Last active
June 24, 2019 13:58
-
-
Save evertrol/e76d3c4b05939661f819619d6003c173 to your computer and use it in GitHub Desktop.
Rhine temperature response
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: base | |
| channels: | |
| - conda-forge | |
| - defaults | |
| dependencies: | |
| - iris=2.2.0=py36_1003 | |
| - iris-sample-data=2.1.0=py_0 | |
| - cftime=1.0.3.4=py36hd352d35_1001 | |
| - numpy=1.16.2=py36_blas_openblash1522bff_0 | |
| - pandas=0.23.4=py36h637b7d7_1000 | |
| - matplotlib=3.0.3=py36_1 | |
| - matplotlib-base=3.0.3=py36h5f35d83_1 |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| { | |
| "cells": [ | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "# Local temperature response\n", | |
| "\n", | |
| "*The local temperature response per GCM/RCP combination in the 2071 -- 2100 timespan, relative to the 1976 -- 2005 timespan, as function of the global temperature response over the same periods.*\n", | |
| "\n", | |
| "This notebook attempts to recreate Figure 2 from Van den Hurk et al.[<sup>1</sup>](#fn1), also used in the [KNMI '14 climate scenarios](http://www.climatescenarios.nl/).\n", | |
| "\n", | |
| "The data used is CMIP 5 data, which is locally available where this notebook was developed.\n", | |
| "\n", | |
| "Iris is our main analysis toolkit, with a few standard libraries such as numpy, pandas and matplotlib.\n", | |
| "\n", | |
| "There are a few caveats to note in the data analysis, mainly done for simplicity:\n", | |
| "- the data is not gridded to a common grid. It is assumed that either all data grids are the same, or there is negligible difference between the grids. It also means that the area of interest, the Rhine basin, will not exactly match the data grids. Again, this should result in a neglible difference.\n", | |
| "- the various available subsets for models are averaged. \n", | |
| "- For some, we are unable to perform this averaging, notable for EC-EARTH, MIROC-ESM-CHEM and HadGEM2-ES. These datasets are left out in further analysis.\n", | |
| "\n", | |
| "Our final results are similar to, but do not exactly match, those by Van den Hurk et al. In paricular, the coefficient of determination, R<sup>2</sup> is smaller for each season. (We also note one model with negative local relative temperature. This is the FIO-ESM model, not used in Van den Hurk et al). We have made several assumptions in the analysis procedure below, which may not precisely correspond to those used by Van den Hurk et al., hence this can explain the differences. Also, our final dataset includes more models and datasets overall than used by Van den Hurk (compare our table near the end with their Table 4 in the Appendix).\n", | |
| "\n", | |
| "A practical note: we have not always included the output of a cell, by putting a semi-colon at the end of the last cell line, as this would become a rather lengthy output and clog up the notebook. It makes it, however, easier for the (interactive) reader to simply remove the semi-colon and see the results.\n", | |
| "\n", | |
| "<span id=\"#fn1\">[1] Van den Hurk, B., Van Oldenborth, G., Lenderink, G., Hazeleger, W., Haarsma, R., de Vries, H., 2014: Drivers of mean climate change around the Netherlands derived from CMIP5. Clim Dyn., 42, 1683--1697. [doi:10.1007/s00382-013-1707-y](https://link.springer.com/article/10.1007/s00382-013-1707-y).</span>" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## Basic imports and settings" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 1, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "import warnings\n", | |
| "from copy import deepcopy\n", | |
| "from collections import defaultdict\n", | |
| "from pathlib import Path\n", | |
| "from collections.abc import Iterable\n", | |
| "\n", | |
| "import numpy as np\n", | |
| "import matplotlib.pyplot as plt\n", | |
| "import pandas as pd\n", | |
| "\n", | |
| "from cftime import datetime, DatetimeNoLeap, DatetimeJulian, Datetime360Day\n", | |
| "import iris\n", | |
| "import iris.coord_categorisation\n", | |
| "from iris.util import unify_time_units\n", | |
| "from iris.experimental.equalise_cubes import equalise_attributes\n", | |
| "\n", | |
| "%matplotlib inline" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 2, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "cmip5dir = Path(\"_data/cmip5\")\n", | |
| "var = \"tas\"\n", | |
| "seasons = ['djf', 'mam', 'jja', 'son']\n", | |
| "coords = {\n", | |
| " 'rhinebasin': {'w': 6, 'e': 9, 'n': 52, 's': 47}\n", | |
| "}" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "The RCPs and models are derived dynamically, directly from the available subdirectories in the file system." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 3, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "data": { | |
| "text/plain": [ | |
| "[26, 45, 60, 85]" | |
| ] | |
| }, | |
| "execution_count": 3, | |
| "metadata": {}, | |
| "output_type": "execute_result" | |
| } | |
| ], | |
| "source": [ | |
| "rcps = sorted([int(path.stem[3:]) for path in cmip5dir.glob(\"rcp*\")])\n", | |
