Sync reproducible research release, batch 33/33

.gitattributes

@@ -69,3 +69,5 @@ data/source/legacy_simulation_archive/230214simulation/wind_simulation.xlsx filt
 data/source/original_pipeline/mydata_0224.xlsx filter=lfs diff=lfs merge=lfs -text
 external_data/ccav_20261007/Vegetation[[:space:]]samples.xlsx filter=lfs diff=lfs merge=lfs -text
 figures/recovery_20261007/recovered_rgb_mask_contact_sheet.pdf filter=lfs diff=lfs merge=lfs -text
+manuscript/Spartina_working_draft_20261007.docx filter=lfs diff=lfs merge=lfs -text
+manuscript/Spartina_working_draft_20261007.pdf filter=lfs diff=lfs merge=lfs -text

legacy/original_code/historical_yearly_match_rate.ipynb

@@ -0,0 +1,796 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "from matplotlib import pyplot as plt\n",
+    "from tqdm import tqdm"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data_pos = pd.read_csv('result_pos1229.csv',index_col = 0)\n",
+    "data_size = pd.read_csv('result_size1229.csv',index_col = 0)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>2014</th>\n",
+       "      <th>2015</th>\n",
+       "      <th>2016</th>\n",
+       "      <th>2017</th>\n",
+       "      <th>2018</th>\n",
+       "      <th>2019</th>\n",
+       "      <th>2020</th>\n",
+       "      <th>2021</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>2473</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>3121</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>9601</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>9658</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>936</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2454</th>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>3763</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2455</th>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>118</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2456</th>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>102</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2457</th>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>2261</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2458</th>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>-1</td>\n",
+       "      <td>2980</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>2459 rows × 8 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "      2014  2015  2016  2017  2018  2019  2020  2021\n",
+       "0     2473    -1    -1    -1    -1    -1    -1    -1\n",
+       "1     3121    -1    -1    -1    -1    -1    -1    -1\n",
+       "2     9601    -1    -1    -1    -1    -1    -1    -1\n",
+       "3     9658    -1    -1    -1    -1    -1    -1    -1\n",
+       "4      936    -1    -1    -1    -1    -1    -1    -1\n",
+       "...    ...   ...   ...   ...   ...   ...   ...   ...\n",
+       "2454    -1    -1    -1    -1    -1    -1    -1  3763\n",
+       "2455    -1    -1    -1    -1    -1    -1    -1   118\n",
+       "2456    -1    -1    -1    -1    -1    -1    -1   102\n",
+       "2457    -1    -1    -1    -1    -1    -1    -1  2261\n",
+       "2458    -1    -1    -1    -1    -1    -1    -1  2980\n",
+       "\n",
+       "[2459 rows x 8 columns]"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_size"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data_new = pd.DataFrame(columns = ['year','size','X','Y','match'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def match(x):\n",
+    "    if x > 0:\n",
+    "        return 1\n",
+    "    else:\n",
+    "        return 0"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|████████████████████████████████████████████████████████████████████████████| 2459/2459 [00:01<00:00, 1247.74it/s]\n"
+     ]
+    }
+   ],
+   "source": [
+    "k = 0 \n",
+    "data_new = pd.DataFrame(columns = ['year','size','X','Y','match'])\n",
+    "for i in tqdm(range(data_size.shape[0])):\n",
+    "    for j in range(data_size.shape[1]-1):\n",
+    "        if (1000 > data_size.iloc[i,j] > 0 and data_size.iloc[i,j-1] < 0) or (j == 0 and 1000>data_size.iloc[i,j]>0):\n",
+    "            data_new.loc[k] = [j+2014,data_size.iloc[i,j],data_pos.iloc[i,2*j],\n",
+    "                              data_pos.iloc[i,2*j+1],match(data_size.iloc[i,j+1])]\n",
+    "            k += 1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 43,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>year</th>\n",
+       "      <th>size</th>\n",
+       "      <th>X</th>\n",
+       "      <th>Y</th>\n",
