Page 98 - Read Online

Page 14 of 23 Yu et al. Carbon Footprints 2025, 4, 17 https://dx.doi.org/10.20517/cf.2025.12

Table 4. Representative studies using time-series models for vehicle-level emission estimation

Input variables
Reference Model Data source Dynamic driving
behavior Vehicle attribute Environment
[72] Wavelet Remote CO Speed, acceleration / The wind speed and direction, the outdoor temperature, the relative humidity,
2
transform, sensor, LPR data the atmospheric pressure, hourly number of vehicles passing by the monitoring
LSTM lanes, the average vehicle length
[73] LSTM, RNN, Dynamometer test Speed Engine family, engine manufacturer, engine model /
GRU data year, vehicle inertia, odometer reading, number of
cylinders, fuel type
[74] LSTM GPS, PEMS, Speed, acceleration / Grade and number of on-board passenger variables
passenger data
[75] LSTM PEMS, GPS Speed, acceleration, VSP / Road slope
[76] RNN, LSTM CO2 sensor, OBD Speed, acceleration, / Mileage
engine RPM, fuel flow,
throttle

LPR: License plate recognition; PEMS: portable emissions measurement system; GPS: global positioning system; OBD: on-board diagnostics; LSTM: long short-term memory; RNN: recurrent neural networks; GRU:
gated recurrent unit.

Table 5 summarizes several representative spatiotemporal emission models at the traffic level, highlighting their diverse data sources and input variables used
to capture the complex spatiotemporal dynamics of traffic emissions.

Discussion
Data requirements
The data requirements for data-driven emission models are broadly comparable to those of classical emission models. However, unlike classical models that

typically rely on predefined emission factor databases, data-driven approaches, particularly those employing machine learning and deep learning, require
independent model training. This entails substantially higher demands in terms of both data volume and quality . Spatiotemporal models represent an
[82]
emerging paradigm in traffic emission modeling, yet they have not been widely adopted in traffic management practice. One major barrier is their
exceptionally high data requirements. For instance, models based on GNNs or GNNs require detailed Geographic Information System (GIS) data of the road

network, high-resolution dynamic traffic flow information, and meteorological parameters for pollutant dispersion. These datasets are typically maintained by
different government agencies or private entities, and they often vary significantly in format, coordinate systems, and temporal resolution. Beyond challenges
in data acquisition, low data quality and the complexity of integrating heterogeneous data sources present substantial obstacles to the effective deployment of

spatiotemporal emission models in real-world road networks.