From c078ac549c03e982fa8fdd8e9ccd0dbfbf66ea62 Mon Sep 17 00:00:00 2001
From: Meaghan Freund <mof25@nau.edu>
Date: Wed, 2 Oct 2024 16:37:38 -0700
Subject: [PATCH 1/6] Basic progress and data estimation implementation.

---
 doc/devlog/2024-08-26-prism-adrio-demo.ipynb |  44 +++--
 epymorph/adrio/prism.py                      | 166 +++++++++++++++----
 2 files changed, 170 insertions(+), 40 deletions(-)

diff --git a/doc/devlog/2024-08-26-prism-adrio-demo.ipynb b/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
index a5685e14..b9322c29 100644
--- a/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
+++ b/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
@@ -47,9 +47,20 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 1,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Loading epymorph.adrio.us_tiger.GeometricCentroid:\n",
+      "  |####################| 100%  (10.638s)\n",
+      "Loading epymorph.adrio.prism.Precipitation:\n",
+      "  |####################| 100%  (2.736s)\n"
+     ]
+    }
+   ],
    "source": [
     "from epymorph.geography.us_census import CountyScope\n",
     "from epymorph import *\n",
@@ -76,13 +87,13 @@
     "        \"precipitation\": prism.Precipitation(time_period),\n",
     "    },\n",
     ")\n",
-    "\n",
-    "precipitation = evaluate_param(rume, \"precipitation\")"
+    "with sim_messaging():\n",
+    "    precipitation = evaluate_param(rume, \"precipitation\")\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [
     {
@@ -164,9 +175,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 5,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Loading epymorph.adrio.prism.Temperature:\n",
+      "  |####################| 100%  (81.437s)\n"
+     ]
+    }
+   ],
    "source": [
     "from epymorph.geography.us_census import StateScope\n",
     "from epymorph import *\n",
@@ -191,20 +211,20 @@
     "        \"temperature\": prism.Temperature(time_period, \"Mean\"),\n",
     "    },\n",
     ")\n",
-    "\n",
-    "temperature = evaluate_param(rume, \"temperature\")"
+    "with sim_messaging():\n",
+    "    temperature = evaluate_param(rume, \"temperature\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "States: ['04']\n",
+      "Counties: ['04']\n",
       "\n",
       "Temperature in Celsius:\n",
       " [[18.10199928]\n",
@@ -242,7 +262,7 @@
     }
    ],
    "source": [
-    "print(f\"States: {geoids}\\n\")\n",
+    "print(f\"Counties: {geoids}\\n\")\n",
     "\n",
     "print(f\"Temperature in Celsius:\\n {temperature}\\n\")"
    ]
diff --git a/epymorph/adrio/prism.py b/epymorph/adrio/prism.py
index 0d8188e9..ef8dcb87 100644
--- a/epymorph/adrio/prism.py
+++ b/epymorph/adrio/prism.py
@@ -10,20 +10,53 @@
 from numpy.typing import NDArray
 from typing_extensions import override
 
-from epymorph.adrio.adrio import Adrio
-from epymorph.cache import load_or_fetch_url, module_cache_path
+from epymorph.adrio.adrio import Adrio, ProgressCallback, adrio_cache
+from epymorph.cache import check_file_in_cache, load_or_fetch_url, module_cache_path
 from epymorph.data_shape import Shapes
 from epymorph.data_type import CentroidType
+from epymorph.data_usage import DataEstimate
 from epymorph.error import DataResourceException
 from epymorph.geography.scope import GeoScope
 from epymorph.geography.us_census import STATE, CensusScope
+from epymorph.geography.us_tiger import CacheEstimate, _url_to_cache_path
 from epymorph.simulation import AttributeDef, TimeFrame
 
 _PRISM_CACHE_PATH = module_cache_path(__name__)
 
 
+def _generate_file_name(
+    attribute: str,
+    latest_date: datetype,
+    last_completed_month: datetype,
+    date: datetype,
+) -> tuple[str, str, str]:
+    """
+    Generates the url for the given date and climate attribute. Returns a tuple
+    of strings with the url, stability, and formatted date for other file name usage.
+    """
+
+    if date.year == latest_date.year and date.month == latest_date.month:
+        stability = "early"
+
+    # if it is before the last finished month
+    elif date > last_completed_month:
+        stability = "provisional"
+
+    # if it is older than 6 completed months
+    else:
+        stability = "stable"
+
+    # format the date for the url
+    formatted_date = date.strftime("%Y%m%d")
+    year = date.year
+
+    url = f"https://ftp.prism.oregonstate.edu/daily/{attribute}/{year}/PRISM_{attribute}_{stability}_4kmD2_{formatted_date}_bil.zip"
+
+    return url, stability, formatted_date
+
+
 def _fetch_raster(
-    attribute: str, date_range: TimeFrame
+    attribute: str, date_range: TimeFrame, progress: ProgressCallback
 ) -> Generator[BytesIO, None, None]:
     """
     Fetches the raster values at the url with the given attribute and date range.
@@ -43,26 +76,14 @@ def _fetch_raster(
     six_months_ago = datetype.today() + relativedelta(months=-6)
     last_completed_month = six_months_ago.replace(day=1) - timedelta(days=1)
 
