zonal stats, rounding.

preprocess_part2.ipynb

@@ -0,0 +1,1165 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "39bf1de3-cba6-475a-a988-ad48e5af4a04",
+   "metadata": {},
+   "source": [
+    "# Get zonal stats "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "ba047a55-642d-4c27-a367-5f35f4406218",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import ibis\n",
+    "import ibis.selectors as s\n",
+    "from ibis import _\n",
+    "import fiona\n",
+    "import geopandas as gpd\n",
+    "import rioxarray\n",
+    "from shapely.geometry import box\n",
+    "\n",
+    "import rasterio\n",
+    "from rasterio.mask import mask\n",
+    "from rasterstats import zonal_stats\n",
+    "import pandas as pd\n",
+    "from joblib import Parallel, delayed\n",
+    "\n",
+    "con = ibis.duckdb.connect()\n",
+    "con.load_extension(\"spatial\")\n",
+    "threads = -1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "8b5656db-2d1d-4ca8-826d-7588126e52e8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# cropping US data to only CA \n",
+    "def crop_raster_to_bounds(tif_file, vector_gdf):\n",
+    "    with rasterio.open(tif_file) as src:\n",
+    "        # Get California's bounding box in the same CRS as the raster\n",
+    "        california_bounds = vector_gdf.total_bounds\n",
+    "        california_bounds = rasterio.coords.BoundingBox(\n",
+    "            *california_bounds\n",
+    "        )\n",
+    "        # Crop the raster to the California bounding box\n",
+    "        out_image, out_transform = mask(src, [california_bounds], crop=True)\n",
+    "        out_meta = src.meta.copy()\n",
+    "        out_meta.update({\n",
+    "            \"driver\": \"GTiff\",\n",
+    "            \"height\": out_image.shape[1],\n",
+    "            \"width\": out_image.shape[2],\n",
+    "            \"transform\": out_transform\n",
+    "        })\n",
+    "        print(\"Unique values in cropped raster:\", np.unique(out_image))\n",
+    "\n",
+    "    return out_image, out_meta\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "9a0e3446-16ac-40b0-9e34-db0157038c5a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def big_zonal_stats(vec_file, tif_file, stats, col_name, n_jobs, verbose=10, timeout=10000):\n",
+    "    gdf = gpd.read_parquet(vec_file)\n",
+    "    if gdf.crs is None:\n",
+    "        gdf = gdf.set_crs(\"EPSG:4326\")\n",
+    "    gdf = gdf.rename(columns={\"geom\": \"geometry\"})\n",
+    "    gdf = gdf.set_geometry(\"geometry\")\n",
+    "    gdf = gdf[gdf[\"geometry\"].notna()].copy()\n",
+    "\n",
+    "    with rasterio.open(tif_file) as src:\n",
+    "        raster_crs = src.crs\n",
+    "        gdf = gdf.to_crs(raster_crs)  # Transform vector to raster CRS\n",
+    "        \n",
+    "        # CA bounding box + convert it to a polygon in raster CRS\n",
+    "        california_polygon = box(*gdf.total_bounds)\n",
+    "        \n",
+    "        out_image, out_transform = mask(src, [california_polygon], crop=True, nodata=src.nodata)\n",
+    "\n",
+    "        # If raster is 3D, select the first band\n",
+    "        if out_image.ndim == 3:\n",
+    "            out_image = out_image[0]\n",
+    "\n",
+    "    # compute zonal statistics for each geometry slice\n",
+    "    def get_stats(geom_slice):\n",
+    "        geom = [geom_slice.geometry]\n",
+    "        stats_result = zonal_stats(\n",
+    "            geom, out_image, stats=stats, affine=out_transform, all_touched=True, nodata=src.nodata\n",
+    "        )\n",
+    "        return stats_result[0] if stats_result and stats_result[0].get(\"mean\") is not None else {'mean': None}\n",
+    "\n",
+    "    output = [get_stats(row) for row in gdf.itertuples()]\n",
+    "    gdf[col_name] = [res['mean'] for res in output]\n",
+    "\n",
+    "    return gdf"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "ce66bae6-bac5-4837-9b01-fde16a00c303",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# getting local copies of data \n",
+    "# aws s3 cp s3://vizzuality/hfp-100/hfp_2021_100m_v1-2_cog.tif . --endpoint-url=https://data.source.coop\n",
+    "# aws s3 cp s3://vizzuality/lg-land-carbon-data/natcrop_bii_100m_cog.tif . --endpoint-url=https://data.source.coop\n",
+    "# aws s3 cp s3://vizzuality/lg-land-carbon-data/natcrop_fii_100m_cog.tif . --endpoint-url=https://data.source.coop\n",
+    "# aws s3 cp s3://vizzuality/lg-land-carbon-data/natcrop_expansion_100m_cog.tif . --endpoint-url=https://data.source.coop\n",
+    "# aws s3 cp s3://vizzuality/lg-land-carbon-data/natcrop_reduction_100m_cog.tif . --endpoint-url=https://data.source.coop\n",
+    "# aws s3 cp s3://cboettig/carbon/cogs/irrecoverable_c_total_2018.tif . --endpoint-url=https://data.source.coop\n",
+    "# aws s3 cp s3://cboettig/carbon/cogs/manageable_c_total_2018.tif . --endpoint-url=https://data.source.coop\n",
