# Simulation Model

## Introduction

Simulation model parameters describe properties of all relevant elements of the observatory. This includes site,
telescopes, and calibration devices.
The management, definition, derivation, verification, and validation of the simulation model is central to simtools.
For simplicity, the term *model parameter* refers to values, (multi-dimensional) vectors of values
(e.g., the mirror reflectivity vs wavelength and photon incident angle), functions, and algorithms
(e.g., the telescope trigger algorithm).

Model parameters describe for example:

- Atmospheric model parameters (profile, transmission)
- Geomagnetic field characteristics
- Optical and mechanical telescope properties
- Detector plane settings
- Trigger configurations
- Camera readout parameters
- Simulation production related software parameters

The major components of the simulation model are:

1. **Model parameter schema files:** The schema files define the model parameters and their properties. Every model parameter is defined in a schema file, see the
   [schema file directory](https://github.com/gammasim/simtools/tree/main/simtools/schemas/model_parameters).
2. **Model parameter repository:** The simulation model is stored in a [gitlab repository](https://gitlab.cta-observatory.org/cta-science/simulations/simulation-model/simulation-models). Review and revision control of the simulation model is done using this repository.
3. **Model parameter databases:** The simulation model is stored in [mongoDB databases](databases.md#databases).
The {ref}`db_handler module <DBHANDLER>` provides reading and writing interfaces to the database.
The model parameter database holds all values from the simulation model repository and is updated for each release of the simulation model repository.

:::{Note}
The simulation model is a central part of the CTAO Simulation Pipeline. The responsibility for the values and the correctness of the simulation model parameters is with the CTAO Simulation Team.
:::

## Description of model parameters through schemas

Model parameter data structures are defined in schema files.
To ensure consistency and correctness of the model parameters, these schema files are used to validate the model parameter files with `simtools-validate-file-using-schema`.
Strict versioning of schemas and metaschema is introduced to ensure that the model parameters are always compatible with the schema files.

The model parameters are stored in json-style in the [model repository](https://gitlab.cta-observatory.org/cta-science/simulations/simulation-model/simulation-models) and database.
A typical model parameter file looks like:

```json
{
    "schema_version": "0.1.0",
    "model_parameter_schema_version": "0.1.0",
    "parameter": "num_gains",
    "instrument": "LSTN-01",
    "site": "North",
    "parameter_version": "1.0.0",
    "unique_id": null,
    "value": 2,
    "unit": null,
    "type": "int64",
    "file": false
}
```

Two types of schema definitions need to be distinguished:

1. **Meta schema:** The meta schema defines the structure of the model parameter file. It is a multi-document YAML file including several schema versions. The meta schema is used to validate the model parameter files. The `schema_version` field is used to identify the correct meta schema.
2. **Model parameter schema:** The model parameter schema defines the properties of the model parameter. It is a human-readable YAML file that describes the model parameter in detail. The `model_parameter_schema_version` field is used to identify the correct model parameter schema (optional field; default value is `0.1.0`).

The metaschema [model_parameter.metaschema.yml](https://github.com/gammasim/simtools/blob/main/src/simtools/schemas/model_parameter.metaschema.yml) defines the json data structure used for model parameters. This meta schema defines e.g., that `parameter` must be a string or `file` a boolean. No additional properties are allowed.

Model parameter schemas are defined in YAML format and located in the `simtools/schemas/model_parameters` directory. Each schema file specifies fields such as `name`, `type`, `format`, `units`, and applicable telescopes.  Schemas also include metadata on default values, validation rules, data sources, and links to relevant simulation software.  Model parameter descriptions are primarily derived from the [sim_telarray manual](https://www.mpi-hd.mpg.de/hfm/~bernlohr/sim_telarray/), ensuring alignment with its definitions and conventions.

