![](\"../Supplementary_data/dea_logo.jpg\"){kind=logo}
\n",
+ "\n",
+ "* [**Sign up to the DEA Sandbox**](https://docs.dea.ga.gov.au/setup/sandbox.html) to run this notebook interactively from a browser\n",
+ "* **Compatibility:** Notebook currently compatible with both the `NCI` and `DEA Sandbox` environments\n",
+ "* **Products used:** \n",
+ "[ga_s2am_ard_3](https://explorer.sandbox.dea.ga.gov.au/ga_s2am_ard_3), \n",
+ "[ga_s2bm_ard_3](https://explorer.sandbox.dea.ga.gov.au/ga_s2bm_ard_3)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "tags": []
+ },
+ "source": [
+ "## Background\n",
+ "\n",
+ "This notebook conducts time-series forecasting of vegetation condition (NDVI) using SARIMAX, a variation on [autoregressive-moving-average (ARMA)](https://en.wikipedia.org/wiki/Autoregressive%E2%80%93moving-average_model#ARMAX) models which includes an integrated (I) component to difference the timeseries so it becomes stationary, a seasonal (S) component, and has the capacity to consider exogenous (X) variables. \n",
+ "\n",
+ "In this example, we will conduct a forecast on a univariate NDVI timeseries. That is, our forecast will be built on temporal patterns in NDVI. Conversely, multivariate approaches can account for influences of variables such as soil moisture and rainfall.\n",
+ "\n",
+ "## Description\n",
+ "\n",
+ "In this notebook, we generate a NDVI timeseries from [Digital Earth Austalia's Sentinel-2 surface reflectance data](../DEA_products/DEA_Sentinel2_Surface_Reflectance.ipynb), then use it develop a forecasting algorithm.\n",
+ "\n",
+ "The following steps are taken:\n",
+ "\n",
+ "1. Load Sentinel-2 data and calculate NDVI.\n",
+ "2. Iterate through SARIMAX parameters and conduct model selection based on cross-validation.\n",
+ "3. Inspect model diagnostics\n",
+ "4. Forecast NDVI into the future and visualise the result.\n",
+ "\n",
+ "***"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Load packages\n",
+ "Import Python packages that are used for the analysis.\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "%matplotlib inline\n",
+ "import datacube\n",
+ "import xarray as xr\n",
+ "import pandas as pd\n",
+ "import numpy as np\n",
+ "from matplotlib import pyplot as plt\n",
+ "from tqdm.notebook import tqdm\n",
+ "from itertools import product\n",
+ "from statsmodels.tsa.api import SARIMAX\n",
+ "from statsmodels.tools.eval_measures import rmse\n",
+ "\n",
+ "from datacube import Datacube\n",
+ "\n",
+ "import sys\n",
+ "\n",
+ "sys.path.insert(1, \"../Tools/\")\n",
+ "from dea_tools.datahandling import load_ard\n",
+ "from dea_tools.plotting import display_map\n",
+ "from dea_tools.bandindices import calculate_indices\n",
+ "from dea_tools.dask import create_local_dask_cluster"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Set up a Dask cluster\n",
+ "\n",
+ "Dask can be used to better manage memory use down and conduct the analysis in parallel. \n",
+ "For an introduction to using Dask with Digital Earth Africa, see the [Dask notebook](../Beginners_guide/08_Parallel_processing_with_dask.ipynb).\n",
+ "\n",
+ ">**Note**: We recommend opening the Dask processing window to view the different computations that are being executed; to do this, see the *Dask dashboard in DE Africa* section of the [Dask notebook](../Beginners_guide/08_parallel_processing_with_dask.ipynb).\n",
+ "\n",
+ "To use Dask, set up the local computing cluster using the cell below."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "Client-1620909f-716d-11ee-81a0-66e207f94d07
\n", + "| Connection method: Cluster object | \n", + "Cluster type: distributed.LocalCluster | \n", + " \n", + "
| \n", + " Dashboard: /user/robbi.bishoptaylor@ga.gov.au/proxy/8787/status\n", + " | \n", + "\n", + " |
38aaf70d
\n", + "| \n", + " Dashboard: /user/robbi.bishoptaylor@ga.gov.au/proxy/8787/status\n", + " | \n", + "\n", + " Workers: 1\n", + " | \n", + "
| \n", + " Total threads: 2\n", + " | \n", + "\n", + " Total memory: 12.21 GiB\n", + " | \n", + "
| Status: running | \n", + "Using processes: True | \n", + "
Scheduler-9cb1453e-76b6-4dbd-862a-d16442fb60bc
\n", + "| \n", + " Comm: tcp://127.0.0.1:43641\n", + " | \n", + "\n", + " Workers: 1\n", + " | \n", + "
| \n", + " Dashboard: /user/robbi.bishoptaylor@ga.gov.au/proxy/8787/status\n", + " | \n", + "\n", + " Total threads: 2\n", + " | \n", + "
| \n", + " Started: Just now\n", + " | \n", + "\n", + " Total memory: 12.21 GiB\n", + " | \n", + "
| \n", + " Comm: tcp://127.0.0.1:34379\n", + " | \n", + "\n", + " Total threads: 2\n", + " | \n", + "
| \n", + " Dashboard: /user/robbi.bishoptaylor@ga.gov.au/proxy/41051/status\n", + " | \n", + "\n", + " Memory: 12.21 GiB\n", + " | \n", + "
| \n", + " Nanny: tcp://127.0.0.1:34979\n", + " | \n", + "\n", + " |
| \n", + " Local directory: /tmp/dask-scratch-space/worker-hzmfste0\n", + " | \n", + "
<xarray.Dataset>\n", + "Dimensions: (time: 280, y: 60, x: 52)\n", + "Coordinates:\n", + " * time (time) datetime64[ns] 2019-01-01T02:06:37.098000 ... 2022-08...\n", + " * y (y) float64 -3.69e+06 -3.69e+06 ... -3.691e+06 -3.691e+06\n", + " * x (x) float64 -9.719e+05 -9.719e+05 ... -9.709e+05 -9.709e+05\n", + " spatial_ref int32 3577\n", + "Data variables:\n", + " nbart_red (time, y, x) float32 dask.array<chunksize=(280, 60, 52), meta=np.ndarray>\n", + " nbart_nir_1 (time, y, x) float32 dask.array<chunksize=(280, 60, 52), meta=np.ndarray>\n", + "Attributes:\n", + " crs: EPSG:3577\n", + " grid_mapping: spatial_ref