+    "# ! aws s3 cp s3://cboettig/justice40/disadvantaged-communities.parquet . --endpoint-url=https://data.source.coop\n",
+    "# minio/shared-biodiversity/redlist/cog/combined_sr_2022.tif\n",
+    "# /home/rstudio/minio/shared-biodiversity/redlist/cog/combined_rwr_2022.tif\n",
+    "# ! aws s3 cp s3://cboettig/social-vulnerability/svi2020_us_tract.parquet . --endpoint-url=https://data.source.coop\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "531e7f88-1ce1-4027-b0ab-aab597e9a2b2",
+   "metadata": {},
+   "source": [
+    "# Biodiversity Data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "66dec912-ad8a-41cf-a5c2-6ec9cc350984",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'SpeciesRichness_All.tif'\n",
+    "vec_file = \"/home/rstudio/github/ca-30x30/ca2024-30m.parquet\"\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'], col_name = \"richness\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "b081ec1a-ea91-485e-95f9-12cd06c2002a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'RSR_All.tif'\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],\n",
+    "                      col_name = \"rsr\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "d5133f36-404e-4f6a-a90b-eb5f098e6f06",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'combined_sr_2022.tif'\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],  col_name = \"all_species_richness\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "2ce56a66-34e3-4f61-95ae-65d1f06bc468",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'combined_rwr_2022.tif'\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],  col_name = \"all_species_rwr\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "6c129894-3775-4842-8767-f81a8f626d2c",
+   "metadata": {},
+   "source": [
+    "# Carbon Data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "19c3e402-8712-450f-b3dd-af9d0c01689c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'irrecoverable_c_total_2018.tif'\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],  col_name = \"irrecoverable_carbon\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "c55c777a-48ce-4403-a171-cfc0d2351df6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'manageable_c_total_2018.tif'\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],  col_name = \"manageable_carbon\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "33ac0fb7-2cde-448d-a634-1973e34ac14f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'deforest_carbon_100m_cog.tif'\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'], \n",
+    "                     col_name = \"deforest_carbon\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "096c00a8-57af-41d7-93cc-85d85414aa4f",
+   "metadata": {},
+   "source": [
+    "# Human Impact Data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "d2a8c10f-e94b-4eef-940f-2af9599edee1",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'natcrop_bii_100m_cog.tif'\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'], \n",
+    "                     col_name = \"biodiversity_intactness_loss\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "1c318f39-7ca0-4f3c-80fb-73f72202e4e0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'natcrop_fii_100m_cog.tif'\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],\n",
+    "                     col_name = \"forest_integrity_loss\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "aef9070a-c87a-463e-81b8-3cc6c5c9d484",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'natcrop_expansion_100m_cog.tif'\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],  col_name = \"crop_expansion\", n_jobs=threads, verbose=0)\n",
+    "gpd.GeoDataFrame(df, geometry=\"geometry\").to_parquet(\"cpad-stats-temp.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "d94f937b-b32c-4de1-b4ac-93ce33f0919f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'natcrop_reduction_100m_cog.tif'\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],  col_name = \"crop_reduction\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "6bdaba61-30c1-49d6-a4e6-db68f1daafa3",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'hfp_2021_100m_v1-2_cog.tif'\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],  col_name = \"human_impact\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f8e037d4-7a34-42bc-941f-0c09ee80ef3b",
+   "metadata": {},
+   "source": [
+    "# Need to convert SVI & Justice40 files to tif"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "c4a19013-65f1-4eef-be2d-0cf1be3d0f7f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import geopandas as gpd\n",
+    "import numpy as np\n",
+    "import rasterio\n",
+    "from rasterio.features import rasterize\n",
+    "from rasterio.transform import from_bounds\n",
+    "\n",
+    "def get_geotiff(gdf, output_file,col):\n",
+    "    gdf = gdf.set_geometry(\"geometry\")\n",
+    "    gdf = gdf.set_crs(\"EPSG:4326\")\n",