+       "      <th>match</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>2014</td>\n",
+       "      <td>936</td>\n",
+       "      <td>609</td>\n",
+       "      <td>2470</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>2014</td>\n",
+       "      <td>228</td>\n",
+       "      <td>919</td>\n",
+       "      <td>1846</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>2014</td>\n",
+       "      <td>136</td>\n",
+       "      <td>2201</td>\n",
+       "      <td>6322</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>2014</td>\n",
+       "      <td>571</td>\n",
+       "      <td>2243</td>\n",
+       "      <td>2820</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>2014</td>\n",
+       "      <td>985</td>\n",
+       "      <td>2359</td>\n",
+       "      <td>2750</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>703</th>\n",
+       "      <td>2020</td>\n",
+       "      <td>571</td>\n",
+       "      <td>16569</td>\n",
+       "      <td>9014</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>704</th>\n",
+       "      <td>2020</td>\n",
+       "      <td>651</td>\n",
+       "      <td>16849</td>\n",
+       "      <td>14265</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>705</th>\n",
+       "      <td>2020</td>\n",
+       "      <td>173</td>\n",
+       "      <td>16881</td>\n",
+       "      <td>10898</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>706</th>\n",
+       "      <td>2020</td>\n",
+       "      <td>303</td>\n",
+       "      <td>16935</td>\n",
+       "      <td>10966</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>707</th>\n",
+       "      <td>2020</td>\n",
+       "      <td>136</td>\n",
+       "      <td>17869</td>\n",
+       "      <td>14112</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>708 rows × 5 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "     year size      X      Y match\n",
+       "0    2014  936    609   2470     0\n",
+       "1    2014  228    919   1846     0\n",
+       "2    2014  136   2201   6322     0\n",
+       "3    2014  571   2243   2820     0\n",
+       "4    2014  985   2359   2750     0\n",
+       "..    ...  ...    ...    ...   ...\n",
+       "703  2020  571  16569   9014     0\n",
+       "704  2020  651  16849  14265     0\n",
+       "705  2020  173  16881  10898     0\n",
+       "706  2020  303  16935  10966     0\n",
+       "707  2020  136  17869  14112     0\n",
+       "\n",
+       "[708 rows x 5 columns]"
+      ]
+     },
+     "execution_count": 43,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_new"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "year  match\n",
+       "2014  0        72\n",
+       "      1        52\n",
+       "2015  0        79\n",
+       "      1        44\n",
+       "2016  0        38\n",
+       "      1         7\n",
+       "2017  0        41\n",
+       "      1         7\n",
+       "2018  0        89\n",
+       "      1        18\n",
+       "2019  0        71\n",
+       "      1        43\n",
+       "2020  0        96\n",
+       "      1        51\n",
+       "Name: match, dtype: int64"
+      ]
+     },
+     "execution_count": 44,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_new.groupby('year')['match'].value_counts()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 45,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "plaque_sum = data_new.groupby('year')['size'].count().values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 46,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "plaque_count = []\n",
+    "for i in range(len(plaque_sum)):\n",
+    "    for j in range(2):\n",
+    "        plaque_count.append(plaque_sum[i])\n",
+    "new_plaque_matched = data_new.groupby('year')['match'].value_counts()/plaque_count"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 47,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[124, 124, 123, 123, 45, 45, 48, 48, 107, 107, 114, 114, 147, 147]"
+      ]
+     },
+     "execution_count": 47,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "plaque_count"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 48,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "new_plaque_matched = pd.DataFrame(new_plaque_matched)\n",
+    "new_plaque_matched.to_csv('new_plaque_matched.csv')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 49,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "new_plaque_matched['count'] = data_new.groupby('year')['match'].value_counts()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 50,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "year     222\n",