| "rcps" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 4, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "modelnames = sorted({path.stem for path in cmip5dir.glob(f\"rcp*/Amon/{var}/*\")}, key=lambda s: s.casefold())\n", | |
| "modelnames ;" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## Set the area and time constraints\n", | |
| "\n", | |
| "The area constraint is simply based on the \"rectangular\" grid defined above.\n", | |
| "\n", | |
| "The time constraint is somewhat more problematic: not all datasets use the same style calendar. Therefore, there are multiple variations for the start and end of the timespan set, and a simple if-else chain tests which variation is appropriate. Since this test all happens inside the lambda function, it will slow down things a bit (since this check would happen for every time stamp); to improve the speed, one could define multiple constraints, one for each calendar style, then check the calendar style of the input data beforehand, and apply the appropriate time constraint to the dataset." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 5, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "data": { | |
| "text/plain": [ | |
| "Constraint(coord_values={'longitude': <function <lambda> at 0x7f39d8053510>, 'latitude': <function <lambda> at 0x7f39d8053598>})" | |
| ] | |
| }, | |
| "execution_count": 5, | |
| "metadata": {}, | |
| "output_type": "execute_result" | |
| } | |
| ], | |
| "source": [ | |
| "coord = coords['rhinebasin']\n", | |
| "rhine = iris.Constraint(\n", | |
| " longitude=lambda c: coord['w'] <= c <= coord['e'],\n", | |
| " latitude=lambda c: coord['s'] <= c <= coord['n'])\n", | |
| "rhine" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 6, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "start1 = DatetimeNoLeap(2071, 1, 1)\n", | |
| "stop1 = DatetimeNoLeap(2099, 12, 31)\n", | |
| "start2 = DatetimeJulian(2071, 1, 1)\n", | |
| "stop2 = DatetimeJulian(2099, 12, 31)\n", | |
| "start3 = Datetime360Day(2071, 1, 1)\n", | |
| "stop3 = Datetime360Day(2099, 12, 30)\n", | |
| "start4 = datetime(2071, 1, 1)\n", | |
| "stop4 = datetime(2100, 12, 31)\n", | |
| "timespan = iris.Constraint(time=lambda cell: \n", | |
| " start1 <= cell.point <= stop1 if isinstance(cell.point, DatetimeNoLeap) \n", | |
| " else start2 <= cell.point <= stop2 if isinstance(cell.point, DatetimeJulian)\n", | |
| " else start3 <= cell.point <= stop3 if isinstance(cell.point, Datetime360Day)\n", | |
| " else start4 <= cell.point <= stop4)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## Load and extract the data\n", | |
| "\n", | |
| "This is where all the actual work happens; everything else is just bookkeeping.\n", | |
| "\n", | |
| "We create a function for this, since we'll be doing this for all combinations of RCPs and models; and, in fact, for the historical values as well.\n", | |
| "\n", | |
| "The dataset is loaded, or possibly multiple datasets, for cases where the data is spread across multiple files (this spread over multiple files is often across time, but we'll ignore any time limits in the dataset initially, and only bother with that when we extract the relevant timespan. Thanks to Iris's lazy loading, this is still relatively cheap, since it avoids loading the full dataset (just the time dimension dataset)).\n", | |
| "\n", | |
| "Two Iris utilities functions are used, `equalise_attributes` and `unify_time_units`, which equalise attributes such as creation timestamps (which are unimportant for our actual data), so that Iris can properly concatenate all the cubes. Iris does the appropriate thing, concatenating cubes along the time axes (by noting the longitude and latitude coordinates are the same). Note that in a lot of cases, there will simply be one cube loaded anyway, and concatenation is not necessary: we can just use that cube.\n", | |
| "\n", | |
| "We then extract the relevant area and timespan. Multiple areas can be provided (in our case, the Rhine basin area and the global world), and we loop over these areas, performing the extraction on copies of the original data (in our specific case, we could first extract the global area, and then the local area, which avoids a copy, but it's rarely guaranteed one area lies within another. This wouldn't work if, say, Australia was included in the list of areas). Note that \"global\" is defined by having an area of `None`. Any other areas and timespan are given by Iris constraints, defined above.\n", | |
| "\n", | |
| "Once we have our extract cube, we add seasons, over which we'll aggregate and calculate the mean seasonal value over the timespan.\n", | |
| "\n", | |
| "We also average the area into a single point, with appropriately weighting for the curved Earth.\n", | |
| "\n", | |
| "The output for a single area is then a single cube, with 4 datapoints: one value for each season.\n", | |
| "\n", | |