+    "    print(gdf.crs)\n",
+    "\n",
+    "    # Set raster properties\n",
+    "    minx, miny, maxx, maxy = gdf.total_bounds  # Get the bounds of the geometry\n",
+    "    pixel_size = 0.01  # Define the pixel size in units of the CRS\n",
+    "    width = int((maxx - minx) / pixel_size)\n",
+    "    height = int((maxy - miny) / pixel_size)\n",
+    "    transform = from_bounds(minx, miny, maxx, maxy, width, height)\n",
+    "    \n",
+    "    # Define rasterization with continuous values\n",
+    "    shapes = ((geom, value) for geom, value in zip(gdf.geometry, gdf[col]))\n",
+    "    raster = rasterize(\n",
+    "        shapes,\n",
+    "        out_shape=(height, width),\n",
+    "        transform=transform,\n",
+    "        fill=0.0,  # Background value for areas outside the geometry\n",
+    "        dtype=\"float32\"  # Set data type to handle continuous values\n",
+    "    )\n",
+    "    print(\"Unique values in raster:\", np.unique(raster))\n",
+    "\n",
+    "    # Define GeoTIFF metadata\n",
+    "    out_meta = {\n",
+    "        \"driver\": \"GTiff\",\n",
+    "        \"height\": height,\n",
+    "        \"width\": width,\n",
+    "        \"count\": 1,\n",
+    "        \"dtype\": raster.dtype,\n",
+    "        \"crs\": gdf.crs,\n",
+    "        \"transform\": transform,\n",
+    "        \"compress\": \"deflate\"  # Use compression to reduce file size\n",
+    "    }\n",
+    "    \n",
+    "    # Write to a GeoTIFF file with COG options\n",
+    "    with rasterio.open(output_file, \"w\", **out_meta) as dest:\n",
+    "        dest.write(raster, 1)\n",
+    "        dest.build_overviews([2, 4, 8, 16], rasterio.enums.Resampling.average)\n",
+    "        dest.update_tags(1, TIFFTAG_RESOLUTION_UNIT=\"Meter\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f4925a74-5ed2-49a4-845b-6a0f0398a43e",
+   "metadata": {},
+   "source": [
+    "# SVI"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "4e678f01-73af-4f99-a565-e9b7f04d9547",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# clean up SVI data\n",
+    "svi_df =  (con\n",
+    " .read_parquet(\"svi2020_us_tract.parquet\")\n",
+    " .select(\"RPL_THEMES\",\"RPL_THEME1\",\"RPL_THEME2\",\"RPL_THEME3\",\"RPL_THEME4\",\"Shape\")\n",
+    " .rename(SVI = \"RPL_THEMES\", socioeconomic = \"RPL_THEME1\", \n",
+    "         household_char = \"RPL_THEME2\", racial_ethnic_minority =  \"RPL_THEME3\",\n",
+    "         housing_transit = \"RPL_THEME4\", geometry = \"Shape\")\n",
+    ".cast({\"geometry\":\"geometry\"})\n",
+    ")\n",
+    "svi_df.execute().to_parquet(\"svi2020_us_tract_clean.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "c5046d6b-9798-46d3-a1bc-548e29414007",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "gdf = gpd.read_parquet(\"svi2020_us_tract_clean.parquet\")\n",
+    "svi = gdf[['SVI','geometry']]\n",
+    "socio = gdf[['socioeconomic','geometry']]\n",
+    "house = gdf[['household_char','geometry']]\n",
+    "minority = gdf[['racial_ethnic_minority','geometry']]\n",
+    "transit = gdf[['housing_transit','geometry']]\n",
+    "\n",
+    "#convert SVI parquet to tif\n",
+    "get_geotiff(svi,\"svi.tif\",\"SVI\")\n",
+    "get_geotiff(socio,\"svi_socioeconomic.tif\",\"socioeconomic\")\n",
+    "get_geotiff(house,\"svi_household.tif\",\"household_char\")\n",
+    "get_geotiff(minority,\"svi_minority.tif\",\"racial_ethnic_minority\")\n",
+    "get_geotiff(transit,\"svi_transit.tif\",\"housing_transit\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "6a36b77f-d0be-45bd-9318-da4b57eaf353",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "tif_file = 'svi.tif'\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],  col_name = \"SVI\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "05ef74e2-3f23-4f69-8cd3-8862cb73a259",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "tif_file = 'svi_socioeconomic.tif'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],  col_name = \"socioeconomic_status\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "23417a03-38c2-4b31-8340-f08ec8a11631",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "tif_file = 'svi_household.tif'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],  col_name = \"household_char\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "de86d7f0-6cdc-4d05-bdee-d9803cbd83bd",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "tif_file = 'svi_minority.tif'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],  col_name = \"racial_ethnic_minority\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "0c49dd50-7dd3-4240-9af8-3e32ec656bc0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "tif_file = 'svi_transit.tif'\n",
+    "df = big_zonal_stats(vec_file, tif_file, stats = ['mean'],  col_name = \"housing_transit\", n_jobs=threads, verbose=0).to_parquet(\"cpad-stats-temp.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ff4b6604-9828-4882-90bd-554c21f5c6e6",
