County_field_boundary .ipynb

County_field_boundary .ipynb

{
  "cells": [
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import ee\nimport geemap\n\nMap = geemap.Map()\n\n# IDENTIFY FIELDS FOR PREDICTION\n#---DEFINE FUNCTIONS-------------------------------#\n# create binary layer for if crop is soybean\ndef isSoy(image):\n  return image.eq(5)\n\n# create binary layer for if field had soybean in last 10 years\ndef soyField(imageColl, countyGeom):\n  # create binary layer if soy\n  historicalCrops = imageColl.map(isSoy)\n  # sum over all years\n  soyFields = historicalCrops.sum()\n  # create mask for fields planted at least 1 year of soy over past 10 years\n  mask = soyFields.gt(0)\n  # update CDL image with soy mask and crop to county\n  soyMask = soyFields.updateMask(mask).clip(counties.geometry())\n  return(soyMask)\n\n#----DEFINE REGION OF INTEREST--------------------------------------#\n# define counties of interest\ncountyList = ['Dade']\n# load state boundary collection\ncounties = ee.FeatureCollection(\"TIGER/2018/Counties\") \\\n    .filter(ee.Filter.inList('NAME', countyList))\n#-------CREATE SOYBEAN FIELD LAYER----------------------------#\n# load cropland data layer from previous 10 years\nhistoricalCrops = ee.ImageCollection('USDA/NASS/CDL') \\\n                  .filter(ee.Filter.date('2010-01-01', '2019-12-31')) \\\n                  .select('cropland') \\\n                  .filterBounds(counties.geometry())\nsoyMap = soyField(historicalCrops, counties)\n#Map.addLayer(ndviCounty, {min: 0, max:1, palette: ['blue', 'white', 'green'], opacity: .5}, 'ndvi')\nMap.addLayer(soyMap, {'palette': [\"B997C7\"], 'opacity': .5}, 'SoyFields')\n# Export the image, specifying scale and region.\n#soyMap_Int =  ee.Image.toInt(soyMap)\n\n######################################################\n#Here below is what is added on top of Rebecca's code\n######################################################\n#/ Unsupervised classification based on training dataset using the county bounary --- for post-processing the soyMap raster layer\nseeds = ee.Algorithms.Image.Segmentation.seedGrid(36)\n#Map.addLayer(seeds.mask(seeds), {}, 'seeds')\n\n# training region is the full county \"soyMap\" image\ntraining = soyMap.sample(**{\n  'region': counties,\n  'scale': 30,\n  'numPixels': 5000\n})\n\n# train cluster on a county image\nsoyMap_clusterer = ee.Clusterer.wekaKMeans(10).train(training)\n# cluster the complete county image\nsoyMap_cluster_result = soyMap.cluster(soyMap_clusterer)\n\nMap.addLayer(soyMap_cluster_result.randomVisualizer(), {}, 'soyMap_cluster_result')\n\n# define the size of a low-pass kernel\nkernel_size = ee.Kernel.square(**{\n  'radius': 3, 'units': 'pixels', 'normalize': True\n})\n\n# Smooth the image by convolving with the low-pass-kernel\nsoyMap_cluster_result_smooth = soyMap_cluster_result.convolve(kernel_size)\nMap.addLayer(soyMap_cluster_result_smooth, {}, 'soyMap_cluster_result_smooth')\n#---------------------------------------------------------------------------------------------------\n# perform impervious layers [using NLCD C21, C22, C23 and C24] masking\n#---------------------------------------------------------------------------------------------------\nnlcd = ee.ImageCollection('USGS/NLCD') \\\n                          .filter(ee.Filter.eq('system:index', 'NLCD2016')) \\\n                          .select(0) \n\n# Map-out developed open space (C21), developed -- low (C22), medium (C23) and high (C24) intensities\n\ndef func_jnl(image):\n  return image.remap([21,22,23,24],\n                      [0, 0, 0, 0], 1)\n\nnlcdImperv = nlcd.map(func_jnl)\n\n#print(nlcd.first())\n#print(nlcdImperv.first())\nMap.addLayer(nlcdImperv, {}, 