| "The returned value is these 4-datapoint cubes appended into a list, one per area. We'll need to manually keep track of which list index corresponds to which area.\n", | |
| "\n", | |
| "For datasets that don't cover the relevant timespan (it's assumed all datasets cover the globe, and thus all areas of interest), `None` is returned." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 7, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "def load_cube(paths, timespan, areas=None):\n", | |
| " if not isinstance(areas, Iterable) or isinstance(areas, (str, bytes)):\n", | |
| " areas = [areas]\n", | |
| " cubes = iris.load([str(path) for path in paths], constraints='air_temperature')\n", | |
| " equalise_attributes(cubes)\n", | |
| " unify_time_units(cubes)\n", | |
| " try:\n", | |
| " cube = cubes.concatenate_cube()\n", | |
| " except iris.exceptions.ConcatenateError as exc:\n", | |
| " warnings.warn(str(exc) + \"\\nfor: \" + str(paths))\n", | |
| " cube = cubes[0] # iris.load always returns a cubelist, so just take the first element\n", | |
| " \n", | |
| " results = []\n", | |
| " for area in areas:\n", | |
| " excube = cube.copy()\n", | |
| " if area:\n", | |
| " excube = excube.extract(area)\n", | |
| " excube = excube.extract(timespan)\n", | |
| " if excube is None:\n", | |
| " results.append(None)\n", | |
| " continue\n", | |
| " \n", | |
| " iris.coord_categorisation.add_season(excube, 'time', name='season')\n", | |
| " \n", | |
| " weights = iris.analysis.cartography.area_weights(excube)\n", | |
| " excube_av = excube.collapsed(['latitude', 'longitude'], iris.analysis.MEAN, weights=weights)\n", | |
| " excube_season = excube_av.aggregated_by('season', iris.analysis.MEAN)\n", | |
| " results.append(excube_season)\n", | |
| "\n", | |
| " return results" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 25, | |
| "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>RCP 2.6</th>\n", | |
| " <th>RCP 4.5</th>\n", | |
| " <th>RCP 6.0</th>\n", | |
| " <th>RCP 8.5</th>\n", | |
| " <th>Name</th>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>Label</th>\n", | |
| " <th></th>\n", | |
| " <th></th>\n", | |
| " <th></th>\n", | |
| " <th></th>\n", | |
| " <th></th>\n", | |
| " </tr>\n", | |
| " </thead>\n", | |
| " <tbody>\n", | |
| " <tr>\n", | |
| " <th>0</th>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>ACCESS1-0</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>1</th>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>ACCESS1-3</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>2</th>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>bcc-csm1-1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>3</th>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>bcc-csm1-1-m</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>4</th>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>BNU-ESM</td>\n", | |
| " </tr>\n", | |
| " </tbody>\n", | |
| "</table>\n", | |
| "</div>" | |
| ], | |
| "text/plain": [ | |
| " RCP 2.6 RCP 4.5 RCP 6.0 RCP 8.5 Name\n", | |
| "Label \n", | |
| "0 0 0 0 0 ACCESS1-0\n", | |
| "1 0 0 0 0 ACCESS1-3\n", | |
| "2 0 0 0 0 bcc-csm1-1\n", | |
| "3 0 0 0 0 bcc-csm1-1-m\n", | |
| "4 0 0 0 0 BNU-ESM" | |
| ] | |
| }, | |
| "execution_count": 25, | |
| "metadata": {}, | |
| "output_type": "execute_result" | |
| } | |
| ], | |
| "source": [ | |
| "columns = {f\"RCP {rcp/10:0.1f}\": 0 for rcp in rcps}\n", | |
| "columns['Name'] = modelnames\n", | |
| "df = pd.DataFrame(columns)\n", | |
| "df.index.rename('Label', inplace=True)\n", | |
| "df.head()" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## Loop over all models\n", | |
| "\n", | |
| "This loops over all possible models, across all RCP variations and across all model subsets.\n", | |
| "\n", | |
| "Results which do not contain the relevant timespan (and thus include `None` instead of a cube, see above) are discarded.\n", | |
| "Models that do not contain any result (in any of their subsets) are completely removed.\n", | |
| "\n", | |
| "The data structure the results are stored in is a nested dictionary. This is perhaps not the most directly insightful, but since not every model is available for every RCP value, and since the number of subsets varies per model, this is the most flexible way to store our values.\n", | |
| "\n", | |
| "The levels of the dictionary are \"rcp\", then \"model\", then the subset, then a list of results per area, then each list item being a cube with 4 seasonal-averaged values. For example, `temp[85]['ACCESS1-0']['1i1p1'][1][2]` would get the summer temperature (`[2]`: the third element of the cube, since the summer is the third season) of the Rhine basin, for the subset 1i1p1 for model ACCESS1-0 for RCP 8.5.\n", | |
| "\n", | |
| "To show progress, we print every RCP iteration, and optionally every model iteration." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 9, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "== RCP: 26 ==\n", | |
| "== RCP: 45 ==\n", | |
| "== RCP: 60 ==\n", | |
| "== RCP: 85 ==\n", | |
| "CPU times: user 12min 53s, sys: 52.4 s, total: 13min 45s\n", | |