+       "size     222\n",
+       "X        222\n",
+       "Y        222\n",
+       "match    222\n",
+       "dtype: int64"
+      ]
+     },
+     "execution_count": 50,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_new[data_new['match'] == 1].count()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th>match</th>\n",
+       "      <th>count</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>year</th>\n",
+       "      <th>match</th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">2014</th>\n",
+       "      <th>0</th>\n",
+       "      <td>0.580645</td>\n",
+       "      <td>72</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.419355</td>\n",
+       "      <td>52</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">2015</th>\n",
+       "      <th>0</th>\n",
+       "      <td>0.642276</td>\n",
+       "      <td>79</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.357724</td>\n",
+       "      <td>44</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">2016</th>\n",
+       "      <th>0</th>\n",
+       "      <td>0.844444</td>\n",
+       "      <td>38</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.155556</td>\n",
+       "      <td>7</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">2017</th>\n",
+       "      <th>0</th>\n",
+       "      <td>0.854167</td>\n",
+       "      <td>41</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.145833</td>\n",
+       "      <td>7</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">2018</th>\n",
+       "      <th>0</th>\n",
+       "      <td>0.831776</td>\n",
+       "      <td>89</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.168224</td>\n",
+       "      <td>18</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">2019</th>\n",
+       "      <th>0</th>\n",
+       "      <td>0.622807</td>\n",
+       "      <td>71</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.377193</td>\n",
+       "      <td>43</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">2020</th>\n",
+       "      <th>0</th>\n",
+       "      <td>0.653061</td>\n",
+       "      <td>96</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.346939</td>\n",
+       "      <td>51</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "               match  count\n",
+       "year match                 \n",
+       "2014 0      0.580645     72\n",
+       "     1      0.419355     52\n",
+       "2015 0      0.642276     79\n",
+       "     1      0.357724     44\n",
+       "2016 0      0.844444     38\n",
+       "     1      0.155556      7\n",
+       "2017 0      0.854167     41\n",
+       "     1      0.145833      7\n",
+       "2018 0      0.831776     89\n",
+       "     1      0.168224     18\n",
+       "2019 0      0.622807     71\n",
+       "     1      0.377193     43\n",
+       "2020 0      0.653061     96\n",
+       "     1      0.346939     51"
+      ]
+     },
+     "execution_count": 51,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "new_plaque_matched"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 52,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 7/7 [00:00<00:00, 7100.39it/s]\n"
+     ]
+    }
+   ],
+   "source": [
+    "possibility = []\n",
+    "for i in tqdm(range(1,15,2)):\n",
+    "    possibility.append(new_plaque_matched.iloc[i,0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 53,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[0.41935483870967744,\n",
+       " 0.35772357723577236,\n",
+       " 0.15555555555555556,\n",
+       " 0.14583333333333334,\n",
+       " 0.16822429906542055,\n",
+       " 0.37719298245614036,\n",
+       " 0.3469387755102041]"
+      ]
+     },
+     "execution_count": 53,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "possibility"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 54,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 1.0, 'Yearly match rate')"
+      ]
+     },
+     "execution_count": 54,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.plot(possibility)\n",
+    "plt.xticks(range(7),range(2014,2021))\n",
+    "plt.title('Yearly match rate')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "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": "3.8.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}

