Abstract Live fish waterless transportation is a low cost and promising strategy. But how to improve survival rate and reduce stress are challenging goal. This paper aims to explore the effect of ambient and stress indicators on fish state in waterless transportation. Multi-sensors ambient indicators and breath rate stress indicator monitoring system were designed and implemented in live salmon waterless transportation. Dynamic probing for fish stress changes in term of acceleration sensor-based that is a significant survey to get optimum temperature and improve fish survival rate. The critical indicators dynamic prediction and live salmon state diagnosis model were developed by using CNN-SVM algorithm. And stress level evaluation model is established based on ambient indicators and stress indicator by using fuzzy comprehensive evaluation method. After 10h waterless transportation, the quality indicators i.e. texture, color and pH changes were measured and evaluated. The results show that multi-sensors real-time monitoring are significant to improve survival rate and fish flesh. Quality properties of live salmon samples decreased more slowly than dead salmon. This will provide an effective and reliable survival evaluation and transportation management.