For above example the schema file [num_gains.schema.yml](https://github.com/gammasim/simtools/blob/main/src/simtools/schemas/model_parameters/num_gains.schema.yml) is:

```yaml
%YAML 1.2
---
title: Schema for num_gains model parameter
version: 0.1.0
meta_schema: simpipe-schema
meta_schema_url: https://raw.githubusercontent.com/gammasim/simtools/main/simtools/schemas/model_parameter_and_data_schema.metaschema.yml
meta_schema_version: 0.1.0
name: num_gains
description: Number of different gains the input signal gets digitized.
data:
  - type: int
    unit: dimensionless
    default: 1
    allowed_range:
      min: 1
      max: 2
instrument:
  class: Camera
  type:
    - LSTN
    - LSTS
    - MSTN
    - MSTS
    - SSTS
    - SCTS
activity:
  setting:
    - SetParameterFromExternal
  validation:
    - ValidateParameterByExpert
source:
  - Initial instrument setup
simulation_software:
  - name: sim_telarray
```

Model parameter schema files follow the fixed meta schema (see `meta_schema`, `meta_schema_url`, and `meta_schema_version` fields in above example and [model_parameter_and_data_schema.metaschema.yml](https://github.com/gammasim/simtools/blob/main/simtools/schemas/model_parameter_and_data_schema.metaschema.yml)).

### Valid Keys

Valid keys are described in detail in [model_parameter_and_data_schema.metaschema.yml](https://github.com/gammasim/simtools/blob/main/simtools/schemas/model_parameter_and_data_schema.metaschema.yml). The most important parameters are:

#### Header section

- `title`: Title of the schema file.
- `version`: Version of this schema file.
- `meta_schema`: Name of the base schema.
- `meta_schema_url`: URL of the base schema.
- `meta_schema_version`: Version of the base schema.

#### Parameter description

- `name`: Name of the parameter.
- `description`: Description of the parameter.
- `short_description`: Short description of the parameter (optional).

#### Parameter data

The `data` field is used to describe the actual type, format, and allowed values of the parameter.

- `type`: Data type of the parameter.
- `unit`: Units of the parameter (compatible with astropy units).
- `default`: Default value of the parameter.
- `allowed_range`: Allowed range of the parameter.
- `allowed_values`: Allowed values of the parameter. Use if values are not continuous and restricted to a small list of allowed values.

Tables of any dimensions can be described using the `data_table` field. Example:

```yaml
data:
  - type: data_table
    table_columns:
      - name: wavelength
        description: Wavelength.
        required: true
        unit: nm
        type: double
        required_range:
          min: 300.0
          max: 700.0
        input_processing:
          - remove_duplicates
          - sort
      - name: photo_detection_efficiency
        description: Average quantum or photon detection efficiency.
        required: true
        unit: dimensionless
        type: double
        allowed_range:
          min: 0.0
          max: 1.0
```

Input processing in form of sorting, removing of duplicates, etc. can be specified using the `input_processing` field.

:::{Warning}
The `data_table` field is not yet implemented in the simtools schema and parameter files.
:::

#### Instrument description

- `class`: Instrument class. Allow values are *Camera*, *Site*, *Structure*, *Telescope*
- `type`: Instrument type following CTAO Naming.
- `site`: CTAO site. Allowed values are *North* and *South*.

### Activity description

Describes setting and validation activities. Each activity corresponds to a workflow as described in the [simtools workflows repository](https://github.com/gammasim/workflows).

:::{Warning}
The implementation of activities and workflows is incomplete and in progress.
:::

### Data source description

Describes the source of the data or parameter (e.g., *Calibration*)

### Simulation software description

Describes the simulation software (e.g., *sim_telarray* or *corsika*) the parameter is used for.

```yaml
simulation_software:
  - name: corsika
```

Allowed is here to add the internal simulation software parameter naming (if different than the name used here in the schema).
This is important to write correct and consistent CORSIKA or sim\_telarray configuration files.

 Example:

```yaml
simulation_software:
  - name: sim_telarray
    internal_parameter_name: secondary_ref_radius
```

## Updating the model database

:::{Danger}
This is for experts only and might impact the simulation model database for all users.
Tests should be done using a sandbox database before applying this to the production databases.
:::

### Update a single model parameter

New model parameter defined in the simtools database format (json file) can be uploaded to the database using the {ref}`simtools-add-value-from-json-to-db <db_add_value_from_json_to_db>` application.
New data files can be uploaded using {ref}`simtools-add-file-to-db <db_add_file_to_db>`.