manuscript/Spartina_working_draft_20261007.docx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0eb8f12ee49f409d6344bfc865a65080b153cb4ed710be31b9d0d811c7cf502d
+size 6933680

manuscript/Spartina_working_draft_20261007.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:434c88ae7dd7b0111d181bd37d687919a2da3e2090468cf476b2fa3fcf12394e
+size 7513561

metadata/data_sources.json

@@ -86,7 +86,7 @@
   ],
   "companion_huggingface_dataset": "https://huggingface.co/datasets/cyd0806/spartina-ai-eco-evolution-data",
   "github_repository": "https://github.com/ydchen0806/spartina-ai-eco-evolution",
-  "source_boundary": "The release includes the recovered simulation workbooks, patch CSV and pinned NASA POWER/GBIF response files. The raw UAV orthomosaics, segmentation masks, original weather/tide files and original simulation source code were not found in the workspace scan.",
+  "source_boundary": "The public package contains source tables, code and derived audits. A separate local recovery on 2026-10-06 restored 15 full image/mask pairs, an additional 2013 image and annotation deliveries. The 38.65 GB restored archive is not bundled in this release. See docs/RECOVERY_VALIDATION_2026-10-07.md. Checkpoint, original training RGB photographs, original split and scenario exporter remain unrecovered.",
   "external_context_files": [
     {
       "path": "external_data/nasa_power_zhangjiang_2014_2021.json",
@@ -127,7 +127,7 @@
   "release_root": "nature_manuscript/public_release",
   "upstream_code_repository": "https://github.com/ydchen0806/ai4FastEvolution",
   "upstream_code_commit": "171c9f516799b8bc90aa98b9274730ead7ae740e",
-  "upstream_recovery_note": "The release preserves selected upstream scripts and source matrices; original UAV TIFF/JPEG pixels, segmentation masks and scenario-export code were absent from upstream Git history and local workspace.",
+  "upstream_recovery_note": "The upstream snapshot at 171c9f516799b8bc90aa98b9274730ead7ae740e was inspected from a shallow clone; this is not an exhaustive Git history search. Separate local media recovery supersedes the earlier missing-image status.",
   "upstream_recovered_files": [
     {
       "path": "data/source/original_pipeline/xiamen.xlsx",
@@ -726,5 +726,46 @@
       "sha256": "934c367938bc12588d834624dc19d76467130b8ee875faa747b4f58be644e465",
       "bytes": 30984
     }
-  ]
+  ],
+  "local_recovery_20261006": {
+    "local_root": "raw_inputs/recovered_local_20261006",
+    "files": 1001,
+    "bytes": 38654767408,
+    "independent_hash_audit": "data/derived/recovery_20261007/fresh_hash_verification.json",
+    "hash_failures": 0,
+    "raw_archive_bundled_in_release": false,
+    "report": "docs/RECOVERY_VALIDATION_2026-10-07.md"
+  },
+  "trajectory_audit_20261007": {
+    "report": "docs/TRAJECTORY_ENDPOINT_AUDIT_2026-10-07.md",
+    "derived_directory": "data/derived/trajectory_audit_20261007",
+    "review_gallery": "figures/transition_review_20261007/index.html",
+    "review_cases": 61,
+    "expert_labels_completed": 0,
+    "source_archive_modified": false,
+    "stored_endpoint_disagreements": 188,
+    "biological_events_independently_validated": false
+  },
+  "strict_benchmark_20261007": {
+    "config": "data/derived/strict_benchmark_20261007/benchmark_config.json",
+    "status": "Completed starting-geometry benchmark on archived labels; no independent biological validation"
+  },
+  "annotation_pilot_20261007": {
+    "protocol": "data/derived/validation_pilot_20261007/sampling_protocol.json",
+    "local_crop_root": "nature_manuscript/validation_pilot_20261007",
+    "crops_in_public_release": false,
+    "expert_labels": 0
+  },
+  "image_center_climate_20261007": {
+    "provider": "NASA POWER",
+    "provenance": "external_data/image_center_climate_20261007/climate_context_provenance.json",
+    "purpose": "Gridded context; not tide or local terrain measurement"
+  },
+  "working_manuscript": "docs/MANUSCRIPT_WORKING_DRAFT.md",
+  "literature_verification_20261007": {
+    "provider": "Crossref REST API",
+    "records": 8,
+    "scope": "DOI metadata and available abstracts; not full-text review",
+    "index": "external_data/literature_verification_20261007/verified_metadata.json"
+  }
 }

requirements-browser.txt

@@ -0,0 +1,2 @@
+# Optional PDF rendering and annotation interface browser checks
+playwright

requirements.txt

@@ -5,3 +5,8 @@ matplotlib
 requests
 openpyxl
 scikit-learn
+rasterio
+opencv-python-headless
+Pillow
+python-docx
+pyproj

tables/original_pipeline_audit.json

@@ -3,8 +3,8 @@
   "upstream_commit": "171c9f516799b8bc90aa98b9274730ead7ae740e",
   "raw_size_position_rows": 3237,
   "raw_image_dates": [
-    "2014-08-01",
-    "2015-08-01",
+    "2014-08",
+    "2015-08",
     "2015-10-17",
     "2016-09-07",
     "2016-10-25",
@@ -27,6 +27,6 @@
   "weather_year_min": 1980,
   "weather_year_max": 2021,
   "image_pixel_files_found_in_release": false,
-  "image_pixel_boundary": "The matrices reference ./mask/*.tif and notebooks reference E:/remote_data/numvis/*.jpg, but those image pixels are absent from the upstream repository and local workspace scan.",
-  "model_checkpoint_boundary": "The upstream repository contains AutoGluon predictor artifacts, but no UAV segmentation checkpoint or training masks."
+  "image_pixel_boundary": "Raw imagery and masks are outside this release snapshot. They were recovered locally on 2026-10-06; see docs/RECOVERY_VALIDATION_2026-10-07.md and data/derived/recovery_20261007/.",
+  "model_checkpoint_boundary": "The upstream repository contains AutoGluon predictor artifacts, but no recovered UAV segmentation checkpoint or original training split. Polygon annotations and 79 delivered PNG masks have since been recovered locally."
 }

