes are tectonically related to the earthquake and may have been caused by the tectonic activity associated with the Nepal earthquake. This process began before the earthquake and therefore might even be potentially premonitory information associated with the Nepal earthquake.

The origins of tremors and their relation to fault slip and impending earthquakes remain unclear. Laboratory studies on tremor‐like acoustic emission events generated during preslip may shed some light on the above issues. We conducted stick‐slip experiments and observed tremor‐like events and preslip in the laboratory. The results show the following. (1) The dominant frequency of tremor‐like events increases with accelerating preslip. (2) The evolution of tremor‐like events is spatiotemporally consistent with that of preslip. (3) Interactions between preslip and high‐frequency events indicate the formation of a positive feedback effect between the expansion of preslip and failure of asperities, which leads to stick‐slip instability. Thus, we propose that tremors may be caused by stress variations derived from asperity interactions along fault surfaces at dry conditions. The increasing dominant frequency of tremors may imply the accelerating preslip of an upcoming earthquake.