-    for single_date in date_list:
-        if (
-            single_date.year == latest_date.year
-            and single_date.month == latest_date.month
-        ):
-            stability = "early"
-
-        # if it is before the last finished month
-        elif single_date > last_completed_month:
-            stability = "provisional"
-
-        # if it is older than 6 completed months
-        else:
-            stability = "stable"
-
-        # format the date for the url
-        formatted_date = single_date.strftime("%Y%m%d")
-        year = single_date.year
+    # for progress tracking
+    processing_steps = len(date_list) + 1
 
-        url = f"https://ftp.prism.oregonstate.edu/daily/{attribute}/{year}/PRISM_{attribute}_{stability}_4kmD2_{formatted_date}_bil.zip"
+    # include i in the loop, representing the processing steps
+    for i, single_date in enumerate(date_list):
+        url, stability, formatted_date = _generate_file_name(
+            attribute, latest_date, last_completed_month, single_date
+        )
 
         # load/fetch the url for the file
         try:
@@ -73,16 +94,22 @@ def _fetch_raster(
 
         file.name = f"PRISM_{attribute}_{stability}_4kmD2_{formatted_date}_bil.bil"
 
+        # if the progress isnt None
+        if progress is not None:
+            # incremement progress (i + 1), and divide by the processing steps
+            progress((i + 1) / processing_steps, None)
+        # (representing the progress by how many total files have been completed)
+
         yield file
 
 
 def _make_centroid_strategy_adrio(
-    attribute: str, date: TimeFrame, centroids: NDArray
+    attribute: str, date: TimeFrame, centroids: NDArray, progress: ProgressCallback
 ) -> NDArray[np.float64]:
     """
     Retrieves the raster value at a centroid of a granularity.
     """
-    raster_files = _fetch_raster(attribute, date)
+    raster_files = _fetch_raster(attribute, date, progress)
     results = []
 
     # read in each file
@@ -128,6 +155,65 @@ def _validate_scope(scope: GeoScope) -> CensusScope:
     return scope
 
 
+def _estimate_size(
+    file_size: int, date_range: TimeFrame, attribute: str
+) -> CacheEstimate:
+    """
+    Estimate the size of all of the files, whether in cache or fetching.
+    """
+    # setup urls as list to check if theyre in the cache
+
+    # set up date variables to check stability
+    first_day = date_range.start_date
+    last_day = date_range.end_date
+    latest_date = datetype.today() - timedelta(days=1)
+
+    # the stability of PRISM data is defined by date, specified around the 6 month mark
+    six_months_ago = datetype.today() + relativedelta(months=-6)
+    last_completed_month = six_months_ago.replace(day=1) - timedelta(days=1)
+
+    date_list = [
+        first_day + timedelta(days=x) for x in range((last_day - first_day).days + 1)
+    ]
+
+    urls, _, _ = [
+        _generate_file_name(attribute, latest_date, last_completed_month, day)
+        for day in date_list
+    ]
+
+    total_files = date_range.duration_days
+    missing_files = total_files - sum(
+        1 for u in urls if check_file_in_cache(_url_to_cache_path(u))
+    )
+
+    est_file_size = Literal[file_size]
+
+    return CacheEstimate(
+        total_cache_size=total_files * est_file_size,
+        missing_cache_size=missing_files * est_file_size,
+    )
+
+
+def _estimate_data(
+    self, file_size: int, date_range: TimeFrame, attribute: str
+) -> DataEstimate:
+    """
+    Grab estimates for the PRISM simulation.
+    """
+    est = _estimate_size(file_size, date_range, attribute)
+
+    key = ""
+    return DataEstimate(
+        name=self.full_name,
+        cache_key=key,
+        new_network_bytes=est.missing_cache_size,
+        new_cache_bytes=est.missing_cache_size,
+        total_cache_bytes=est.total_cache_size,
+        max_bandwidth=None,
+    )
+
+
+@adrio_cache
 class Precipitation(Adrio[np.float64]):
     """
     Creates an TxN matrix of floats representing the amount of precipitation in an area,
@@ -141,12 +227,19 @@ class Precipitation(Adrio[np.float64]):
     def __init__(self, date_range: TimeFrame):
         self.date_range = _validate_dates(date_range)
 
+    # estimating data simulation
+    def estimate_sim(self) -> DataEstimate:
+        file_size = 1_200_00
+        return _estimate_data(self, file_size, self.date_range, "ppt")
+
     @override
     def evaluate_adrio(self) -> NDArray[np.float64]:
         scope = self.scope
         scope = _validate_scope(scope)
         centroids = self.data("centroid")
-        raster_vals = _make_centroid_strategy_adrio("ppt", self.date_range, centroids)
+        raster_vals = _make_centroid_strategy_adrio(
+            "ppt", self.date_range, centroids, self.progress
+        )
         return raster_vals
 