tables/original_pipeline_image_inventory.csv

@@ -1,16 +1,16 @@
-image_column,date,year,n_nonmissing_size,n_nonmissing_x,n_nonmissing_y
-./mask/201408ddyw.tif,2014-08-01,2014,568,568,568
-./mask/201508ddyw.tif,2015-08-01,2015,845,845,845
-./mask/20151017sp.tif,2015-10-17,2015,364,364,364
-./mask/20160907.tif,2016-09-07,2016,365,365,365
-./mask/20161025.tif,2016-10-25,2016,449,449,449
-./mask/20170714.tif,2017-07-14,2017,327,327,327
-./mask/20170728.tif,2017-07-28,2017,328,328,328
-./mask/20171002.tif,2017-10-02,2017,211,211,211
-./mask/20180729.tif,2018-07-29,2018,326,326,326
-./mask/20180828.tif,2018-08-28,2018,414,414,414
-./mask/20180926.tif,2018-09-26,2018,340,340,340
-./mask/20190828.tif,2019-08-28,2019,296,296,296
-./mask/20190923.tif,2019-09-23,2019,307,307,307
-./mask/20200727.tif,2020-07-27,2020,319,319,319
-./mask/20210717.tif,2021-07-17,2021,287,287,287
+image_column,date,date_precision,year,n_nonmissing_size,n_nonmissing_x,n_nonmissing_y
+./mask/201408ddyw.tif,2014-08,month,2014,568,568,568
+./mask/201508ddyw.tif,2015-08,month,2015,845,845,845
+./mask/20151017sp.tif,2015-10-17,day,2015,364,364,364
+./mask/20160907.tif,2016-09-07,day,2016,365,365,365
+./mask/20161025.tif,2016-10-25,day,2016,449,449,449
+./mask/20170714.tif,2017-07-14,day,2017,327,327,327
+./mask/20170728.tif,2017-07-28,day,2017,328,328,328
+./mask/20171002.tif,2017-10-02,day,2017,211,211,211
+./mask/20180729.tif,2018-07-29,day,2018,326,326,326
+./mask/20180828.tif,2018-08-28,day,2018,414,414,414
+./mask/20180926.tif,2018-09-26,day,2018,340,340,340
+./mask/20190828.tif,2019-08-28,day,2019,296,296,296
+./mask/20190923.tif,2019-09-23,day,2019,307,307,307
+./mask/20200727.tif,2020-07-27,day,2020,319,319,319
+./mask/20210717.tif,2021-07-17,day,2021,287,287,287