+   "metadata": {},
+   "source": [
+    "# Justice40 "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3678a91f-72f7-4339-a409-a97776cba043",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#clean up\n",
+    "justice40 =  (con\n",
+    "        .read_parquet(\"disadvantaged-communities.parquet\")\n",
+    "        .rename(geometry = \"SHAPE\",justice40=\"Disadvan\")\n",
+    "        .filter(_.StateName == \"California\")\n",
+    "        .mutate(geometry = _.geometry.convert(\"ESRI:102039\",\"EPSG:4326\"))\n",
+    "        .select(\"justice40\",\"geometry\")\n",
+    "        )\n",
+    "justice40.execute().to_parquet(\"ca_justice40.parquet\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "8faa425f-6f9c-4189-a53a-24dd0250c539",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# #justice40 is either 0 or 1, so we want to get the percentage of polygon where justice40 = 1.  \n",
+    "\n",
+    "def big_zonal_stats_binary(vec_file, justice40_file, col_name,projected_crs=\"EPSG:3310\"):\n",
+    "    # Read both vector files as GeoDataFrames\n",
+    "    gdf = gpd.read_parquet(vec_file)\n",
+    "    justice40_gdf = gpd.read_parquet(justice40_file)\n",
+    "    \n",
+    "    # Set CRS if not already set (assuming both should be in EPSG:4326, modify if needed)\n",
+    "    if gdf.crs is None:\n",
+    "        gdf = gdf.set_crs(\"EPSG:4326\")\n",
+    "    if justice40_gdf.crs is None:\n",
+    "        justice40_gdf = justice40_gdf.set_crs(\"EPSG:4326\")\n",
+    "     # Ensure both GeoDataFrames are in the same CRS and reproject to a projected CRS for area calculations\n",
+    "    gdf = gdf.to_crs(projected_crs)\n",
+    "    justice40_gdf = justice40_gdf.to_crs(projected_crs)\n",
+    "    \n",
+    "    # Ensure both GeoDataFrames are in the same CRS\n",
+    "    gdf = gdf.to_crs(justice40_gdf.crs)\n",
+    "    \n",
+    "    # Filter justice40 polygons where justice40 == 1\n",
+    "    justice40_gdf = justice40_gdf[justice40_gdf['justice40'] == 1].copy()\n",
+    "    \n",
+    "    # Prepare a list to hold percentage of justice40 == 1 for each polygon\n",
+    "    percentages = []\n",
+    "    \n",
+    "    # Iterate over each polygon in the main GeoDataFrame\n",
+    "    for geom in gdf.geometry:\n",
+    "        # Find intersecting justice40 polygons\n",
+    "        justice40_intersections = justice40_gdf[justice40_gdf.intersects(geom)].copy()\n",
+    "        \n",
+    "        # Calculate the intersection area\n",
+    "        if not justice40_intersections.empty:\n",
+    "            justice40_intersections['intersection'] = justice40_intersections.intersection(geom)\n",
+    "            total_intersection_area = justice40_intersections['intersection'].area.sum()\n",
+    "            \n",
+    "            # Calculate percentage based on original polygon's area\n",
+    "            percentage_1 = (total_intersection_area / geom.area) \n",
+    "        else:\n",
+    "            percentage_1 = 0.0  # No intersection with justice40 == 1 polygons\n",
+    "        \n",
+    "        # Append result\n",
+    "        percentages.append(percentage_1)\n",
+    "    \n",
+    "    # Add results to the original GeoDataFrame\n",
+    "    gdf[col_name] = percentages\n",
+    "    return gdf\n",
+    "\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "fe80fc28-73ce-4a26-9925-851c2798e467",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "vec_file = './cpad-stats-temp.parquet'\n",
+    "\n",
+    "df = big_zonal_stats_binary(vec_file, \"ca_justice40.parquet\", col_name=\"percent_disadvantaged\")\n",
+    "df.to_parquet(\"cpad-stats-temp.parquet\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5438a4f4-377e-41fe-800b-8ffc1f33caa0",
+   "metadata": {},
+   "source": [
+    "# Fire"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "4bd83b4d-01df-49d8-99e1-6740d365c833",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import geopandas as gpd\n",
+    "\n",
+    "#get percentage of polygon with fire occurrence \n",
+    "def fire_stats(file_name, fire_df, col_name):\n",
+    "    gdf = gpd.read_parquet(file_name)\n",
+    "    \n",
+    "    percentages = []\n",
+    "    # Find all fires that intersect with the current protected area    \n",
+    "    for geom in gdf.geometry:\n",
+    "        fire_intersections = fire_df[fire_df.intersects(geom)].copy()\n",
+    "        if not fire_intersections.empty:\n",
+    "            # If there is only one intersecting fire, compute the intersection area\n",
+    "            if len(fire_intersections) == 1:\n",
+    "                intersection_area = fire_intersections.geometry.iloc[0].intersection(geom).area\n",
+    "            else:\n",
+    "                # If there are multiple intersecting fires, use a union to avoid double-counting\n",
+    "                unioned_fires = fire_intersections.unary_union\n",