 
@@ -163,13 +256,18 @@ class DewPoint(Adrio[np.float64]):
     def __init__(self, date_range: TimeFrame):
         self.date_range = _validate_dates(date_range)
 
+    # estimating data simulation
+    def estimate_sim(self) -> DataEstimate:
+        file_size = 1_700_00
+        return _estimate_data(self, file_size, self.date_range, "tdmean")
+
     @override
     def evaluate_adrio(self) -> NDArray[np.float64]:
         scope = self.scope
         scope = _validate_scope(scope)
         centroids = self.data("centroid")
         raster_vals = _make_centroid_strategy_adrio(
-            "tdmean", self.date_range, centroids
+            "tdmean", self.date_range, centroids, self.progress
         )
         return raster_vals
 
@@ -198,6 +296,11 @@ def __init__(self, date_range: TimeFrame, temp_var: TemperatureType):
         self.temp_var = temp_var
         self.date_range = _validate_dates(date_range)
 
+    # estimating data simulation
+    def estimate_sim(self) -> DataEstimate:
+        file_size = 1_700_00
+        return _estimate_data(self, file_size, self.date_range, self.temp_var)
+
     @override
     def evaluate_adrio(self) -> NDArray[np.float64]:
         scope = self.scope
@@ -205,7 +308,7 @@ def evaluate_adrio(self) -> NDArray[np.float64]:
         temp_var = self.temp_variables[self.temp_var]
         centroids = self.data("centroid")
         raster_vals = _make_centroid_strategy_adrio(
-            temp_var, self.date_range, centroids
+            temp_var, self.date_range, centroids, self.progress
         )
 
         return raster_vals
@@ -231,11 +334,18 @@ def __init__(self, date_range: TimeFrame, vpd_var: VPDType):
         self.vpd_var = vpd_var
         self.date_range = _validate_dates(date_range)
 
+    # estimating data simulation
+    def estimate_sim(self) -> DataEstimate:
+        file_size = 1_600_00
+        return _estimate_data(self, file_size, self.date_range, self.vpd_var)
+
     @override
     def evaluate_adrio(self) -> NDArray[np.float64]:
         scope = self.scope
         scope = _validate_scope(scope)
         vpd_var = self.vpd_variables[self.vpd_var]
         centroids = self.data("centroid")
-        raster_vals = _make_centroid_strategy_adrio(vpd_var, self.date_range, centroids)
+        raster_vals = _make_centroid_strategy_adrio(
+            vpd_var, self.date_range, centroids, self.progress
+        )
         return raster_vals

From 304d8cbc02cebfe6ea00e36b4ede6eb3ce13defe Mon Sep 17 00:00:00 2001
From: Meaghan Freund <mof25@nau.edu>
Date: Fri, 4 Oct 2024 09:45:54 -0700
Subject: [PATCH 2/6] Corrected estimate_data and functions.

---
 doc/devlog/2024-08-26-prism-adrio-demo.ipynb | 60 ++++++++++++---
 epymorph/adrio/prism.py                      | 79 ++++++++++----------
 2 files changed, 87 insertions(+), 52 deletions(-)