templates/annotation_pilot.html

@@ -0,0 +1,48 @@
+<!doctype html>
+<html lang="zh"><meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1">
+<title>互花米草 · 独立影像标注试点</title>
+<style>
+body{font:16px system-ui,sans-serif;color:#18333d;background:#f3f6f5;margin:24px;max-width:1280px}h1{font-size:26px}p{line-height:1.6}button,select,input{font:inherit;padding:7px;margin:3px;border:1px solid #aebfbc;border-radius:5px}button{cursor:pointer;background:white}button.primary{background:#116d68;color:white}main{display:grid;grid-template-columns:minmax(300px,760px) 1fr;gap:24px}canvas{max-width:100%;height:auto;border:1px solid #657d77;background:#ddd;cursor:crosshair;touch-action:none}aside{background:white;padding:18px;border-radius:8px}.status{color:#116d68}textarea{width:95%;height:80px}#samples{max-width:100%}small{line-height:1.5;display:block}a{color:#116d68}@media(max-width:850px){main{grid-template-columns:1fr}body{margin:12px}}
+</style>
+<h1>互花米草 · 独立影像标注试点</h1>
+<p>先在原图上判断可观测性，再描绘目标植被边界。无法确认物种或看不清边界时标为“不确定”。此页面不显示历史 mask。样本没有完成专家标注，所有结果需经复核。</p>
+<div><button id="prev">上一张</button><select id="samples"></select><button id="next">下一张</button><span id="progress"></span></div>
+<main><section><canvas id="canvas" width="512" height="512"></canvas><p id="status" class="status"></p></section>
+<aside>
+<label>标注人 <input id="reviewer" placeholder="姓名或稳定编号"></label>
+<p><label>区域类别 <select id="category"><option value="cordgrass_candidate">互花米草候选</option><option value="other_vegetation">其他植被</option><option value="uncertain">不确定区域</option><option value="unobservable">不可观测区域</option></select></label></p>
+<small>沿边界单击添加顶点；至少3点后完成多边形。坐标按原图像素保存。不要把阴影、潮水或拼接缺口自动标为无植被。</small>
+<p><button id="close" class="primary">完成多边形</button><button id="undo">撤销顶点</button><button id="cancel">取消当前轮廓</button><button id="remove">删除最后多边形</button></p>
+<label>整图状态 <select id="visibility"><option value="unreviewed">尚未复核</option><option value="observable">可观测</option><option value="partly_observable">部分可观测</option><option value="unobservable">不可观测</option><option value="uncertain">无法判断</option></select></label>
+<p><label><input type="checkbox" id="completed">本图已检查完毕（允许无目标）</label></p>
+<label>说明<textarea id="notes" placeholder="边界、潮位、遮挡或物种判断的不确定性"></textarea></label>
+<p><button id="export" class="primary">导出当前标注 JSON</button></p>
+<label>恢复已导出的标注 <input id="import" type="file" accept="application/json,.json"></label>
+<small>页面尝试保存在当前浏览器本地；离开前请导出 JSON。浏览器本地存储不可用时仍可导出。正式测试标签应由独立保管者保存，不用于训练或模型选择。标注坐标不是经过验证的地理坐标。</small>
+</aside></main>
+<script>
+'use strict';
+const samples=__SAMPLES_JSON__;
+const key='spartina-pilot-20261008-v1';
+const el=id=>document.getElementById(id);
+const canvas=el('canvas'),ctx=canvas.getContext('2d');
+const colors={cordgrass_candidate:'#ed008b',other_vegetation:'#00b575',uncertain:'#ffad00',unobservable:'#5158d8'};
+let records={},index=0,points=[],loaded=false,im=new Image(),storageWarning='';
+try{const saved=JSON.parse(localStorage.getItem(key)||'{}');records=saved.records||{};el('reviewer').value=saved.reviewer||'';}catch(e){storageWarning='浏览器本地存储不可用，请及时导出。';}
+function current(){const s=samples[index];if(!records[s.case_id])records[s.case_id]={case_id:s.case_id,width:s.width,height:s.height,polygons:[],visibility:'unreviewed',completed:false,notes:'',reviewer:''};return records[s.case_id];}
+function save(){const r=current();r.visibility=el('visibility').value;r.completed=el('completed').checked;r.notes=el('notes').value;r.reviewer=el('reviewer').value;r.updated_at=new Date().toISOString();try{localStorage.setItem(key,JSON.stringify({records,reviewer:el('reviewer').value}));}catch(e){storageWarning='本地保存失败，请立即导出 JSON。';}progress();}
+function progress(){el('progress').textContent=`${index+1} / ${samples.length}　已完成 ${Object.values(records).filter(r=>r.completed).length}`;}