+    "                intersection_area = unioned_fires.intersection(geom).area\n",
+    "    \n",
+    "            percentage_1 = round((intersection_area / geom.area),3)\n",
+    "        else:\n",
+    "            percentage_1 = 0.0  \n",
+    "\n",
+    "        percentages.append(percentage_1)\n",
+    "    \n",
+    "    gdf[col_name] = percentages\n",
+    "    return gdf\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "4ce35cea-8897-42c0-b1f6-01b414a5b556",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#historical fire perimeters \n",
+    "fire_20 =  (con\n",
+    "        .read_parquet(\"firep22_1.parquet\")\n",
+    "        .rename(year = \"YEAR_\")\n",
+    "        .filter(_.STATE == \"CA\", _.year != '')\n",
+    "        .cast({\"year\":\"int\"})\n",
+    "        .filter(_.year>=2003)\n",
+    "        .select(\"year\",\"geometry\")\n",
+    "        .mutate(\n",
+    "            geometry=ibis.ifelse(\n",
+    "            _.geometry.is_valid(),\n",
+    "                _.geometry,  # Keep the geometry if it's valid\n",
+    "                _.geometry.buffer(0)  # Apply buffer(0) to fix invalid geometries\n",
+    "                )\n",
+    "            )\n",
+    "        )\n",
+    "fire_20.execute().to_parquet(\"ca-fire-20yrs.parquet\")\n",
+    "fire_10 = fire_20.filter(_.year>=2013)\n",
+    "fire_5 =  fire_20.filter(_.year>=2018)\n",
+    "fire_2 =  fire_20.filter(_.year>=2022)\n",
+    "\n",
+    "\n",
+    "fire_20_df = fire_20.execute().set_crs(\"EPSG:3310\")\n",
+    "fire_10_df = fire_10.execute().set_crs(\"EPSG:3310\")\n",
+    "fire_5_df = fire_5.execute().set_crs(\"EPSG:3310\")\n",
+    "fire_2_df = fire_2.execute().set_crs(\"EPSG:3310\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "0a041210-6ffe-49b0-b4a7-3a9220acedb9",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#prescribed burns\n",
+    "rxburn_20 =  (con\n",
+    "        .read_parquet(\"rxburn22_1.parquet\")\n",
+    "        .rename(year = \"YEAR_\")\n",
+    "        .filter(_.STATE == \"CA\", _.year != ' ', _.year != '')\n",
+    "        .cast({\"year\":\"int\"})\n",
+    "        .filter(_.year>=2003)\n",
+    "        .select(\"year\",\"geometry\")\n",
+    "        .mutate(\n",
+    "            geometry=ibis.ifelse(\n",
+    "            _.geometry.is_valid(),\n",
+    "                _.geometry,  # Keep the geometry if it's valid\n",
+    "                _.geometry.buffer(0)  # Apply buffer(0) to fix invalid geometries\n",
+    "                )\n",
+    "            )\n",
+    "        )\n",
+    "\n",
+    "rxburn_20.execute().to_parquet(\"ca-rxburn-20yrs.parquet\")\n",
+    "rxburn_10 =  (rxburn_20.filter(_.year>=2013))\n",
+    "rxburn_5 =  (rxburn_20.filter(_.year>=2018))\n",
+    "rxburn_2 =  (rxburn_20.filter(_.year>=2022))\n",
+    "\n",
+    "rxburn_20_df = rxburn_20.execute().set_crs(\"EPSG:3310\")\n",
+    "rxburn_10_df = rxburn_10.execute().set_crs(\"EPSG:3310\")\n",
+    "rxburn_5_df = rxburn_5.execute().set_crs(\"EPSG:3310\")\n",
+    "rxburn_2_df = rxburn_2.execute().set_crs(\"EPSG:3310\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "fc955b02-efc1-4ae3-b8e4-ea424d491a68",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# need to validate geometries, using epsg:3310 to match fire polygons\n",
+    "ca = (con\n",
+    "      .read_parquet('cpad-stats-temp.parquet')\n",
+    "      .mutate(geom = _.geom.convert(\"EPSG:4326\",\"EPSG:3310\"))\n",
+    "      .mutate(\n",
+    "            geometry=ibis.ifelse(\n",
+    "            _.geom.is_valid(),\n",
+    "                _.geom,  # Keep the geometry if it's valid\n",
+    "                _.geom.buffer(0)  # Apply buffer(0) to fix invalid geometries\n",
+    "                )\n",
+    "            )\n",
+    "      .drop('geom')\n",
+    "     )\n",
+    "gdf = ca.execute()\n",
+    "gdf = gdf.set_crs('EPSG:3310')\n",
+    "gdf.to_parquet('cpad-stats-temp-EPSG3310.parquet')\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "68e25266-efc8-4378-afc5-95c7a769ca81",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%time\n",
+    "file_name = 'cpad-stats-temp-EPSG3310.parquet'\n",
+    "\n",
+    "names = [\"percent_fire_20yr\", \"percent_fire_10yr\", \"percent_fire_5yr\",\n",
+    "         \"percent_fire_2yr\",\"percent_rxburn_20yr\", \"percent_rxburn_10yr\", \n",
+    "         \"percent_rxburn_5yr\",\"percent_rxburn_2yr\"]\n",
+    "dfs = [fire_20_df,fire_10_df,fire_5_df,fire_2_df,rxburn_20_df,rxburn_10_df,rxburn_5_df,rxburn_2_df]\n",
+    "\n",
+    "for df,name in zip(dfs,names):\n",
+    "    df_stat = fire_stats(file_name,df, col_name=name)\n",
+    "    df_stat.to_parquet(file_name)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "cd4acb35-d1a3-4632-ae30-c6e3e923e94c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#save data back to cpad-stats-temp\n",