| "Wall time: 40min 8s\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "%%time\n", | |
| "# This takes about 40 minutes\n", | |
| "\n", | |
| "temp = {}\n", | |
| "for rcp in rcps:\n", | |
| " print(\"== RCP:\", rcp, \"==\")\n", | |
| " temp[rcp] = {}\n", | |
| " subdir = cmip5dir / f\"rcp{rcp}/Amon/{var}\"\n", | |
| " for model in modelnames:\n", | |
| " temp[rcp][model] = {}\n", | |
| " modeldir = subdir / model\n", | |
| " for rdir in modeldir.glob(\"r*\"):\n", | |
| " paths = list(rdir.glob(\"*.nc\"))\n", | |
| " with warnings.catch_warnings():\n", | |
| " warnings.simplefilter(\"ignore\", UserWarning)\n", | |
| " results = load_cube(paths, timespan, areas=[None, rhine])\n", | |
| " # Skip cubes that don't contain the appropriate timespan for any of the regions\n", | |
| " if any(result is None for result in results):\n", | |
| " continue\n", | |
| " temp[rcp][model][rdir.stem] = results\n", | |
| " # Remove cubes that didn't have any appropriate data\n", | |
| " nvalues = len(temp[rcp][model].values())\n", | |
| " if nvalues == 0:\n", | |
| " del temp[rcp][model]\n", | |
| " #print(model, \":\", nvalues, \"value(s)\")" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## Read the historical data\n", | |
| "\n", | |
| "Reading the historical data is very similar to the above code. We first set a new time constraint: 1976 until 2005." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 10, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "start1 = DatetimeNoLeap(1976, 1, 1)\n", | |
| "stop1 = DatetimeNoLeap(2005, 12, 31)\n", | |
| "start2 = DatetimeJulian(1976, 1, 1)\n", | |
| "stop2 = DatetimeJulian(2005, 12, 31)\n", | |
| "start3 = Datetime360Day(1976, 1, 1)\n", | |
| "stop3 = Datetime360Day(2005, 12, 30)\n", | |
| "start4 = datetime(1976, 1, 1)\n", | |
| "stop4 = datetime(2005, 12, 31)\n", | |
| "timespan = iris.Constraint(time=lambda cell: \n", | |
| " start1 <= cell.point <= stop1 if isinstance(cell.point, DatetimeNoLeap) \n", | |
| " else start2 <= cell.point <= stop2 if isinstance(cell.point, DatetimeJulian)\n", | |
| " else start3 <= cell.point <= stop3 if isinstance(cell.point, Datetime360Day)\n", | |
| " else start4 <= cell.point <= stop4)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "We don't need to iterate over different RCPs. Other than that, we can use the same code, just with one iteration level less." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 11, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "ACCESS1-0 : 3 value(s)\n", | |
| "ACCESS1-3 : 3 value(s)\n", | |
| "bcc-csm1-1 : 3 value(s)\n", | |
| "bcc-csm1-1-m : 3 value(s)\n", | |
| "BNU-ESM : 1 value(s)\n", | |
| "CanCM4 : 10 value(s)\n", | |
| "CanESM2 : 5 value(s)\n", | |
| "CCSM4 : 8 value(s)\n", | |
| "CESM1-BGC : 1 value(s)\n", | |
| "CESM1-CAM5 : 3 value(s)\n", | |
| "CESM1-CAM5-1-FV2 : 4 value(s)\n", | |
| "CESM1-WACCM : 7 value(s)\n", | |
| "CMCC-CESM : 1 value(s)\n", | |
| "CMCC-CM : 1 value(s)\n", | |
| "CMCC-CMS : 1 value(s)\n", | |
| "CNRM-CM5 : 10 value(s)\n", | |
| "CSIRO-Mk3-6-0 : 10 value(s)\n", | |
| "CSIRO-Mk3L-1-2 : 3 value(s)\n", | |
| "EC-EARTH : 7 value(s)\n", | |
| "FGOALS-g2 : 5 value(s)\n", | |
| "FGOALS-s2 : 2 value(s)\n", | |
| "FIO-ESM : 3 value(s)\n", | |
| "GFDL-CM2p1 : 10 value(s)\n", | |
| "GFDL-CM3 : 5 value(s)\n", | |
| "GFDL-ESM2G : 1 value(s)\n", | |
| "GFDL-ESM2M : 1 value(s)\n", | |
| "GISS-E2-H : 18 value(s)\n", | |
| "GISS-E2-H-CC : 1 value(s)\n", | |
| "GISS-E2-R : 26 value(s)\n", | |
| "GISS-E2-R-CC : 1 value(s)\n", | |
| "HadCM3 : 10 value(s)\n", | |
| "HadGEM2-AO : 1 value(s)\n", | |
| "HadGEM2-CC : 3 value(s)\n", | |
| "HadGEM2-ES : 5 value(s)\n", | |
| "inmcm4 : 1 value(s)\n", | |
| "IPSL-CM5A-LR : 6 value(s)\n", | |
| "IPSL-CM5A-MR : 3 value(s)\n", | |
| "IPSL-CM5B-LR : 1 value(s)\n", | |
| "MIROC-ESM : 3 value(s)\n", | |
| "MIROC-ESM-CHEM : 1 value(s)\n", | |
| "MIROC4h : 3 value(s)\n", | |
| "MIROC5 : 5 value(s)\n", | |
| "MPI-ESM-LR : 3 value(s)\n", | |
| "MPI-ESM-MR : 3 value(s)\n", | |
| "MRI-CGCM3 : 5 value(s)\n", | |
| "MRI-ESM1 : 1 value(s)\n", | |
| "NorESM1-M : 3 value(s)\n", | |
| "NorESM1-ME : 2 value(s)\n", | |
| "CPU times: user 8min 32s, sys: 21.5 s, total: 8min 54s\n", | |
| "Wall time: 12min 24s\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "%%time\n", | |
| "# this takes about 20 minutes\n", | |
| "\n", | |
| "histtemp = defaultdict(dict)\n", | |
| "basedir = cmip5dir / f\"historical/Amon/{var}\"\n", | |
| "for model in modelnames:\n", | |
| " modeldir = basedir / model\n", | |
| " for subset in modeldir.glob(\"r*\"):\n", | |
| " paths = subset.glob(\"*.nc\")\n", | |
| " with warnings.catch_warnings():\n", | |
| " warnings.simplefilter(\"ignore\", UserWarning)\n", | |
| " results = load_cube(paths, timespan, areas=[None, rhine])\n", | |
| " if any(result is None for result in results):\n", | |
| " continue\n", | |
| " histtemp[model][subset.stem] = results\n", | |
| " nvalues = len(histtemp[model].values())\n", | |
| " if nvalues == 0:\n", | |