diff --git a/doc/devlog/2024-08-26-prism-adrio-demo.ipynb b/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
index b9322c29..ea3ab1f2 100644
--- a/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
+++ b/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
@@ -47,17 +47,19 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Loading epymorph.adrio.us_tiger.GeometricCentroid:\n",
-      "  |####################| 100%  (10.638s)\n",
-      "Loading epymorph.adrio.prism.Precipitation:\n",
-      "  |####################| 100%  (2.736s)\n"
+      "ADRIO data usage estimation:\n",
+      "- epymorph.adrio.us_tiger.GeometricCentroid will download 0 Bytes and write 0 Bytes to disk\n",
+      "- epymorph.adrio.prism.Precipitation will download 1.2 MB and write 1.2 MB to disk\n",
+      "In total we will:\n",
+      "- Download 1.2 MB, taking a second (assuming 1.0 MB/s)\n",
+      "- Write 1.2 MB to disk cache (you have 249.2 GB free space)\n"
      ]
     }
    ],
@@ -88,12 +90,22 @@
     "    },\n",
     ")\n",
     "with sim_messaging():\n",
-    "    precipitation = evaluate_param(rume, \"precipitation\")\n"
+    "    rume.estimate_data()\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "with sim_messaging():\n",
+    "    precipitation = evaluate_param(rume, \"precipitation\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
    "metadata": {},
    "outputs": [
     {
@@ -175,19 +187,24 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Loading epymorph.adrio.prism.Temperature:\n",
-      "  |####################| 100%  (81.437s)\n"
+      "ADRIO data usage estimation:\n",
+      "- epymorph.adrio.prism.Temperature will download 51.0 MB and write 51.0 MB to disk\n",
+      "- epymorph.adrio.us_tiger.GeometricCentroid will download 0 Bytes and write 0 Bytes to disk\n",
+      "In total we will:\n",
+      "- Download 51.0 MB, taking 51 seconds (assuming 1.0 MB/s)\n",
+      "- Write 51.0 MB to disk cache (you have 249.2 GB free space)\n"
      ]
     }
    ],
    "source": [
+    "from epymorph.rume import estimate_report\n",
     "from epymorph.geography.us_census import StateScope\n",
     "from epymorph import *\n",
     "from epymorph.adrio import acs5, us_tiger\n",
@@ -206,18 +223,37 @@
     "    scope=state_scope,\n",
     "    time_frame=time_period,\n",
     "    params={\n",
+    "        \"temperature\": prism.Temperature(time_period, \"Mean\"),\n",
     "        \"population\": acs5.Population(),\n",
     "        \"centroid\": us_tiger.GeometricCentroid(),\n",
-    "        \"temperature\": prism.Temperature(time_period, \"Mean\"),\n",
     "    },\n",
     ")\n",
+    "with sim_messaging():\n",
+    "    rume.estimate_data()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Loading epymorph.adrio.prism.Temperature:\n",
+      "  |####################| 100%  (0.499s)\n"
+     ]
+    }
+   ],
+   "source": [
     "with sim_messaging():\n",
     "    temperature = evaluate_param(rume, \"temperature\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [
     {
diff --git a/epymorph/adrio/prism.py b/epymorph/adrio/prism.py
index ef8dcb87..7766d8f4 100644
--- a/epymorph/adrio/prism.py
+++ b/epymorph/adrio/prism.py
@@ -98,7 +98,6 @@ def _fetch_raster(
         if progress is not None:
             # incremement progress (i + 1), and divide by the processing steps
             progress((i + 1) / processing_steps, None)
-        # (representing the progress by how many total files have been completed)
 
         yield file
 
@@ -155,53 +154,44 @@ def _validate_scope(scope: GeoScope) -> CensusScope:
     return scope
 
 
-def _estimate_size(
-    file_size: int, date_range: TimeFrame, attribute: str
-) -> CacheEstimate:
+def _estimate_prism(
+    self, file_size: int, date_range: TimeFrame, attribute: str
+) -> DataEstimate:
     """
-    Estimate the size of all of the files, whether in cache or fetching.
+    Grab estimates for the PRISM simulation.
     """
+    est_file_size = file_size
+    total_files = date_range.duration_days
+
     # setup urls as list to check if theyre in the cache
 
-    # set up date variables to check stability
+    # setup date variables
     first_day = date_range.start_date
     last_day = date_range.end_date
     latest_date = datetype.today() - timedelta(days=1)
-
-    # the stability of PRISM data is defined by date, specified around the 6 month mark
     six_months_ago = datetype.today() + relativedelta(months=-6)
     last_completed_month = six_months_ago.replace(day=1) - timedelta(days=1)
-
     date_list = [
         first_day + timedelta(days=x) for x in range((last_day - first_day).days + 1)
     ]
 
-    urls, _, _ = [
-        _generate_file_name(attribute, latest_date, last_completed_month, day)
+    # get url names to check in cache
+    urls = [
+        _generate_file_name(attribute, latest_date, last_completed_month, day)[0]
         for day in date_list
     ]
 
-    total_files = date_range.duration_days
+    # sum the missing files
     missing_files = total_files - sum(
         1 for u in urls if check_file_in_cache(_url_to_cache_path(u))
     )
 
-    est_file_size = Literal[file_size]
-
-    return CacheEstimate(
+    # calculate the cache estimate
+    est = CacheEstimate(
         total_cache_size=total_files * est_file_size,
         missing_cache_size=missing_files * est_file_size,
     )
 
-
-def _estimate_data(
-    self, file_size: int, date_range: TimeFrame, attribute: str
-) -> DataEstimate:
-    """
-    Grab estimates for the PRISM simulation.
-    """
-    est = _estimate_size(file_size, date_range, attribute)
-
     key = ""
     return DataEstimate(
         name=self.full_name,
@@ -227,10 +217,10 @@ class Precipitation(Adrio[np.float64]):
     def __init__(self, date_range: TimeFrame):
         self.date_range = _validate_dates(date_range)
 
-    # estimating data simulation
-    def estimate_sim(self) -> DataEstimate:
-        file_size = 1_200_00
-        return _estimate_data(self, file_size, self.date_range, "ppt")
+    def estimate_data(self) -> DataEstimate:
+        file_size = 1_200_000
+        est = _estimate_prism(self, file_size, self.date_range, "ppt")
+        return est
 
     @override
     def evaluate_adrio(self) -> NDArray[np.float64]:
@@ -256,10 +246,13 @@ class DewPoint(Adrio[np.float64]):
     def __init__(self, date_range: TimeFrame):
         self.date_range = _validate_dates(date_range)
 
