diff --git a/input/tasks.json b/input/tasks.json
index bd490d2..7bb3cb4 100644
--- a/input/tasks.json
+++ b/input/tasks.json
@@ -543,6 +543,17 @@
         "always_journal": true,
         "priority": 209
     },
+    "essence_spectra": {
+        "nice_name": "%pre ESSENCE spectra",
+        "active": true,
+        "update": false,
+        "module": "supernovae.tasks.essence",
+        "function": "do_essence_spectra",
+        "groups": ["spectra"],
+        "repo": "input/sne-external-spectra",
+        "always_journal": true,
+        "priority": 210
+    },
     "hst": {
         "nice_name": "Coordinates from HST pointings",
         "active": true,
diff --git a/tasks/essence.py b/tasks/essence.py
index d4f0111..2db89ad 100644
--- a/tasks/essence.py
+++ b/tasks/essence.py
@@ -1,11 +1,16 @@
 '''Import tasks for ESSENCE.
 '''
 import csv
+import datetime
 import os
 from glob import glob
 
 from astrocats.catalog.photometry import PHOTOMETRY, set_pd_mag_from_counts
-from astrocats.catalog.utils import pbar
+from astrocats.catalog.spectrum import SPECTRUM
+from astrocats.catalog.utils import is_number, pbar, pbar_strings
+from astropy.time import Time as astrotime
+
+from cdecimal import Decimal
 
 from ..supernova import SUPERNOVA
 
@@ -85,3 +90,97 @@ def do_essence_photo(catalog):
 
     catalog.journal_entries()
     return
+
+
+def do_essence_spectra(catalog):
+    task_str = catalog.get_current_task_str()
+
+    insdict = {
+        "lris": "LRIS",
+        "esi": "ESI",
+        "deimos": "DEIMOS",
+        "gmos": "GMOS",
+        "fors1": "FORS1",
+        "bluechannel": "Blue Channel",
+        "ldss2": "LDSS-2",
+        "ldss3": "LDSS-3",
+        "imacs": "IMACS",
+        "fast": "FAST"
+    }
+
+    teldict = {
+        "lris": "Keck",
+        "esi": "Keck",
+        "deimos": "Keck",
+        "gmos": "Gemini",
+        "fors1": "VLT",
+        "bluechannel": "MMT",
+        "ldss2": "Magellan Clay & Baade",
+        "ldss3": "Magellan Clay & Baade",
+        "imacs": "Magellan Clay & Baade",
+        "fast": "FLWO 1.5m"
+    }
+
+    file_names = glob(
+        os.path.join(catalog.get_current_task_repo(), 'ESSENCE', '*'))
+    oldname = ''
+    for fi, fname in enumerate(pbar_strings(file_names, task_str)):
+        filename = os.path.basename(fname)
+        fileparts = filename.split('_')
+        name = 'ESSENCE ' + fileparts[0]
+        name = catalog.get_preferred_name(name)
+        if oldname and name != oldname:
+            catalog.journal_entries()
+        oldname = name
+
+        if is_number(fileparts[1]):
+            doffset = 1
+        else:
+            if fileparts[1] != 'comb':
+                continue
+            doffset = 2
+
+        dstr = fileparts[doffset]
+        mjd = str(
+            astrotime(
+                datetime.datetime(
+                    year=int(dstr[:4]),
+                    month=int(dstr[4:6]),
+                    day=int(dstr[6:8])) + datetime.timedelta(days=float(dstr[
+                        8:]))).mjd)
+
+        instrument = fileparts[-1].split('.')[0]
+        telescope = teldict.get(instrument, '')
+        instrument = insdict.get(instrument, '')
+
+        with open(fname, 'r') as f:
+            data = csv.reader(f, delimiter=' ', skipinitialspace=True)
+            data = [list(i) for i in zip(*data)]
+            wavelengths = data[0]
+            fluxes = [str(Decimal('1.0e-15') + Decimal(x)) for x in data[1]]
+
+        name, source = catalog.new_entry(name, bibcode='2016ApJS..224....3N')
+
+        specdict = {
+            SPECTRUM.TIME: mjd,
+            SPECTRUM.U_TIME: 'MJD',
+            SPECTRUM.U_WAVELENGTHS: 'Angstrom',
+            SPECTRUM.WAVELENGTHS: wavelengths,
+            SPECTRUM.FLUXES: fluxes,
+            SPECTRUM.U_FLUXES: 'erg/s/cm^2/Angstrom',
+            SPECTRUM.FILENAME: filename,
+            SPECTRUM.SOURCE: source
+        }
+
+        if instrument:
+            specdict[SPECTRUM.INSTRUMENT] = instrument
+        if telescope:
+            specdict[SPECTRUM.TELESCOPE] = telescope
+
+        catalog.entries[name].add_spectrum(**specdict)
+
+        if catalog.args.travis and fi >= catalog.TRAVIS_QUERY_LIMIT:
+            break
+
+    catalog.journal_entries()
+    return