+    "# (not really necessary but I want to reuse the same code)\n",
+    "ca = (con\n",
+    "      .read_parquet(file_name)\n",
+    "     .mutate(geometry = _.geometry.convert(\"EPSG:3310\",\"EPSG:4326\"))\n",
+    "     )\n",
+    "gdf = ca.execute()\n",
+    "gdf= gdf.set_crs('EPSG:4326')\n",
+    "gdf.to_parquet(\"cpad-stats-temp.parquet\")\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "e3083b85-1322-4188-ac08-e73c2570978c",
+   "metadata": {},
+   "source": [
+    "# Cleaning up + Rounding floats"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "2e4de199-82d4-4e2b-8572-6fe19b57d1ee",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "## clean up\n",
+    "con = ibis.duckdb.connect(extensions=[\"spatial\"])\n",
+    "ca_geom = con.read_parquet(\"ca2024-30m.parquet\").cast({\"geom\":\"geometry\"}).select(\"id\",\"geom\")\n",
+    "\n",
+    "ca = (con\n",
+    "        .read_parquet(\"cpad-stats-temp.parquet\")\n",
+    "        .cast({\n",
+    "                \"crop_expansion\": \"int64\",\n",
+    "                \"crop_reduction\": \"int64\",\n",
+    "                \"manageable_carbon\": \"int64\",\n",
+    "                \"irrecoverable_carbon\": \"int64\"\n",
+    "            })\n",
+    "        .mutate(\n",
+    "                richness=_.richness.round(3),\n",
+    "                rsr=_.rsr.round(3),\n",
+    "                all_species_rwr=_.all_species_rwr.round(3),\n",
+    "                all_species_richness=_.all_species_richness.round(3),\n",
+    "                percent_disadvantaged=(_.percent_disadvantaged).round(3),\n",
+    "                svi=_.svi.round(3),\n",
+    "                svi_socioeconomic_status=_.socioeconomic_status.round(3),\n",
+    "                svi_household_char=_.household_char.round(3),\n",
+    "                svi_racial_ethnic_minority=_.racial_ethnic_minority.round(3),\n",
+    "                svi_housing_transit=_.housing_transit.round(3),\n",
+    "                human_impact=_.human_impact.round(3),\n",
+    "                deforest_carbon=_.deforest_carbon.round(3),\n",
+    "                biodiversity_intactness_loss=_.biodiversity_intactness_loss.round(3),\n",
+    "                forest_integrity_loss=_.forest_integrity_loss.round(3),\n",
+    "                percent_fire_20yr = _.percent_fire_20yr.round(3),\n",
+    "                percent_fire_10yr = _.percent_fire_10yr.round(3),\n",
+    "                percent_fire_5yr = _.percent_fire_5yr.round(3),\n",
+    "                percent_fire_2yr = _.percent_fire_2yr.round(3),\n",
+    "                percent_rxburn_20yr = _.percent_rxburn_20yr.round(3),\n",
+    "                percent_rxburn_10yr = _.percent_rxburn_10yr.round(3),\n",
+    "                percent_rxburn_5yr = _.percent_rxburn_5yr.round(3),\n",
+    "                percent_rxburn_2yr = _.percent_rxburn_2yr.round(3),\n",
+    "            )\n",
+    "          # only grabbing columns we are making charts with \n",
+    "          .select('established', 'reGAP', 'name', 'access_type', 'manager', 'manager_type', 'Easement', 'Acres', 'id', 'type','richness', \n",
+    "                 'rsr', 'irrecoverable_carbon', 'manageable_carbon', 'percent_fire_20yr', 'percent_fire_10yr', 'percent_fire_5yr','percent_fire_2yr',\n",
+    "                'percent_rxburn_20yr', 'percent_rxburn_10yr', 'percent_rxburn_5yr','percent_rxburn_2yr', 'percent_disadvantaged',\n",
+    "                  'svi', 'svi_socioeconomic_status', 'svi_household_char', 'svi_racial_ethnic_minority',\n",
+    "                  'svi_housing_transit', 'deforest_carbon','human_impact'\n",
+    "                 )\n",
+    "          .join(ca_geom, \"id\", how=\"inner\")\n",
+    "        )\n",
+    "\n",
+    "ca.head(5).execute()\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "3780de2c-3a68-442c-bb3b-64c792418979",
+   "metadata": {},
+   "source": [
+    "# Save as PMTiles + Upload data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "05c791c9-888a-483a-9dbb-a2ba7eb1bce2",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import subprocess\n",
+    "import os\n",
+    "from huggingface_hub import HfApi, login\n",
+    "import streamlit as st\n",
+    "\n",
+    "login(st.secrets[\"HF_TOKEN\"])\n",
+    "# api = HfApi(add_to_git_credential=False)\n",
+    "api = HfApi()\n",
+    "\n",
+    "def hf_upload(file, repo_id,repo_type):\n",
+    "    info = api.upload_file(\n",
+    "            path_or_fileobj=file,\n",
+    "            path_in_repo=file,\n",
+    "            repo_id=repo_id,\n",
+    "            repo_type=repo_type,\n",
+    "        )\n",
+    "def generate_pmtiles(input_file, output_file, max_zoom=12):\n",
+    "    # Ensure Tippecanoe is installed\n",
+    "    if subprocess.call([\"which\", \"tippecanoe\"], stdout=subprocess.DEVNULL) != 0:\n",
+    "        raise RuntimeError(\"Tippecanoe is not installed or not in PATH\")\n",
+    "\n",
