Accurate, sensitive, and selective detection of explosives is of vital importance in antiterrorism and homeland security. Fluorescence sensors are prevalent for sensitive and fast in-field explosive detection but are sometimes compromised by accuracy and stability due to the similar structures of explosives, photobleaching, and complex sample matrixes. Herein, we developed a first bimodal methodology capable of both sensitive in-field fluorescence detection and accurate laboratory mass spectrometric quantification of 2,4,6-trinitrotoluene (TNT) by utilizing the characteristic fluorescent and mass spectrometric response of copper nanoparticles (CuNPs). An excellent selectivity was also realized by involving aptamer recognition. The methodology is capable of detecting TNT at subpart per trillion (PPT) levels, with a detection limit of 0.32 pg mL-1 by inductively coupled plasma mass spectrometry (ICPMS) and 0.17 ng mL-1 by fluorimetry. The signal response was accurate and stable for at least 60 days by ICPMS. Thanks to the biospecificity of the aptamer, this bimodal methodology is potentially applicable to a large panel of explosives.