+function polygon(pts,color,fill){if(!pts.length)return;ctx.beginPath();ctx.moveTo(...pts[0]);for(const p of pts.slice(1))ctx.lineTo(...p);if(fill)ctx.closePath();ctx.strokeStyle=color;ctx.lineWidth=1.5;ctx.stroke();if(fill){ctx.globalAlpha=.16;ctx.fillStyle=color;ctx.fill();ctx.globalAlpha=1;}for(const p of pts){ctx.fillStyle=color;ctx.fillRect(p[0]-1.5,p[1]-1.5,3,3);}}
+function draw(){ctx.clearRect(0,0,canvas.width,canvas.height);if(loaded)ctx.drawImage(im,0,0);for(const p of current().polygons)polygon(p.points,colors[p.category]||'#ffad00',true);polygon(points,colors[el('category').value],false);el('status').textContent=`${current().polygons.length} 个多边形；当前轮廓 ${points.length} 个顶点。 ${storageWarning}`;progress();}
+function show(next){if(points.length){el('status').textContent='请先完成或取消当前轮廓，再切换图片。';return;}index=Math.max(0,Math.min(samples.length-1,next));el('samples').value=index;const s=samples[index],r=current();canvas.width=s.width;canvas.height=s.height;el('visibility').value=r.visibility;el('completed').checked=r.completed;el('notes').value=r.notes;loaded=false;im=new Image();const active=im;im.onload=()=>{if(im===active){loaded=true;draw();}};im.onerror=()=>{el('status').textContent='图片读取失败，请保留 images 文件夹与本页面的相对位置。';};im.src=s.image_path;draw();}
+samples.forEach((s,i)=>{const o=document.createElement('option');o.value=i;o.textContent=s.case_id;el('samples').appendChild(o);});
+canvas.addEventListener('click',e=>{if(!loaded)return;const b=canvas.getBoundingClientRect();points.push([Math.max(0,Math.min(canvas.width,(e.clientX-b.left)*canvas.width/b.width)),Math.max(0,Math.min(canvas.height,(e.clientY-b.top)*canvas.height/b.height))]);draw();});
+el('close').onclick=()=>{if(points.length<3){el('status').textContent='至少需要3个顶点。';return;}current().polygons.push({category:el('category').value,points:points.map(p=>p.map(v=>Math.round(v*100)/100))});points=[];save();draw();};
+el('undo').onclick=()=>{points.pop();draw();};el('cancel').onclick=()=>{points=[];draw();};el('remove').onclick=()=>{current().polygons.pop();save();draw();};
+el('next').onclick=()=>{save();show(index+1);};el('prev').onclick=()=>{save();show(index-1);};el('samples').onchange=e=>{const next=Number(e.target.value);save();show(next);el('samples').value=index;};
+for(const id of ['visibility','completed','notes','reviewer'])el(id).addEventListener('change',save);
+el('category').onchange=draw;
+el('export').onclick=()=>{save();const payload={schema:'spartina-annotation-pilot-v1',protocol_seed:20261008,exported_at:new Date().toISOString(),reviewer:el('reviewer').value,coordinate_system:'original crop pixel coordinates',records:Object.values(records)};const url=URL.createObjectURL(new Blob([JSON.stringify(payload,null,2)],{type:'application/json'}));const a=document.createElement('a');a.href=url;a.download='spartina_pilot_annotations.json';a.click();setTimeout(()=>URL.revokeObjectURL(url),1000);};
+el('import').onchange=async e=>{const f=e.target.files[0];if(!f)return;try{const d=JSON.parse(await f.text());if(d.schema!=='spartina-annotation-pilot-v1'||!Array.isArray(d.records))throw Error('标注格式不匹配');const valid=new Map(samples.map(s=>[s.case_id,s]));for(const r of d.records){const s=valid.get(r.case_id);if(!s||s.width!==r.width||s.height!==r.height||!Array.isArray(r.polygons))throw Error('样本或尺寸不匹配');for(const poly of r.polygons){if(!(poly.category in colors)||!Array.isArray(poly.points)||poly.points.length<3||poly.points.some(p=>!Array.isArray(p)||p.length!==2||p.some(v=>!Number.isFinite(v))||p[0]<0||p[0]>s.width||p[1]<0||p[1]>s.height))throw Error('多边形坐标无效');}}if(!confirm('导入将替换同编号图片的当前标注，继续？'))return;for(const r of d.records)records[r.case_id]=r;el('reviewer').value=d.reviewer||'';points=[];show(index);save();}catch(err){el('status').textContent='导入失败：'+err.message;}};
+show(0);
+</script></html>