+    "    # Construct the Tippecanoe command\n",
+    "    command = [\n",
+    "        \"tippecanoe\",\n",
+    "        \"-o\", output_file,\n",
+    "        \"-zg\",\n",
+    "        \"--extend-zooms-if-still-dropping\",\n",
+    "        \"--force\",\n",
+    "        \"--projection\", \"EPSG:4326\",  \n",
+    "        \"-L\",\"layer:\"+input_file,\n",
+    "    ]\n",
+    "    # Run Tippecanoe\n",
+    "    try:\n",
+    "        subprocess.run(command, check=True)\n",
+    "        print(f\"Successfully generated PMTiles file: {output_file}\")\n",
+    "    except subprocess.CalledProcessError as e:\n",
+    "        print(f\"Error running Tippecanoe: {e}\")\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "1f2d179d-6d47-4e84-83c6-7cb3d969fc00",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "gdf = ca.execute().set_crs(\"EPSG:4326\")\n",
+    "gdf.to_file(\"cpad-stats.geojson\")\n",
+    "\n",
+    "generate_pmtiles(\"cpad-stats.geojson\", \"cpad-stats.pmtiles\")\n",
+    "hf_upload(\"cpad-stats.pmtiles\", \"boettiger-lab/ca-30x30\",\"dataset\")\n",
+    "\n",
+    "gdf.to_parquet(\"cpad-stats.parquet\")\n",
+    "hf_upload(\"cpad-stats.parquet\", \"boettiger-lab/ca-30x30\",\"dataset\")\n",
+    "hf_upload(\"cpad-stats.parquet\", \"boettiger-lab/ca-30x30\",\"space\")\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "09467342-c160-413b-9cdc-31a4bec968cf",
+   "metadata": {},
+   "source": [
+    "# Redoing fire polygons pmtiles to have each range be its own layer  "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "2161c50b-0328-474f-aa57-215e14fe33c2",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def generate_pmtiles(input_file1, input_file2, input_file3, input_file4, output_file, max_zoom=12):\n",
+    "    # Ensure Tippecanoe is installed\n",
+    "    if subprocess.call([\"which\", \"tippecanoe\"], stdout=subprocess.DEVNULL) != 0:\n",
+    "        raise RuntimeError(\"Tippecanoe is not installed or not in PATH\")\n",
+    "\n",
+    "    # Construct the Tippecanoe command\n",
+    "    command = [\n",
+    "        \"tippecanoe\",\n",
+    "        \"-o\", output_file,\n",
+    "        \"-zg\",\n",
+    "        \"--extend-zooms-if-still-dropping\",\n",
+    "        \"--force\",\n",
+    "        \"--projection\", \"EPSG:4326\",  \n",
+    "        \"-L\",\"layer1:\"+input_file1,\n",
+    "        \"-L\",\"layer2:\"+input_file2,\n",
+    "        \"-L\",\"layer3:\"+input_file3,\n",
+    "        \"-L\",\"layer4:\"+input_file4,\n",
+    "\n",
+    "    ]\n",
+    "    # Run Tippecanoe\n",
+    "    try:\n",
+    "        subprocess.run(command, check=True)\n",
+    "        print(f\"Successfully generated PMTiles file: {output_file}\")\n",
+    "    except subprocess.CalledProcessError as e:\n",
+    "        print(f\"Error running Tippecanoe: {e}\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3a15d11f-ef32-4af3-8b72-b43acd43cf08",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "rxburn_20 =  (con\n",
+    "        .read_parquet(\"rxburn22_1.parquet\")\n",
+    "        .rename(year = \"YEAR_\")\n",
+    "        .filter(_.STATE == \"CA\", _.year != ' ', _.year != '')\n",
+    "        .cast({\"year\":\"int\"})\n",
+    "        .filter(_.year>=2003)\n",
+    "        .mutate(\n",
+    "            geometry=ibis.ifelse(\n",
+    "            _.geometry.is_valid(),\n",
+    "                _.geometry,  # Keep the geometry if it's valid\n",
+    "                _.geometry.buffer(0)  # Apply buffer(0) to fix invalid geometries\n",
+    "                )\n",
+    "            )\n",
+    "        .mutate(geometry = _.geometry.convert(\"EPSG:3310\",\"EPSG:4326\"))\n",
+    "        )\n",
+    "\n",
+    "rxburn_10 =  (rxburn_20.filter(_.year>=2013))\n",
+    "rxburn_5 =  (rxburn_20.filter(_.year>=2018))\n",
+    "rxburn_2 =  (rxburn_20.filter(_.year>=2022))\n",
+    "\n",
+    "rxburn_20_df = rxburn_20.execute().set_crs(\"EPSG:4326\").to_file(\"rxburn_20.geojson\")\n",
+    "rxburn_10_df = rxburn_10.execute().set_crs(\"EPSG:4326\").to_file(\"rxburn_10.geojson\")\n",
+    "rxburn_5_df = rxburn_5.execute().set_crs(\"EPSG:4326\").to_file(\"rxburn_5.geojson\")\n",
+    "rxburn_2_df = rxburn_2.execute().set_crs(\"EPSG:4326\").to_file(\"rxburn_2.geojson\")\n",
+    "\n",
+    "\n",
+    "generate_pmtiles(\"rxburn_20.geojson\",\"rxburn_10.geojson\",\"rxburn_5.geojson\",\"rxburn_2.geojson\",\"cal_rxburn_2022.pmtiles\")\n",
+    "hf_upload(\"cal_rxburn_2022.pmtiles\", \"boettiger-lab/ca-30x30\",\"dataset\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "1220c348-c68b-4475-ba0f-ef563fea7345",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "fire_20 =  (con\n",
+    "        .read_parquet(\"firep22_1.parquet\")\n",
+    "        .rename(year = \"YEAR_\")\n",
+    "        .filter(_.STATE == \"CA\", _.year != '')\n",
+    "        .cast({\"year\":\"int\"})\n",
+    "        .filter(_.year>=2003)\n",