-    # estimating data simulation
-    def estimate_sim(self) -> DataEstimate:
-        file_size = 1_700_00
-        return _estimate_data(self, file_size, self.date_range, "tdmean")
+    def estimate_data(self) -> DataEstimate:
+        year = self.date_range.end_date.year
+        if year > 2020:
+            file_size = 1_800_000
+        else:
+            file_size = 1_400_000
+        return _estimate_prism(self, file_size, self.date_range, "tdmean")
 
     @override
     def evaluate_adrio(self) -> NDArray[np.float64]:
@@ -296,10 +289,13 @@ def __init__(self, date_range: TimeFrame, temp_var: TemperatureType):
         self.temp_var = temp_var
         self.date_range = _validate_dates(date_range)
 
-    # estimating data simulation
-    def estimate_sim(self) -> DataEstimate:
-        file_size = 1_700_00
-        return _estimate_data(self, file_size, self.date_range, self.temp_var)
+    def estimate_data(self) -> DataEstimate:
+        year = self.date_range.end_date.year
+        if year > 2020:
+            file_size = 1_700_000
+        else:
+            file_size = 1_400_000
+        return _estimate_prism(self, file_size, self.date_range, self.temp_var)
 
     @override
     def evaluate_adrio(self) -> NDArray[np.float64]:
@@ -334,10 +330,13 @@ def __init__(self, date_range: TimeFrame, vpd_var: VPDType):
         self.vpd_var = vpd_var
         self.date_range = _validate_dates(date_range)
 
-    # estimating data simulation
-    def estimate_sim(self) -> DataEstimate:
-        file_size = 1_600_00
-        return _estimate_data(self, file_size, self.date_range, self.vpd_var)
+    def estimate_data(self) -> DataEstimate:
+        year = self.date_range.end_date.year
+        if year > 2020:
+            file_size = 1_700_000
+        else:
+            file_size = 1_300_000
+        return _estimate_prism(self, file_size, self.date_range, self.vpd_var)
 
     @override
     def evaluate_adrio(self) -> NDArray[np.float64]:

From 14aa35517ece8e322700e08d174a5f09227c1ae2 Mon Sep 17 00:00:00 2001
From: Meaghan Freund <mof25@nau.edu>
Date: Fri, 4 Oct 2024 10:49:54 -0700
Subject: [PATCH 3/6] Fixed caching estimation.

---
 doc/devlog/2024-08-26-prism-adrio-demo.ipynb | 37 +++++++++++++++-----
 epymorph/adrio/prism.py                      | 23 ++++++------
 2 files changed, 41 insertions(+), 19 deletions(-)

diff --git a/doc/devlog/2024-08-26-prism-adrio-demo.ipynb b/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
index ea3ab1f2..5ea61d34 100644
--- a/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
+++ b/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
@@ -47,7 +47,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 1,
    "metadata": {},
    "outputs": [
     {
@@ -95,9 +95,20 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 2,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Loading epymorph.adrio.us_tiger.GeometricCentroid:\n",
+      "  |####################| 100%  (15.719s)\n",
+      "Loading epymorph.adrio.prism.Precipitation:\n",
+      "  |####################| 100%  (1.521s)\n"
+     ]
+    }
+   ],
    "source": [
     "with sim_messaging():\n",
     "    precipitation = evaluate_param(rume, \"precipitation\")"
@@ -105,7 +116,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [
     {
@@ -187,7 +198,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
@@ -204,7 +215,6 @@
     }
    ],
    "source": [
-    "from epymorph.rume import estimate_report\n",
     "from epymorph.geography.us_census import StateScope\n",
     "from epymorph import *\n",
     "from epymorph.adrio import acs5, us_tiger\n",
@@ -234,15 +244,24 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
+      "Loading epymorph.adrio.us_tiger.GeometricCentroid:\n",
+      "  |                    | 0% \r"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "  |####################| 100%  (2.716s)\n",
       "Loading epymorph.adrio.prism.Temperature:\n",
-      "  |####################| 100%  (0.499s)\n"
+      "  |####################| 100%  (24.449s)\n"
      ]
     }
    ],
@@ -253,7 +272,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
diff --git a/epymorph/adrio/prism.py b/epymorph/adrio/prism.py
index 7766d8f4..76b38d5d 100644
--- a/epymorph/adrio/prism.py
+++ b/epymorph/adrio/prism.py
@@ -18,7 +18,7 @@
 from epymorph.error import DataResourceException
 from epymorph.geography.scope import GeoScope
 from epymorph.geography.us_census import STATE, CensusScope
-from epymorph.geography.us_tiger import CacheEstimate, _url_to_cache_path
+from epymorph.geography.us_tiger import CacheEstimate
 from epymorph.simulation import AttributeDef, TimeFrame
 