| " del histtemp[model]\n", | |
| " print(model, \":\", nvalues, \"value(s)\")" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## Calculate the average tempearture\n", | |
| "\n", | |
| "We average over the different subsets for each model/RCP combination. We do this by using the iris `merge_cube` option: it will merge across a common, increasing coordinate. We add a new \"subset\" coordinate for this purpose, then merge the list of cubes into one, then average across this subset coordinate.\n", | |
| "\n", | |
| "Some merges fail; for now, these are thrown out and ignored (with a short message printed).\n", | |
| "\n", | |
| "The averaged cube is stored in the `avtemp` variable.\n", | |
| "\n", | |
| "This is also the point where we add the historical data to the original `temp` dictionary, as a special kind of RCP. We alos make a backup (deep copy) of the original data, since we don't want to re-read all the data if we make a mistake below." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 12, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "# Add the historical temperatures to the other temperature dict, and make a copy of the whole thing.\n", | |
| "temp['historical'] = histtemp\n", | |
| "backup = deepcopy(temp)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 13, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "# Start from this cell (the data backup) if a mistake was made anywhere below\n", | |
| "temp = deepcopy(backup)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 14, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "name": "stderr", | |
| "output_type": "stream", | |
| "text": [ | |
| "/opt/conda/lib/python3.6/site-packages/iris/coords.py:1355: UserWarning: Collapsing a non-contiguous coordinate. Metadata may not be fully descriptive for 'subset'.\n", | |
| " warnings.warn(msg.format(self.name()))\n", | |
| "/opt/conda/lib/python3.6/site-packages/iris/coords.py:1355: UserWarning: Collapsing a non-contiguous coordinate. Metadata may not be fully descriptive for 'latitude'.\n", | |
| " warnings.warn(msg.format(self.name()))\n" | |
| ] | |
| }, | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "Failed to merge cubes for EC-EARTH -- 45\n", | |
| "Failed to merge cubes for EC-EARTH -- 45\n", | |
| "Failed to merge cubes for MIROC-ESM-CHEM -- 45\n", | |
| "Failed to merge cubes for MIROC-ESM-CHEM -- 45\n", | |
| "Failed to merge cubes for HadGEM2-ES -- 60\n", | |
| "Failed to merge cubes for HadGEM2-ES -- 60\n", | |
| "Failed to merge cubes for EC-EARTH -- 85\n", | |
| "Failed to merge cubes for EC-EARTH -- 85\n", | |
| "Failed to merge cubes for HadGEM2-CC -- 85\n", | |
| "Failed to merge cubes for HadGEM2-CC -- 85\n", | |
| "Failed to merge cubes for CESM1-WACCM -- historical\n", | |
| "Failed to merge cubes for CESM1-WACCM -- historical\n", | |
| "Failed to merge cubes for EC-EARTH -- historical\n", | |
| "Failed to merge cubes for EC-EARTH -- historical\n", | |
| "Failed to merge cubes for GISS-E2-R -- historical\n", | |
| "Failed to merge cubes for GISS-E2-R -- historical\n", | |
| "Failed to merge cubes for HadGEM2-CC -- historical\n", | |
| "Failed to merge cubes for HadGEM2-CC -- historical\n", | |
| "Failed to merge cubes for HadGEM2-ES -- historical\n", | |
| "Failed to merge cubes for HadGEM2-ES -- historical\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "# Average results over subsets\n", | |
| "# Skip failed merges\n", | |
| "avtemp = defaultdict(dict)\n", | |
| "for rcp in temp.keys():\n", | |
| " for model in temp[rcp].keys():\n", | |
| " avtemp[rcp][model] = []\n", | |
| " values = list(temp[rcp][model].values())\n", | |
| " for iregion in range(len(values[0])):\n", | |
| " data = np.array([value[iregion].data for value in values])\n", | |
| " #print(data)\n", | |
| " cubes = iris.cube.CubeList([value[iregion] for value in values])\n", | |
| " if rcp != 'historical':\n", | |
| " df.loc[df['Name'] == model, f\"RCP {int(rcp)/10:0.1f}\"] = len(cubes)\n", | |
| "\n", | |
| " for i, cube in enumerate(cubes):\n", | |
| " coordnames = [coord.long_name for coord in cube.coords()]\n", | |
| " if 'subset' in coordnames:\n", | |
| " cube.remove_coord('subset') \n", | |
| " cube.add_aux_coord(iris.coords.AuxCoord(i, long_name='subset', units='no_unit'))\n", | |
| " equalise_attributes(cubes)\n", | |
| " try:\n", | |
| " cube = cubes.merge_cube()\n", | |
| " except iris.exceptions.MergeError as exc:\n", | |
| " print('Failed to merge cubes for', model, \" -- \", rcp)\n", | |
| " continue\n", | |
| " # We don't need to average is there was only 1 cube (subset) to begin with. \n", | |
| " # In fact, Iris will complain and throw an exception if we try to average across a dimension with just 1 point.\n", | |
| " if len(cubes) > 1:\n", | |
| " cube = cube.collapsed(['subset'], iris.analysis.MEAN)\n", | |
| " avtemp[rcp][model].append(cube)\n", | |
| "avtemp ;" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "Show what data we actually have" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 15, | |