+    "        .select(\"year\",\"geometry\")\n",
+    "        .mutate(\n",
+    "            geometry=ibis.ifelse(\n",
+    "            _.geometry.is_valid(),\n",
+    "                _.geometry,  # Keep the geometry if it's valid\n",
+    "                _.geometry.buffer(0)  # Apply buffer(0) to fix invalid geometries\n",
+    "                )\n",
+    "            )\n",
+    "        .mutate(geometry = _.geometry.convert(\"EPSG:3310\",\"EPSG:4326\"))\n",
+    "        )\n",
+    "\n",
+    "fire_10 =  (fire_20.filter(_.year>=2013))\n",
+    "fire_5 =  (fire_20.filter(_.year>=2018))\n",
+    "fire_2 =  (fire_20.filter(_.year>=2022))\n",
+    "\n",
+    "fire_20_df = fire_20.execute().set_crs(\"EPSG:4326\").to_file(\"fire_20.geojson\")\n",
+    "fire_10_df = fire_10.execute().set_crs(\"EPSG:4326\").to_file(\"fire_10.geojson\")\n",
+    "fire_5_df = fire_5.execute().set_crs(\"EPSG:4326\").to_file(\"fire_5.geojson\")\n",
+    "fire_2_df = fire_2.execute().set_crs(\"EPSG:4326\").to_file(\"fire_2.geojson\")\n",
+    "\n",
+    "\n",
+    "generate_pmtiles(\"fire_20.geojson\",\"fire_10.geojson\",\"fire_5.geojson\",\"fire_2.geojson\",\"cal_fire_2022.pmtiles\")\n",
+    "hf_upload(\"cal_fire_2022.pmtiles\", \"boettiger-lab/ca-30x30\",\"dataset\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "41ddf636-812e-4f0d-81db-64cf80cb2d4d",
+   "metadata": {},
+   "source": [
+    "# Renaming variables, adding new columns, etc"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "8eb85005-856f-4cc5-ba8d-e3efb24cdb32",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "ca = (con\n",
+    "      .read_parquet(\"https://huggingface.co/spaces/boettiger-lab/ca-30x30/resolve/main/cpad-stats.parquet\")\n",
+    "      .rename(easement = \"Easement\")\n",
+    "      .rename(acres = \"Acres\")\n",
+    "      .drop('percent_fire_20yr', 'percent_fire_5yr','percent_fire_2yr','percent_rxburn_20yr', 'percent_rxburn_5yr','percent_rxburn_2yr')\n",
+    "      .cast({\"established\":\"str\"})\n",
+    "      .mutate(easement = _.easement.substitute({\"Easement\": \"True\", \"Fee\":\"False\"}),\n",
+    "              established = _.established.substitute({\"2023\": \"pre-2024\" }),\n",
+    "              )\n",
+    "     )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "78eef2b6-5f34-49b6-937e-4744fd64cea8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "hf_upload(\"cpad-stats.parquet\", \"boettiger-lab/ca-30x30\",\"space\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "652152fd-da31-44a0-bc50-9d3aa0fe6491",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "gdf = ca.execute().set_crs(\"EPSG:4326\")\n",
+    "gdf.to_parquet(\"cpad-stats.parquet\")\n",
+    "# hf_upload(\"cpad-stats.parquet\", \"boettiger-lab/ca-30x30\",\"dataset\")\n",
+    "hf_upload(\"cpad-stats.parquet\", \"boettiger-lab/ca-30x30\",\"space\")\n",
+    "\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "80537a24-da0c-4016-9d8b-736bce30eb40",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "gdf.to_file(\"cpad-stats.geojson\")\n",
+    "generate_pmtiles(\"cpad-stats.geojson\",\"cpad-stats.pmtiles\")\n",
+    "hf_upload(\"cpad-stats.pmtiles\", \"boettiger-lab/ca-30x30\",\"dataset\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "b0a5521b-8159-495b-a9a1-b78574fe2ceb",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "hf_upload(\"cpad-stats.parquet\", \"boettiger-lab/ca-30x30-folium\",\"space\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "7727c253-813a-40e6-b73a-e973514606f3",
+   "metadata": {},
+   "source": [
+    "# Rounding acres "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "9f427c9d-6b87-4bc0-a5d7-66f16a9bec77",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# foliumap tooltip looks messy so I am rounding the acres value.\n",
+    "parquet = \"cpad-stats.parquet\"\n",
+    "ca = (con\n",
+    "      .read_parquet(parquet)\n",
+    "      .mutate(acres = _.acres.round(4)\n",
+    "              )\n",
+    "     )\n",
+    "\n",
+    "gdf = ca.execute().set_crs(\"EPSG:4326\")\n",
+    "gdf.to_parquet(\"cpad-stats.parquet\")\n",
+    "## didn't need to upload parquet since the rounding doesn't impact this?\n",
+    "hf_upload(\"cpad-stats.parquet\", \"boettiger-lab/ca-30x30\",\"dataset\")\n",
+    "# hf_upload(\"cpad-stats.parquet\", \"boettiger-lab/ca-30x30\",\"space\")\n",
+    "# hf_upload(\"cpad-stats.parquet\", \"boettiger-lab/ca-30x30-folium\",\"space\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "9d949c80-c572-4ee2-aa73-563c9ac5a649",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "gdf.to_file(\"cpad-stats.geojson\")\n",
+    "generate_pmtiles(\"cpad-stats.geojson\",\"cpad-stats.pmtiles\")\n",
+    "hf_upload(\"cpad-stats.pmtiles\", \"boettiger-lab/ca-30x30\",\"dataset\")\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.12.7"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}