The following pipeline will generate Forest Cover Maps for a given region, which are used to estimate changes in forest cover and train a model that can predict future changes. During the implementation of the pipeline, the following maps will be produced:
- Forest Covers
- Edge to Distance of the Forest
- Jurisdiction Maps
- Predicted Deforestation Maps There are 7 steps involved in the process, and at the end of each step, a physical action related to file movement is required (such as installing the file from the folder in the Google Drive).
Create a Python Environment and install all the dependencies with the help of requirements.txt available in the repository
<<<<<<< HEAD If in case osgeo/GDAL library's wheel is not installing, please install the pre-built wheel and install the package directly. You can find the prebuilt library for Windows (AMD 64bit) systems with Python 3.13 <a = https://drive.google.com/drive/folders/11syOJuhaX7qSCeN-mvk003r3JkrfuFyP?usp=drive_link>here.
If the osgeo/GDAL library's wheel is not installing, please install the pre-built wheel and install the package directly. You can find the prebuilt library for Windows (AMD 64-bit) systems with Python 3.13 <a = https://drive.google.com/drive/folders/11syOJuhaX7qSCeN-mvk003r3JkrfuFyP?usp=drive_link> here .
66d06095681033f16ad4bb8b9cf3a77a137eacf0
Google Earth Engine is used to generate the Forest Covers and Jurisdiction Maps and derive Edge to Distance of the Forest Maps. To run the GEE Operations please run the following CLI operation
python step.py 1 <DISTRICT = "Dhenkanal"> <START_YEAR = "2010"> <END_OF_MODELING_YEAR = "2015"> <PREDICTION_YEAR = "2020">
All the files produced will be saved in the Drive in the folder associated with the GEE account. The name of the folder is "GEE_exports_{}"; if no DISTRICT was provided, the folder will be named "GEE_exports_Dhenkanal".
Please install the entire folder in the "./data/" folder.
After the necessary folder has been installed in the "./data/" folder, running the following command will generate the prediction maps for the given region for the year 2020. The current implementation produces the results based on the guidelines provided by ClarkLabs {CITE}
python ./scripts/step.py 0 <DISTRICT = "Dhenkanal"> <START_YEAR = "2010"> <END_OF_MODELING_YEAR = "2015"> <PREDICTION_YEAR = "2020"
All the necessary prediction files will be saved in the "./data/GEE_exports_{}" folder.
Deforestation and Afforestation predictions are necessary for further analysis, hence running the following command calculates the deforestation and afforestation regions for the three different periods.
python ./scripts/FC_change.py
For the region of interest inside the district, for which sampling plots have been measured, it can be clustered to estimate the biomass.
A given sampling survey will contain the different trees recorded in the sample and the size of the region. Each tree species has an associated allometric equation, which can be used to construct an allometric model of the tree based on its characteristic features.
Upload the following spreadsheet containing the sampling survey for a region to the
"./data/GEE_exports_" folder:
- REGION + "_allometric.csv (REGION should be in uppercase)
All data in this repository was collected using the following survey datasheet format:
| Field Name | Description | Example | Required |
|---|---|---|---|
| Habdistrict | District Site Belongs to | Dhenkanal | Yes |
| Habpanchayat | Sampling Site | Pangatira | Yes |
| Plot No | Unique Identifier for the Sampling Plot | 1, 2, 3 ... | Yes |
| Plot Lat | GPS latitude of site in decimal degrees | 21.1618159 | Yes |
| Plot Long | GPS longitude of site in decimal degrees | 85.3657417 | Yes |
| Scientific Name | Scientific Name of the Species Sampled | Kendu Diospyros melanoxylon | Yes |
| Height | Height of the Species Sampled | 11 | Yes |
| DBH | Diameter at Breast Height of the Species | 6.366197724 | Yes |
| Scientific Name | Scientific Name of the Species Sampled | Kendu Diospyros melanoxylon | Yes |
Run the following command line to estimate the biomass:
python ./scripts/parser.py <REGION = "Pangatira">
The clustering step creates different clusters for the sampled region. Run the following command to save the clusters as a GEOTIF:
python ./scripts/KMeansClusters.py
Additionality is the biomass added or lost during the time period from the start of the modeling and prediction year. Run the following code to save the Additionality:
python ./scripts/Additionality2.py
For the final step, run the following step to calculate the total change in the weight of CO2 and Carbon that occurred during the period of modelling and prediction. Run the following command to estimate this change:
python ./scripts/step2.py
The following command line will estimate the biomass estimated in Step 5 with the prediciton results produced from a Regression Model that is trained on the GEDI Dataset for each cluster. <<<<<<< HEAD
The following pipeline will generate Forest Cover Maps for a given region, that are used to estimate the change in the forest cover and train a model that can predict the change in the near future. During the implemententioant of hte pipeline, the following maps will be produced:
- Forest Covers
- Edge to Distance of the Forest
- Jurisdiction Maps
- Predicted Deforestation Maps There are 1 steps involved in the process, and at the end of each step, a physical action related to file movement is required (such as installing the file from the folder in the Google Drive).
Google Earth Engine is used to generate the Forest Covers and Jurisdiction Maps and derive Edge to Distance of the Forest Maps. To run the GEE Operations please run the following CLI operation
python step.py 1 <DISTRICT = "Dhenkanal"> <START_YEAR = "2010"> <END_OF_MODELING_YEAR = "2015"> <PREDICTION_YEAR = "2020">
All the files produced will be saved in the Drive in the folder associated with the GEE account. The name of the folder is "GEE_exports_{}"; if no DISTRICT was provided, the folder will be named "GEE_exports_Dhenkanal".
Please install the entire folder in the "./data/" folder.
After the necessary folder has been installed in the "./data/" folder, running the following command will generate the prediction maps for the given region for the year 2020. The current implementation produces the results based on the guidelines provided from ClarkLabs {CITE}
python ./scripts/step.py 0 <DISTRICT = "Dhenkanal"> <START_YEAR = "2010"> <END_OF_MODELING_YEAR = "2015"> <PREDICTION_YEAR = "2020"
All the necessary prediction files will be saved in the "./data/GEE_exports_{}" folder.
Deforestation and Afforestation predictions are necessary for further analysis, hence running the following command calculates the deforestation and afforestation regions for the three different periods.
python ./scripts/FC_change.py
For the region of interest inside the district, for which sampling plots have been measured, it can be clustered to estimate the biomass.
A given sampling survey will contain the different trees recorded in the sample and the size of the region. Each tree species has an associated allometric equation, which can build allometric equation of the tree based on characteristic features of the tree.
Upload the following spreadsheet containing the sampling survey for a region to the
"./data/GEE_exports_" folder:
- REGION + "_allometric.csv (REGION should be in uppercase)
Run the following the command line to estimate the biomass:
python ./scripts/parser.py <REGION = "Pangatira">
The clustering step creates different clusters for the sampled region. Run the following command to save the clusters as a GEOTIF:
python ./scripts/KMeansClusters.py
Additionality is the biomass added or lost during the time period from the start of the modeling and prediction year. Run the following code to save the Additionality:
python ./scripts/Additionality2.py
For the final step, run the following step to calculate the total change in the weight of CO2 and Carbon that occured during the period of modeling and prediction. Run the following command to estimate this change:
python ./scripts/step2.py
The following command line will estimate the biomass estimated in Step 5 with the prediciton results produced from a Regression Model that is trained on the GEDI Dataset for each cluster.
66d06095681033f16ad4bb8b9cf3a77a137eacf0
python ./scripts/Evaluator.py