### Upload a complete simulation model from model repository to model database

The application `db_add_model_parameters_from_repository_to_db.py` allows to export the simulation model parameters from a
model repository to the model database for a given model version.
See the [database](databases.md#databases) section for implementation details on databases in simtools.

Example:

```bash
simtools-add_model-parameters-from-repository-to-db \
      --input_path /path/to/repository \
      --db_name new_db_name
```

This application loops over all subdirectories in `input_path` and uploads all json files to the database `new_db_name` (or updates an existing database with the same name):

- subdirectories starting with `OBS` are uploaded to the `sites` collection
- json files from the subdirectory `configuration_sim_telarray/configuration_corsika` are uploaded to the `configuration_sim_telarray/configuration_corsika` collection
- `Files` are added to the `files` collection
- all other json files are uploaded to collection defined in the array element description in [simtools/schemas/array_elements.yml](https://github.com/gammasim/simtools/blob/main/simtools/schemas/array_elements.yml). Allowed values are e.g., `telescopes`, `calibration_devices`.

## Import simulation model parameters

### Import Prod6 model parameters

Prod6 model parameters can be extracted from `sim_telarray` using the following commands.

```bash
./sim_telarray/bin/sim_telarray \
   -c sim_telarray/cfg/CTA/CTA-PROD6-LaPalma.cfg \
   -C limits=no-internal -C initlist=no-internal \
   -C list=no-internal -C typelist=no-internal \
   -C maximum_telescopes=30 -DNSB_AUTOSCALE \
   -DNECTARCAM -DHYPER_LAYOUT -DNUM_TELESCOPES=30 \
   /dev/null 2>|/dev/null | grep '(@cfg)'  >| all_telescope_config_la_palma.cfg

./sim_telarray/bin/sim_telarray \
   -c sim_telarray/cfg/CTA/CTA-PROD6-Paranal.cfg \
   -C limits=no-internal -C initlist=no-internal \
   -C list=no-internal -C typelist=no-internal \
   -C maximum_telescopes=87 -DNSB_AUTOSCALE \
   -DNECTARCAM -DHYPER_LAYOUT -DNUM_TELESCOPES=87 \
   /dev/null 2>|/dev/null | grep '(@cfg)'  >| all_telescope_config_paranal.cfg
```

### Import Prod5 model parameters

Prod5 model parameters can be extracted from `sim_telarray` using the following commands.

```bash
./sim_telarray/bin/sim_telarray \
   -c sim_telarray/cfg/CTA/CTA-PROD5-LaPalma.cfg \
   -C limits=no-internal -C initlist=no-internal \
   -C list=no-internal -C typelist=no-internal \
   -C maximum_telescopes=85 -DNSB_AUTOSCALE \
   -DNECTARCAM -DHYPER_LAYOUT -DNUM_TELESCOPES=85 \
   /dev/null 2>|/dev/null | grep '(@cfg)'  >| all_telescope_config_la_palma_prod5.cfg

./sim_telarray/bin/sim_telarray \
   -c sim_telarray/cfg/CTA/CTA-PROD5-Paranal.cfg \
   -C limits=no-internal -C initlist=no-internal \
   -C list=no-internal -C typelist=no-internal \
   -C maximum_telescopes=120 -DNSB_AUTOSCALE \
   -DNECTARCAM -DHYPER_LAYOUT -DNUM_TELESCOPES=120 \
   /dev/null 2>|/dev/null | grep '(@cfg)'  >| all_telescope_config_paranal_prod5.cfg
```

### Import telescope positions

Positions of array elements like telescopes are provided by CTAO in form of tables (typically ecsv files).
To import these positions into the model parameter repository, see the following example:

```bash
simtools-write-array-element-positions-to-repository \
    --input /path/to/positions.txt \
    --repository_path /path/to/repository \
    --model_version 1.0.0 \
    --coordinate_system ground \
    --site North
```