 _PRISM_CACHE_PATH = module_cache_path(__name__)
@@ -29,7 +29,7 @@ def _generate_file_name(
     latest_date: datetype,
     last_completed_month: datetype,
     date: datetype,
-) -> tuple[str, str, str]:
+) -> tuple[str, str]:
     """
     Generates the url for the given date and climate attribute. Returns a tuple
     of strings with the url, stability, and formatted date for other file name usage.
@@ -52,7 +52,9 @@ def _generate_file_name(
 
     url = f"https://ftp.prism.oregonstate.edu/daily/{attribute}/{year}/PRISM_{attribute}_{stability}_4kmD2_{formatted_date}_bil.zip"
 
-    return url, stability, formatted_date
+    bil_name = f"PRISM_{attribute}_{stability}_4kmD2_{formatted_date}_bil.bil"
+
+    return url, bil_name
 
 
 def _fetch_raster(
@@ -81,7 +83,7 @@ def _fetch_raster(
 
     # include i in the loop, representing the processing steps
     for i, single_date in enumerate(date_list):
-        url, stability, formatted_date = _generate_file_name(
+        url, bil_name = _generate_file_name(
             attribute, latest_date, last_completed_month, single_date
         )
 
@@ -92,11 +94,11 @@ def _fetch_raster(
         except Exception as e:
             raise DataResourceException("Unable to fetch PRISM data.") from e
 
-        file.name = f"PRISM_{attribute}_{stability}_4kmD2_{formatted_date}_bil.bil"
+        file.name = bil_name
 
         # if the progress isnt None
         if progress is not None:
-            # incremement progress (i + 1), and divide by the processing steps
+            # progress by one, increasing percentage done
             progress((i + 1) / processing_steps, None)
 
         yield file
@@ -181,9 +183,9 @@ def _estimate_prism(
         for day in date_list
     ]
 
-    # sum the missing files
+    # sum the files needed to download
     missing_files = total_files - sum(
-        1 for u in urls if check_file_in_cache(_url_to_cache_path(u))
+        1 for u in urls if check_file_in_cache(_PRISM_CACHE_PATH / Path(u).name)
     )
 
     # calculate the cache estimate
@@ -192,7 +194,7 @@ def _estimate_prism(
         missing_cache_size=missing_files * est_file_size,
     )
 
-    key = ""
+    key = f"prism:{attribute}:{date_range}"
     return DataEstimate(
         name=self.full_name,
         cache_key=key,
@@ -291,11 +293,12 @@ def __init__(self, date_range: TimeFrame, temp_var: TemperatureType):
 
     def estimate_data(self) -> DataEstimate:
         year = self.date_range.end_date.year
+        temp_var = self.temp_variables[self.temp_var]
         if year > 2020:
             file_size = 1_700_000
         else:
             file_size = 1_400_000
-        return _estimate_prism(self, file_size, self.date_range, self.temp_var)
+        return _estimate_prism(self, file_size, self.date_range, temp_var)
 
     @override
     def evaluate_adrio(self) -> NDArray[np.float64]:

From ed151e988efbbed3a720dbb22ea958486f807571 Mon Sep 17 00:00:00 2001
From: Meaghan Freund <mof25@nau.edu>
Date: Fri, 4 Oct 2024 18:18:22 -0700
Subject: [PATCH 4/6] Minor comment changes.

---
 epymorph/adrio/prism.py | 33 +++++++++++++++++++--------------
 1 file changed, 19 insertions(+), 14 deletions(-)

diff --git a/epymorph/adrio/prism.py b/epymorph/adrio/prism.py
index 76b38d5d..67ceb6d4 100644
--- a/epymorph/adrio/prism.py
+++ b/epymorph/adrio/prism.py
@@ -32,7 +32,7 @@ def _generate_file_name(
 ) -> tuple[str, str]:
     """
     Generates the url for the given date and climate attribute. Returns a tuple
-    of strings with the url, stability, and formatted date for other file name usage.
+    of strings with the url and the name of the bil file within the zip file.
     """
 
     if date.year == latest_date.year and date.month == latest_date.month:
@@ -81,7 +81,6 @@ def _fetch_raster(
     # for progress tracking
     processing_steps = len(date_list) + 1
 
-    # include i in the loop, representing the processing steps
     for i, single_date in enumerate(date_list):
         url, bil_name = _generate_file_name(
             attribute, latest_date, last_completed_month, single_date
@@ -94,13 +93,13 @@ def _fetch_raster(
         except Exception as e:
             raise DataResourceException("Unable to fetch PRISM data.") from e
 
-        file.name = bil_name
-
         # if the progress isnt None
         if progress is not None:
             # progress by one, increasing percentage done
             progress((i + 1) / processing_steps, None)
 
+        file.name = bil_name
+
         yield file
 
 
@@ -160,7 +159,7 @@ def _estimate_prism(
     self, file_size: int, date_range: TimeFrame, attribute: str
 ) -> DataEstimate:
     """
-    Grab estimates for the PRISM simulation.
+    Calculate estimates for downloading PRISM files.
     """
     est_file_size = file_size
     total_files = date_range.duration_days
@@ -220,7 +219,7 @@ def __init__(self, date_range: TimeFrame):
         self.date_range = _validate_dates(date_range)
 
     def estimate_data(self) -> DataEstimate:
-        file_size = 1_200_000
+        file_size = 1_200_000  # no significant change in size, average to about 1.2MB
         est = _estimate_prism(self, file_size, self.date_range, "ppt")
         return est
 
