ISWAT - International Space Weather Action Teams 

Objects in Low Earth Orbit (LEO) are affected by atmospheric drag, which is the major source of uncertainty in the prediction of future locations of functional and de-activated satellites as well as orbital debris. Drag is heavily dependent on the density of the neutral atmosphere, which at orbital altitudes are the thermosphere and the exosphere. Neutral density is highly variable, both spatially and temporally, due to the highly variable solar radiation heating, geomagnetic heating, radiative cooling, and to a lesser extent tides and dissipating waves originating in Earth’s lower atmospheric layers.

Roadmap recommendations:

1. What are the driver-response relationships, the temporal and spatial scales of the density and wind variability due to EUV/UV radiation and geomagnetic disturbances, radiative cooling, and, to a lesser degree, forcing due to disturbances propagating up from lower altitudes? The result of this activity will be a complete qualitative and quantitative description of variations in thermosphere density due to heating of all origins.
2. Which observations/proxies/indices are most representative of upper atmosphere heating due to solar EUV/UV emissions and geomagnetic activity? Are they perennial, uninterrupted and accurately and robustly measured? What is the quality of their forecast? Is a new measurement (type) required? The result of this activity will be recommendations on the observations and indices/proxies to use in thermosphere modeling, and score cards on hours – 1 month (45 days) forecast horizons.
3. What is the accuracy and precision of upper atmosphere density and wind models? The result of this activity will be a standard assessment procedure, which requires appropriate metrics as well as high-quality and high-resolution data over many years. Model quality will be measured, quantifiable and comparable, and provided in the form of scorecards. Standard benchmarking is necessary in order to track progress objectively. It will also inform model developers which weak points must be tackled.
4. Which density and temperature datasets are available, or critically needed, to deal with items 1-3? Are the datasets coherent and appropriately edited? The result of this activity will be a database containing all datasets, which are clean (no outliers, etc) and internally unbiased (i.e., not in an absolute sense), as well as prioritized recommendations on required observations for future instruments and missions.
5. How should atmospheric drag be computed in a consistent and standard manner? The result of this activity should be reference models for satellite shape and aerodynamic coefficient computation, probably distinguishing between simple ones for (operational) orbit computation and more realistic and complex ones for density derivation. However, simple and complex models must produce results that on average, over one orbit, are identical to avoid introducing biases. Pure model performance, i.e. using observed solar and geomagnetic input, is not constant but a function of time scale (solar cycle to minutes), solar and geomagnetic activity levels, and location. When forecasting density, thermosphere model errors increase due to the error in the forecast of the drivers. Forecast horizons vary from a few days to months, to years, to complete solar cycles in mission design. It is essential that solar and geomagnetic activity forecast models and thermosphere models are improved in concert, and this is foreseen in the present I-SWAT cluster design. Scoreboards for thermosphere models, solar and geomagnetic activity forecast models will be created. Item 5, the satellite model, is the most problematic one because it is a topic that goes well beyond the reach of I-SWAT. Standard models, such as CIRA for the atmosphere, are not available and there is no consensus in the orbit computation community. Our ultimate goal is satellite drag forecasting; improvement can be quantified too by means of precise orbit determination, i.e. scorecards can be produced. However, computation of satellite shape and aerodynamic coefficients must be harmonized first, either through applying identical models, or for example through scaling.  

Data Utilization

• Maintaining and augmenting the thermosphere density database.

• Identification and preparation of critical datasets for assimilation and assessment, and optimized exploitation (e.g., machine learning).

Understanding, Theory, Modeling, Data Utilization

• Advance thermosphere modeling capability (empirical, first principles, data assimilation).

• Understand and quantify thermosphere response to drivers from above and below. 

• Analysis, modeling, and prediction of neutral density variations and changes in composition in the thermosphere at all spatial and temporal scales. 

  • altitudes below 250 km, for imminent re-entry, 

  • altitudes between 250 km and 750 km

  • altitudes 750-850 km, where most Earth observations satellites operate.

Assessment

• Assessment of thermosphere neutral density modeling capability and tracing progress over time.

Data Utilization, Forecasting

• Development and evaluation of model drivers (solar EUV and geomagnetic activity measurements, wave activity from the lower atmosphere), and their forecast.

Understanding, Theory, Modeling, Assessment

• Satellite drag modeling (aerodynamic coefficient, shape model). 

• Uncertainty assessment due to drag in orbit determination.

Roadmap Update

• Global space weather roadmap for satellite drag.