'nlcd_impervLayer')\nnlcd_test = nlcdImperv. filterBounds(counties.geometry())\n#Map.addLayer(nlcd_test, {}, 'nlcd_test')\n\ndef isImperv(image):\n  return image.eq(1)\n\n#---------------------------------------------------------------------------------------------------\n#/ Export final output as either a raster or a vector (.shp or .GeoJSON) file format\n#---------------------------------------------------------------------------------------------------\n# Convert the raster data type to integer\nsoyMap_result_smooth_Int = soyMap_cluster_result_smooth.toInt()\nMap.addLayer(soyMap_result_smooth_Int, {}, 'soyMap_result_smooth_Int')\n\n# Export as a raster format\n# Export.image.toCloudStorage({\n#   'image': soyMap_result_smooth_Int,\n#   'description': 'soyMap_result_smooth_Int',\n#   'scale': 30,\n#   #region: 'counties',\n#   'fileFormat': 'GeoTIFF',\n#   'bucket': 'gee-eval'\n\n# })\n#print(soyMap_result_smooth_Int)\n\n# Convert raster to vector\n#clusterValue = maskUrban.select('cluster')\nsoyMap_result_smooth_Int_vector = soyMap_result_smooth_Int.reduceToVectors(**{\n  'reducer': ee.Reducer.countEvery(),\n  'geometry': counties,\n  'scale': 30,\n  'maxPixels': 1e8\n})\nsoyMap_result_smooth_Int_vector2 = ee.FeatureCollection(soyMap_result_smooth_Int_vector)\nMap.addLayer(soyMap_result_smooth_Int_vector2.union(1), {},'soyMap_result_smooth_Int_vector2')\n\n# Export the vector data\n# Export.table.toCloudStorage({\n#   'collection': soyMap_result_smooth_Int_vector2.union(1),\n#   'description': 'county_soyFields_final',\n#   #fileFormat: 'SHP',\n#   'fileFormat': 'GeoJSON',\n#   'bucket': 'gee-eval'\n# })\n##################################################\n##################################################\nMap\n",
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "Map(center=[40, -100], controls=(WidgetControl(options=['position'], widget=HBox(children=(ToggleButton(value=…",
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "e9d3b8e961ac4aba947ce2a63847e3c5"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3",
      "language": "python"
    },
    "language_info": {
      "name": "python",
      "version": "3.8.2",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "varInspector": {
      "window_display": false,
      "cols": {
        "lenName": 16,
        "lenType": 16,
        "lenVar": 40
      },
      "kernels_config": {
        "python": {
          "library": "var_list.py",
          "delete_cmd_prefix": "del ",
          "delete_cmd_postfix": "",
          "varRefreshCmd": "print(var_dic_list())"
        },
        "r": {
          "library": "var_list.r",
          "delete_cmd_prefix": "rm(",
          "delete_cmd_postfix": ") ",
          "varRefreshCmd": "cat(var_dic_list()) "
        }
      },
      "types_to_exclude": [
        "module",
        "function",
        "builtin_function_or_method",
        "instance",
        "_Feature"
      ]
    },
    "hide_input": false,
    "toc": {
      "nav_menu": {},
      "number_sections": true,
      "sideBar": true,
      "skip_h1_title": true,
      "base_numbering": 1,
      "title_cell": "Table of Contents",
      "title_sidebar": "Table of Contents",
      "toc_cell": false,
      "toc_position": {},
      "toc_section_display": true,
      "toc_window_display": true
    },
    "gist": {
      "id": "eb98b1203a50ae1079bead62f655d4a7",
      "data": {
        "description": "County_field_boundary .ipynb",
        "public": true
      }
    },
    "_draft": {
      "nbviewer_url": "https://gist.github.com/eb98b1203a50ae1079bead62f655d4a7"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 4
}