diff --git a/doc/tutorial.ipynb b/doc/tutorial.ipynb
index 8f9e213..b81818f 100644
--- a/doc/tutorial.ipynb
+++ b/doc/tutorial.ipynb
@@ -29,7 +29,8 @@
     "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
-    }
+    },
+    "tags": []
    },
    "outputs": [],
    "source": [
@@ -38,8 +39,7 @@
     "%matplotlib inline\n",
     "\n",
     "nt, nx = 100, 30\n",
-    "da = xr.DataArray(np.random.randn(nt, nx), dims=['time', 'x'],\n",
-    "                  name='foo') # all inputs need a name\n",
+    "da = xr.DataArray(np.random.randn(nt, nx), dims=['time', 'x'])\n",
     "display(da)\n",
     "da.plot()"
    ]
@@ -55,16 +55,60 @@
    "cell_type": "code",
    "execution_count": null,
    "metadata": {
-    "collapsed": false,
-    "jupyter": {
-     "outputs_hidden": false
-    }
+    "tags": []
    },
    "outputs": [],
    "source": [
     "from xhistogram.xarray import histogram\n",
     "\n",
     "bins = np.linspace(-4, 4, 20)\n",
+    "h = histogram(da, bins=[bins])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This error occurred because no name was given to `da` (e.g. `da.name == None`) and `histogram` needs that to determine the name of the output dimension.\n",
+    "We can solve this by either renaming `da` at the `histogram` input as\n",
+    "\n",
+    "```python\n",
+    "h = histogram(da.rename(\"foo\"), bins=[bins])\n",
+    "```\n",
+    "\n",
+    "or redefining `da` by giving the name at the input"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "da = xr.DataArray(np.random.randn(nt, nx), dims=['time', 'x'], name = \"foo\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now we can calculate the histogram and visualize it"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    },
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
     "h = histogram(da, bins=[bins])\n",
     "display(h)\n",
     "h.plot()"
@@ -90,7 +134,8 @@
     "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
-    }
+    },
+    "tags": []
    },
    "outputs": [],
    "source": [
@@ -113,7 +158,8 @@
     "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
-    }
+    },
+    "tags": []
    },
    "outputs": [],
    "source": [
@@ -136,7 +182,8 @@
     "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
-    }
+    },
+    "tags": []
    },
    "outputs": [],
    "source": [
@@ -158,7 +205,8 @@
     "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
-    }
+    },
+    "tags": []
    },
    "outputs": [],
    "source": [
@@ -182,7 +230,8 @@
     "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
-    }
+    },
+    "tags": []
    },
    "outputs": [],
    "source": [
@@ -219,7 +268,8 @@
     "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
-    }
+    },
+    "tags": []
    },
    "outputs": [],
    "source": [
@@ -232,7 +282,7 @@
     "    xr.open_dataset(Temp_url).tmn.load(),\n",
     "    xr.open_dataset(Salt_url).smn.load(),\n",
     "    xr.open_dataset(Oxy_url).omn.load()])\n",
-    "ds"
+    "display(ds)"
    ]
   },
   {
@@ -249,7 +299,8 @@
     "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
-    }
+    },
+    "tags": []
    },
    "outputs": [],
    "source": [
@@ -264,7 +315,8 @@
     "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
-    }
+    },
+    "tags": []
    },
    "outputs": [],
    "source": [
@@ -292,7 +344,8 @@
     "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
-    }
+    },
+    "tags": []
    },
    "outputs": [],
    "source": [
@@ -301,7 +354,7 @@
     "\n",
     "# Create a dz variable\n",
     "dz = np.diff(ds.lev)\n",
-    "dz =np.insert(dz, 0, dz[0])\n",
+    "dz = np.insert(dz, 0, dz[0])\n",
     "dz = xr.DataArray(dz, coords= {'lev':ds.lev}, dims='lev')\n",
     "\n",
     "# weight by volume of grid cell (resolution = 5degree, 1degree=110km)\n",
@@ -319,6 +372,75 @@
     "The ridges of this above plot are indicative of T/S classes with a lot of volume, and some of them are indicative of Mode Waters (example Eighteen Degree water with T$\\sim18^oC$, and S$\\sim36.5psu$. "
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now let's suppose that we have a mask for different regions in the planet. For the sake of simplicity, we will create a mask to separate the globe into three regions.\n",
+    "\n",
+    "- Tropical: `np.abs(latitude) <= 30`\n",
+    "- Temperate: `30 < np.abs(latitude) <= 60`\n",
+    "- Polar: `60 < np.abs(latitude)`"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "zones = np.abs(ds.lat / 30).round().rename(\"zones\")\n",
+    "display(zones)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now there are two ways of doing that, we can create a new dimension on the histogram"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "zonebins = np.arange(-0.5, 4, 1)\n",
+    "hTSw = histogram(ds.smn, ds.tmn, zones, bins=[sbins, tbins, zonebins], weights=dVol)\n",
+    "np.log10(hTSw.T).plot(levels=31, vmin=11.5, vmax=16, cmap='brg', col = \"zones_bin\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Or use groupby function"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "hTSw = ds.assign(dVol = dVol).groupby(zones).apply(lambda ds: histogram(ds.smn, ds.tmn, bins=[sbins, tbins], weights=ds.dVol))\n",
+    "np.log10(hTSw.T).plot(levels=31, vmin=11.5, vmax=16, cmap='brg', col = \"zones\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The second option is more verbose, but more efficient and also works with text or mask values that does not vary monotonically."
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -339,7 +461,8 @@
     "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
-    }
+    },
+    "tags": []
    },
    "outputs": [],
    "source": [
@@ -381,9 +504,9 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "coringa0.3.0",
    "language": "python",
-   "name": "python3"
+   "name": "coringa0.3.0"
   },
   "language_info": {
    "codemirror_mode": {
@@ -395,7 +518,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.6"
+   "version": "3.9.15"
   }
  },
  "nbformat": 4,