@@ -250,10 +249,12 @@ def __init__(self, date_range: TimeFrame):
 
     def estimate_data(self) -> DataEstimate:
         year = self.date_range.end_date.year
+
+        # file sizes are larger after the year 2020
         if year > 2020:
-            file_size = 1_800_000
+            file_size = 1_800_000  # average to 1.8MB after 2020
         else:
-            file_size = 1_400_000
+            file_size = 1_400_000  # average to 1.4MB 2020 and before
         return _estimate_prism(self, file_size, self.date_range, "tdmean")
 
     @override
@@ -270,7 +271,7 @@ def evaluate_adrio(self) -> NDArray[np.float64]:
 class Temperature(Adrio[np.float64]):
     """
     Creates an TxN matrix of floats representing the temperature in an area, represented
-      in degrees Celsius (°C).
+    in degrees Celsius (°C).
     """
 
     date_range: TimeFrame
@@ -294,10 +295,12 @@ def __init__(self, date_range: TimeFrame, temp_var: TemperatureType):
     def estimate_data(self) -> DataEstimate:
         year = self.date_range.end_date.year
         temp_var = self.temp_variables[self.temp_var]
+
+        # file sizes are larger after the year 2020
         if year > 2020:
-            file_size = 1_700_000
+            file_size = 1_700_000  # average to 1.7MB after 2020
         else:
-            file_size = 1_400_000
+            file_size = 1_400_000  # average to 1.4MB 2020 and before
         return _estimate_prism(self, file_size, self.date_range, temp_var)
 
     @override
@@ -316,7 +319,7 @@ def evaluate_adrio(self) -> NDArray[np.float64]:
 class VaporPressureDeficit(Adrio[np.float64]):
     """
     Creates an TxN matrix of floats representing the vapor pressure deficit in an area,
-      represented in hectopascals (hPa).
+    represented in hectopascals (hPa).
     """
 
     date_range: TimeFrame
@@ -335,10 +338,12 @@ def __init__(self, date_range: TimeFrame, vpd_var: VPDType):
 
     def estimate_data(self) -> DataEstimate:
         year = self.date_range.end_date.year
+
+        # file sizes are larger after the year 2020
         if year > 2020:
-            file_size = 1_700_000
+            file_size = 1_700_000  # average to 1.7MB after 2020
         else:
-            file_size = 1_300_000
+            file_size = 1_300_000  # average to 1.3MB 2020 and before
         return _estimate_prism(self, file_size, self.date_range, self.vpd_var)
 
     @override

From 625d69c5a5d24583596ef85bfef5691e9e91aa3b Mon Sep 17 00:00:00 2001
From: Meaghan Freund <mof25@nau.edu>
Date: Fri, 4 Oct 2024 21:38:41 -0700
Subject: [PATCH 5/6] Removed estimate_data in devlog.

---
 doc/devlog/2024-08-26-prism-adrio-demo.ipynb | 75 +++-----------------
 1 file changed, 11 insertions(+), 64 deletions(-)

diff --git a/doc/devlog/2024-08-26-prism-adrio-demo.ipynb b/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
index 5ea61d34..7ac30ea9 100644
--- a/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
+++ b/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
@@ -54,12 +54,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "ADRIO data usage estimation:\n",
-      "- epymorph.adrio.us_tiger.GeometricCentroid will download 0 Bytes and write 0 Bytes to disk\n",
-      "- epymorph.adrio.prism.Precipitation will download 1.2 MB and write 1.2 MB to disk\n",
-      "In total we will:\n",
-      "- Download 1.2 MB, taking a second (assuming 1.0 MB/s)\n",
-      "- Write 1.2 MB to disk cache (you have 249.2 GB free space)\n"
+      "Loading epymorph.adrio.us_tiger.GeometricCentroid:\n",
+      "  |####################| 100%  (11.971s)\n",
+      "Loading epymorph.adrio.prism.Precipitation:\n",
+      "  |####################| 100%  (0.227s)\n"
      ]
     }
    ],
@@ -90,28 +88,7 @@
     "    },\n",
     ")\n",
     "with sim_messaging():\n",
-    "    rume.estimate_data()\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Loading epymorph.adrio.us_tiger.GeometricCentroid:\n",
-      "  |####################| 100%  (15.719s)\n",
-      "Loading epymorph.adrio.prism.Precipitation:\n",
-      "  |####################| 100%  (1.521s)\n"
-     ]
-    }
-   ],
-   "source": [
-    "with sim_messaging():\n",
-    "    precipitation = evaluate_param(rume, \"precipitation\")"
+    "    precipitation = evaluate_param(rume, \"precipitation\")\n"
    ]
   },
   {
@@ -198,19 +175,17 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "ADRIO data usage estimation:\n",
-      "- epymorph.adrio.prism.Temperature will download 51.0 MB and write 51.0 MB to disk\n",
-      "- epymorph.adrio.us_tiger.GeometricCentroid will download 0 Bytes and write 0 Bytes to disk\n",
-      "In total we will:\n",
-      "- Download 51.0 MB, taking 51 seconds (assuming 1.0 MB/s)\n",
-      "- Write 51.0 MB to disk cache (you have 249.2 GB free space)\n"
+      "Loading epymorph.adrio.us_tiger.GeometricCentroid:\n",
+      "  |####################| 100%  (3.046s)\n",
+      "Loading epymorph.adrio.prism.Temperature:\n",
+      "  |####################| 100%  (22.394s)\n"
      ]
     }
    ],
@@ -238,41 +213,13 @@
     "        \"centroid\": us_tiger.GeometricCentroid(),\n",
     "    },\n",
     ")\n",
-    "with sim_messaging():\n",
-    "    rume.estimate_data()\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Loading epymorph.adrio.us_tiger.GeometricCentroid:\n",
-      "  |                    | 0% \r"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "  |####################| 100%  (2.716s)\n",
-      "Loading epymorph.adrio.prism.Temperature:\n",
-      "  |####################| 100%  (24.449s)\n"
-     ]
-    }
-   ],
-   "source": [
     "with sim_messaging():\n",
     "    temperature = evaluate_param(rume, \"temperature\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [
     {