| "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>Name</th>\n", | |
| " <th>RCP 2.6</th>\n", | |
| " <th>RCP 4.5</th>\n", | |
| " <th>RCP 6.0</th>\n", | |
| " <th>RCP 8.5</th>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>Label</th>\n", | |
| " <th></th>\n", | |
| " <th></th>\n", | |
| " <th></th>\n", | |
| " <th></th>\n", | |
| " <th></th>\n", | |
| " </tr>\n", | |
| " </thead>\n", | |
| " <tbody>\n", | |
| " <tr>\n", | |
| " <th>0</th>\n", | |
| " <td>ACCESS1-0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>1</th>\n", | |
| " <td>ACCESS1-3</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>2</th>\n", | |
| " <td>bcc-csm1-1</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>bcc-csm1-1-m</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>BNU-ESM</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>5</th>\n", | |
| " <td>CanCM4</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>6</th>\n", | |
| " <td>CanESM2</td>\n", | |
| " <td>5</td>\n", | |
| " <td>5</td>\n", | |
| " <td>0</td>\n", | |
| " <td>5</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>7</th>\n", | |
| " <td>CCSM4</td>\n", | |
| " <td>6</td>\n", | |
| " <td>6</td>\n", | |
| " <td>6</td>\n", | |
| " <td>6</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>8</th>\n", | |
| " <td>CESM1-BGC</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>9</th>\n", | |
| " <td>CESM1-CAM5</td>\n", | |
| " <td>3</td>\n", | |
| " <td>3</td>\n", | |
| " <td>3</td>\n", | |
| " <td>3</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>10</th>\n", | |
| " <td>CESM1-CAM5-1-FV2</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>11</th>\n", | |
| " <td>CESM1-WACCM</td>\n", | |
| " <td>1</td>\n", | |
| " <td>3</td>\n", | |
| " <td>0</td>\n", | |
| " <td>3</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>12</th>\n", | |
| " <td>CMCC-CESM</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>13</th>\n", | |
| " <td>CMCC-CM</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>14</th>\n", | |
| " <td>CMCC-CMS</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>15</th>\n", | |
| " <td>CNRM-CM5</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>5</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>16</th>\n", | |
| " <td>CSIRO-Mk3-6-0</td>\n", | |
| " <td>10</td>\n", | |
| " <td>10</td>\n", | |
| " <td>10</td>\n", | |
| " <td>10</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>17</th>\n", | |
| " <td>CSIRO-Mk3L-1-2</td>\n", | |
| " <td>0</td>\n", | |
| " <td>3</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>18</th>\n", | |
| " <td>EC-EARTH</td>\n", | |
| " <td>2</td>\n", | |
| " <td>7</td>\n", | |
| " <td>0</td>\n", | |
| " <td>6</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>19</th>\n", | |
| " <td>FGOALS-g2</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>20</th>\n", | |
| " <td>FGOALS-s2</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>21</th>\n", | |
| " <td>FIO-ESM</td>\n", | |
| " <td>3</td>\n", | |
| " <td>3</td>\n", | |
| " <td>3</td>\n", | |
| " <td>3</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>22</th>\n", | |
| " <td>GFDL-CM2p1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>23</th>\n", | |
| " <td>GFDL-CM3</td>\n", | |
| " <td>1</td>\n", | |
| " <td>3</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>24</th>\n", | |
| " <td>GFDL-ESM2G</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>25</th>\n", | |
| " <td>GFDL-ESM2M</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>26</th>\n", | |
| " <td>GISS-E2-H</td>\n", | |
| " <td>3</td>\n", | |
| " <td>16</td>\n", | |
| " <td>3</td>\n", | |
| " <td>5</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>27</th>\n", | |
| " <td>GISS-E2-H-CC</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>28</th>\n", | |
| " <td>GISS-E2-R</td>\n", | |
| " <td>3</td>\n", | |
| " <td>17</td>\n", | |
| " <td>3</td>\n", | |
| " <td>5</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>29</th>\n", | |
| " <td>GISS-E2-R-CC</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>30</th>\n", | |
| " <td>HadCM3</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>31</th>\n", | |
| " <td>HadGEM2-AO</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>32</th>\n", | |
| " <td>HadGEM2-CC</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>3</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>33</th>\n", | |
| " <td>HadGEM2-ES</td>\n", | |
| " <td>4</td>\n", | |
| " <td>4</td>\n", | |
| " <td>4</td>\n", | |
| " <td>3</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>34</th>\n", | |
| " <td>inmcm4</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>35</th>\n", | |
| " <td>IPSL-CM5A-LR</td>\n", | |
| " <td>4</td>\n", | |
| " <td>4</td>\n", | |
| " <td>1</td>\n", | |
| " <td>4</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>36</th>\n", | |
| " <td>IPSL-CM5A-MR</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>37</th>\n", | |
| " <td>IPSL-CM5B-LR</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>38</th>\n", | |
