Predicting the electron content of ionosphere can provide guidance for satellite navigation and high-frequency communication frequency selection. Deep learning-based models have been shown to be effective in predicting ionospheric changes. This paper proposes a Spatiotemporal causal convolutional network model (STCCN) based on the causal convolutional network and Rational based function (RBF) network, aiming to predict Total Electron Content (TEC) parameters in the ionosphere more accurately using multiple feature data. By using the measured ionospheric parameters of two geographic regions, the results show that the proposed model can effectively capture the correlation in ionospheric TEC series. Compared with TCN, LSTM and GRU, the prediction error of the proposed STCCN model is reduced by 22.8%, 14.4% and 18.1% respectively. The experimental results show that the proposed model can more deeply mine the spatial correlation existing in TEC sequences.