From c21652b6b60814dcefecbb4d0dedd9792e77f371 Mon Sep 17 00:00:00 2001
From: Meaghan Freund <mof25@nau.edu>
Date: Fri, 4 Oct 2024 21:45:15 -0700
Subject: [PATCH 6/6] Reformatted estimate_data in devlog.

---
 doc/devlog/2024-08-26-prism-adrio-demo.ipynb | 61 +++++++++++++++-----
 1 file changed, 48 insertions(+), 13 deletions(-)

diff --git a/doc/devlog/2024-08-26-prism-adrio-demo.ipynb b/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
index 7ac30ea9..35b66f31 100644
--- a/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
+++ b/doc/devlog/2024-08-26-prism-adrio-demo.ipynb
@@ -47,17 +47,19 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Loading epymorph.adrio.us_tiger.GeometricCentroid:\n",
-      "  |####################| 100%  (11.971s)\n",
-      "Loading epymorph.adrio.prism.Precipitation:\n",
-      "  |####################| 100%  (0.227s)\n"
+      "ADRIO data usage estimation:\n",
+      "- epymorph.adrio.us_tiger.GeometricCentroid will download 0 Bytes and write 0 Bytes to disk\n",
+      "- epymorph.adrio.prism.Precipitation will download 1.2 MB and write 1.2 MB to disk\n",
+      "In total we will:\n",
+      "- Download 1.2 MB, taking a second (assuming 1.0 MB/s)\n",
+      "- Write 1.2 MB to disk cache (you have 249.1 GB free space)\n"
      ]
     }
    ],
@@ -87,13 +89,23 @@
     "        \"precipitation\": prism.Precipitation(time_period),\n",
     "    },\n",
     ")\n",
+    "\n",
+    "rume.estimate_data()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
     "with sim_messaging():\n",
-    "    precipitation = evaluate_param(rume, \"precipitation\")\n"
+    "    precipitation = evaluate_param(rume, \"precipitation\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -175,17 +187,19 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Loading epymorph.adrio.us_tiger.GeometricCentroid:\n",
-      "  |####################| 100%  (3.046s)\n",
-      "Loading epymorph.adrio.prism.Temperature:\n",
-      "  |####################| 100%  (22.394s)\n"
+      "ADRIO data usage estimation:\n",
+      "- epymorph.adrio.prism.Temperature will download 51.0 MB and write 51.0 MB to disk\n",
+      "- epymorph.adrio.us_tiger.GeometricCentroid will download 0 Bytes and write 0 Bytes to disk\n",
+      "In total we will:\n",
+      "- Download 51.0 MB, taking 51 seconds (assuming 1.0 MB/s)\n",
+      "- Write 51.0 MB to disk cache (you have 249.1 GB free space)\n"
      ]
     }
    ],
@@ -213,13 +227,34 @@
     "        \"centroid\": us_tiger.GeometricCentroid(),\n",
     "    },\n",
     ")\n",
+    "\n",
+    "rume.estimate_data()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Loading epymorph.adrio.us_tiger.GeometricCentroid:\n",
+      "  |####################| 100%  (4.787s)\n",
+      "Loading epymorph.adrio.prism.Temperature:\n",
+      "  |####################| 100%  (21.742s)\n"
+     ]
+    }
+   ],
+   "source": [
     "with sim_messaging():\n",
     "    temperature = evaluate_param(rume, \"temperature\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [
     {