| " <td>MIROC-ESM</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>39</th>\n", | |
| " <td>MIROC-ESM-CHEM</td>\n", | |
| " <td>1</td>\n", | |
| " <td>9</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>40</th>\n", | |
| " <td>MIROC4h</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>41</th>\n", | |
| " <td>MIROC5</td>\n", | |
| " <td>3</td>\n", | |
| " <td>3</td>\n", | |
| " <td>3</td>\n", | |
| " <td>3</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>42</th>\n", | |
| " <td>MPI-ESM-LR</td>\n", | |
| " <td>3</td>\n", | |
| " <td>3</td>\n", | |
| " <td>0</td>\n", | |
| " <td>3</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>43</th>\n", | |
| " <td>MPI-ESM-MR</td>\n", | |
| " <td>1</td>\n", | |
| " <td>3</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>44</th>\n", | |
| " <td>MRI-CGCM3</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>45</th>\n", | |
| " <td>MRI-ESM1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>46</th>\n", | |
| " <td>NorESM1-M</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>47</th>\n", | |
| " <td>NorESM1-ME</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " </tr>\n", | |
| " </tbody>\n", | |
| "</table>\n", | |
| "</div>" | |
| ], | |
| "text/plain": [ | |
| " Name RCP 2.6 RCP 4.5 RCP 6.0 RCP 8.5\n", | |
| "Label \n", | |
| "0 ACCESS1-0 0 1 0 1\n", | |
| "1 ACCESS1-3 0 1 0 1\n", | |
| "2 bcc-csm1-1 1 1 1 1\n", | |
| "3 bcc-csm1-1-m 1 1 1 1\n", | |
| "4 BNU-ESM 1 1 0 1\n", | |
| "5 CanCM4 0 0 0 0\n", | |
| "6 CanESM2 5 5 0 5\n", | |
| "7 CCSM4 6 6 6 6\n", | |
| "8 CESM1-BGC 0 1 0 1\n", | |
| "9 CESM1-CAM5 3 3 3 3\n", | |
| "10 CESM1-CAM5-1-FV2 0 1 0 1\n", | |
| "11 CESM1-WACCM 1 3 0 3\n", | |
| "12 CMCC-CESM 0 0 0 1\n", | |
| "13 CMCC-CM 0 1 0 1\n", | |
| "14 CMCC-CMS 0 1 0 1\n", | |
| "15 CNRM-CM5 1 1 0 5\n", | |
| "16 CSIRO-Mk3-6-0 10 10 10 10\n", | |
| "17 CSIRO-Mk3L-1-2 0 3 0 0\n", | |
| "18 EC-EARTH 2 7 0 6\n", | |
| "19 FGOALS-g2 1 1 0 1\n", | |
| "20 FGOALS-s2 0 0 0 1\n", | |
| "21 FIO-ESM 3 3 3 3\n", | |
| "22 GFDL-CM2p1 0 0 0 0\n", | |
| "23 GFDL-CM3 1 3 1 1\n", | |
| "24 GFDL-ESM2G 1 1 1 1\n", | |
| "25 GFDL-ESM2M 1 1 1 1\n", | |
| "26 GISS-E2-H 3 16 3 5\n", | |
| "27 GISS-E2-H-CC 0 1 0 1\n", | |
| "28 GISS-E2-R 3 17 3 5\n", | |
| "29 GISS-E2-R-CC 0 1 0 1\n", | |
| "30 HadCM3 0 0 0 0\n", | |
| "31 HadGEM2-AO 1 1 1 1\n", | |
| "32 HadGEM2-CC 0 1 0 3\n", | |
| "33 HadGEM2-ES 4 4 4 3\n", | |
| "34 inmcm4 0 1 0 1\n", | |
| "35 IPSL-CM5A-LR 4 4 1 4\n", | |
| "36 IPSL-CM5A-MR 1 1 1 1\n", | |
| "37 IPSL-CM5B-LR 0 1 0 1\n", | |
| "38 MIROC-ESM 1 1 1 1\n", | |
| "39 MIROC-ESM-CHEM 1 9 1 1\n", | |
| "40 MIROC4h 0 0 0 0\n", | |
| "41 MIROC5 3 3 3 3\n", | |
| "42 MPI-ESM-LR 3 3 0 3\n", | |
| "43 MPI-ESM-MR 1 3 0 1\n", | |
| "44 MRI-CGCM3 1 1 1 1\n", | |
| "45 MRI-ESM1 0 0 0 1\n", | |
| "46 NorESM1-M 1 1 1 1\n", | |
| "47 NorESM1-ME 0 1 0 0" | |
| ] | |
| }, | |
| "execution_count": 15, | |
| "metadata": {}, | |
| "output_type": "execute_result" | |
| } | |
| ], | |
| "source": [ | |
| "df[['Name', 'RCP 2.6', 'RCP 4.5', 'RCP 6.0', 'RCP 8.5']]" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## Calculate relative temperature responses\n", | |
| "\n", | |
| "We now need to calculate the relative temperature responses, by subtracting the historical averages.\n", | |
| "\n", | |
| "For convenience later on (plotting), we also juggle our dictionary \"order\" around: since there will be separate subplots for each season, and each plot will have separate colours per RPC, we move these keys to the front. Our new dict, `reltemp`, now follows the order \"season\", \"rcp\", \"region\", \"model\" (note that the region corresponds to x and y on the plots). See also the last line in the cell below." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 16, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "# Calculate relative temperatures\n", | |
| "# Also flip the season and region to more outer keys\n", | |
| "\n", | |
| "# in `temp` and `avtemp`, the regions are list-based; they'll now be dict based, using the 'global' and 'local' keywords\n", | |
| "# Ditto for the seasons: cube access is by index `i`, dict access by abbrevation.\n", | |
| "regions = ['global', 'local'] \n", | |
| "reltemp = {}\n", | |
| "for i, season in enumerate(seasons): \n", | |
| " reltemp[season] = {}\n", | |
| " for rcp in rcps:\n", | |
| " reltemp[season][rcp] = {region: {} for region in regions}\n", | |
| " for model in avtemp[rcp].keys():\n", | |
| " for region, pred, hist in zip(regions, avtemp[rcp][model], avtemp['historical'][model]):\n", | |
| " reltemp[season][rcp][region][model] = pred[i] - hist[i]" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## Plots\n", | |
| "\n", | |
| "We create the plots in the style of Van den Hurk et al., with some minor modifications:\n", | |
| "- the identical substring for each subplot is moved out of the individual titles, into the main title\n", | |
| "- the axis labels and tick labels are shared\n", | |
| "- the x and y scales are identical across the figures\n", | |
| "\n", | |
| "We also calculate the best fit and R<sup>2</sup> in the plotting loop, so that we can directly plot the result as well.\n", | |
| "\n", | |
| "The plotting requires a nested loop:\n", | |
| "- an outer loop to step through the four subplots, one for each season\n", | |
| "- an inner loop to step through the individual RCPs, so each set of RCP data points is plotted in its own colour.\n", | |
| "\n", | |
| "There are some Matplotlib technicalities as well to consider:\n", | |
| "- the linear fit is done first, so that the individual data points can be overplot on top of the fitted lines. This requires an extra inner loop, to collect the data points of the individual RCPs into a single array before fitting.\n", | |
| "- matplotlib does not have an option to provide each data point with a numeric symbol. Therefore, the opacity for the data points is turned to zero, and we annotate each point with the relevant number (centered at the point position)\n", | |
| "- matplotlib normally uses a symbol or line (in the relevant colour) with black text for the legend. By setting `markersize` to 0, the symbol is not plotted. We set the label text with the appropriate colour, since this defaults to black.\n", | |
| "- Matplotlib allows for LaTeX inside labels and titles (text in general), provided LaTeX is installed\n", | |
| "- A subplot counter `i` is used to determine which subplots get axis labels or a legend." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 24, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<Figure size 864x864 with 4 Axes>" | |
| ] | |
| }, | |
| "metadata": { | |
| "needs_background": "light" | |
| }, | |
| "output_type": "display_data" | |
| } | |
| ], | |
| "source": [ | |
| "plt.style.use('bmh')\n", | |
| "\n", | |
| "# Fix the colours, to be similar to the KNMI figure\n", | |
| "colors = ['green', 'blue', 'orange', 'red']\n", | |
| "\n", | |
| "fig, axes = plt.subplots(2, 2, sharex=True, sharey=True, figsize=(12, 12), gridspec_kw={'hspace': 0.2, 'wspace': 0.1})\n", | |
| "plt.suptitle(\"Local versus global temperature response for 2071-2100\", y=0.95, fontsize=20)\n", | |
| "axes = axes.flatten()\n", | |
| "for iplot, (season, ax) in enumerate(zip(seasons, axes)):\n", | |
| " \n", | |
| " # Linear fit\n", | |
| " xtot = []\n", | |
| " ytot = []\n", | |
| " for rcp in rcps:\n", | |
| " xtot.append([cube.data for cube in reltemp[season][rcp]['global'].values()])\n", | |
| " ytot.append([cube.data for cube in reltemp[season][rcp]['local'].values()])\n", | |
| " x = np.concatenate(xtot)\n", | |
| " y = np.concatenate(ytot)\n", | |
| " coeffs, covar = np.polyfit(x, y, 1, cov=True)\n", | |
| " stderrs = np.sqrt([covar[0,0], covar[1,1]])\n", | |
| " poly = np.poly1d(coeffs)\n", | |
| " \n", | |
| " yfit = poly(x)\n", | |
| " r2 = np.sum((yfit - y.mean())**2) / np.sum((y - y.mean())**2) \n", | |
| " plt.text(0.1, 0.88, f\"slope = ${coeffs[0]:.2f} \\pm {stderrs[0]:.2f}$\", fontsize=14, transform=ax.transAxes)\n", | |
| " plt.text(0.1, 0.8, f\"$R^2 = {r2:.3f}$\", fontsize=14, transform=ax.transAxes)\n", | |
| " x = [-0.5, 5]\n", | |
| " ax.plot(x, poly(x), '-', color='black')\n", | |
| " poly = np.poly1d([1, 0])\n", | |
| " ax.plot(x, poly(x), '--', color='#333333')\n", | |
| "\n", | |
| " # Plot individual RCPs\n", | |
| " for ircp, rcp in enumerate(rcps):\n", | |
| " names = [name for name in reltemp[season][rcp]['global'].keys()]\n", | |
| " x = xtot[ircp]\n", | |
| " y = ytot[ircp]\n", | |
| " xtot.append(np.array(x))\n", | |
| " ytot.append(np.array(y))\n", | |
| " inames = [i for i, name in enumerate(modelnames) if name in names]\n", | |
| " plot = ax.plot(x, y, 'o', label=f\"RCP {rcp/10:.1f}\", alpha=0, color=colors[ircp])\n", | |
| " # Alternative way to obtain the plot colours, if we had let Matplotlib use its defaults\n", | |
| " #color = plot[0].get_color()\n", | |
| " for iname, xx, yy in zip(inames, x, y):\n", | |
| " ax.annotate(f\"{iname+1:02d}\", (xx, yy), color=colors[ircp], fontsize=12, horizontalalignment='center', verticalalignment='center')\n", | |
| " \n", | |
| " # subplot details\n", | |
| " ax.set_title(season.upper(), fontsize=16)\n", | |
| " ax.axis([-0.5, 5, -2, 9]) \n", | |
| " if iplot in [2, 3]:\n", | |
| " ax.set_xlabel('$\\Delta \\\\mathrm{T}_{\\\\mathrm{global}} [\\mathrm{K}]$', fontsize=14)\n", | |
| " if iplot in [0, 2]:\n", | |
| " ax.set_ylabel('$\\Delta \\\\mathrm{T}_{\\\\mathrm{local}}[\\mathrm{K}]$', fontsize=14)\n", | |
| " if iplot == 0:\n", | |
| " legend = ax.legend(loc='lower right', frameon=False, markerscale=0)\n", | |
| " for color, text in zip(colors, legend.get_texts()):\n", | |
| " text.set_color(color)\n", | |
| " \n", | |
| "# Save the figure to file as well\n", | |
| "plt.savefig('temp-response-rhine.png')" | |
| ] | |
| } | |
| ], | |
| "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": "3.6.7" | |
| } | |
| }, | |
| "nbformat": 4, | |
| "nbformat_minor": 2 | |
| } |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment