diff --git a/.github/workflows/test-build-push.yml b/.github/workflows/test-build-push.yml
index ec7463c2..03858ecf 100644
--- a/.github/workflows/test-build-push.yml
+++ b/.github/workflows/test-build-push.yml
@@ -14,24 +14,19 @@ jobs:
   pytest:
     strategy:
       matrix:
-        os:
-          - label: Linux
-            runner: ubuntu-latest
-          - label: macOS
-            runner: macos-latest
-
+        os: [ubunutu-latest, macos-latest]
         deps:
           - label: Latest
-            spec: ""
+            spec: >-
+              isce3
           - label: Minimum
-            spec: |
+            spec: >-
               python=3.8
+              isce3
               gdal=3.5
               h5py=3.6
-              h5netcdf=1.0
               numpy=1.20
               numba=0.54
-              pillow==7.0
               pydantic=2.1
               pymp-pypi=0.4.5
               pyproj=3.3
@@ -40,10 +35,14 @@ jobs:
               scipy=1.5
               shapely=1.8
               threadpoolctl>=3.0
+        exclude:  # TODO: Remove this once pymp is gone
+          - os: macos-latest
+            deps:
+              label: Latest
 
       fail-fast: false
-    name: ${{ matrix.os.label }} • ${{ matrix.deps.label }}
-    runs-on: ${{ matrix.os.runner }}
+    name: ${{ matrix.os }} • ${{ matrix.deps.label }}
+    runs-on: ${{ matrix.os }}
     defaults:
       run:
         shell: bash -l {0}
@@ -51,12 +50,14 @@ jobs:
       - name: Checkout
         uses: actions/checkout@v3
       - name: Setup environment
-        uses: mamba-org/provision-with-micromamba@main
+        uses: mamba-org/setup-micromamba@v1
         with:
           environment-file: conda-env.yml
           environment-name: dolphin-env
-          extra-specs: ${{ matrix.deps.spec }}
-          channels: conda-forge
+          create-args: ${{ matrix.deps.spec }}
+          condarc: |
+            channels:
+              - conda-forge
       - name: Install
         run: |
           pip install --no-deps .
@@ -81,10 +82,10 @@ jobs:
         uses: actions/checkout@v3
         with:
           fetch-depth: 0
-      - name: Set up Python 3.10
+      - name: Set up Python 3.11
         uses: actions/setup-python@v4
         with:
-          python-version: '3.10'
+          python-version: '3.11'
       - name: Set environment variables for docker build
         run: |
           pip install setuptools_scm # Install setuptools_scm to get version number
diff --git a/CHANGELOG.md b/CHANGELOG.md
index c49a750e..47fa5ebf 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,5 +1,22 @@
 # Unreleased
 
+**Changed**
+
+- Split apart OPERA-specific needs from more general library/workflow functionality
+- Removed the final NetCDF product creation
+  - Many rasters in the `scratch/` folder are of general interest after running the workflow
+  - Changed folder structure so that there's not longer a top-level `scratch/` and `output/` by default
+- Changed the required dependencies so the `isce3` unwrapper is optional, as people may wish to implement their own custom parallel unwrapping
+
+**Dependencies**
+
+Dropped:
+- h5netcdf
+- pillow
+
+Now optional:
+- isce3 (for unwrapping)
+
 # [0.3.0](https://github.com/opera-adt/dolphin/compare/v0.2.0...v0.3.0) - 2023-08-23
 
 **Added**
diff --git a/MANIFEST.in b/MANIFEST.in
index cd7184a7..12f0422d 100644
--- a/MANIFEST.in
+++ b/MANIFEST.in
@@ -1,2 +1,3 @@
 include src/dolphin/shp/glrt_cutoffs.csv
 include src/dolphin/shp/kld_cutoffs.csv
+include src/dolphin/py.typed
diff --git a/README.md b/README.md
index 34e1ccd2..ec5fd659 100644
--- a/README.md
+++ b/README.md
@@ -12,11 +12,19 @@ High resolution wrapped phase estimation for InSAR using combined PS/DS processi
 `dolphin` is available on conda:
 
 ```bash
+# if mamba is not already installed: conda install -n base mamba
 mamba install -c conda-forge dolphin
 ```
-
 (Note: [using `mamba`](https://mamba.readthedocs.io/en/latest/mamba-installation.html#mamba-install) is recommended for conda-forge packages, but miniconda can also be used.)
 
+
+`dolphin` has the ability to unwrap interferograms, but requires the optional dependency of `isce3` to use the python bindings to [SNAPHU](https://web.stanford.edu/group/radar/softwareandlinks/sw/snaphu/).
+To install both dolphin and isce3 through conda-forge, run
+```bash
+mamba install -c conda-forge isce3 dolphin
+```
+
+
 To install locally:
 
 1. Download source code:
diff --git a/conda-env-unwrapping.yml b/conda-env-unwrapping.yml
new file mode 100644
index 00000000..51f90777
--- /dev/null
+++ b/conda-env-unwrapping.yml
@@ -0,0 +1,9 @@
+# Can be used to update the environment with the following command:
+#
+#     conda env update --name dolphin-env --file conda-env-unwrapping.yml
+# https://docs.conda.io/projects/conda/en/latest/commands/update.html
+channels:
+  - conda-forge
+dependencies:
+  - isce3>=0.14
+  # - tophu  # Once tophu added to conda forge, this will be installed
diff --git a/conda-env.yml b/conda-env.yml
index 9ca8f469..3abba6d5 100644
--- a/conda-env.yml
+++ b/conda-env.yml
@@ -5,14 +5,12 @@ dependencies:
   - python>=3.8
   - pip>=21.3  # https://pip.pypa.io/en/stable/reference/build-system/pyproject-toml/#editable-installation
   - git  # for pip install, due to setuptools_scm
-  - isce3        # >=0.14.0 # Right now, isce3 is messes up conda's solvers. Should move to optional.
+  # - isce3        # isce3 for unwrapping has been moved to optional
   - gdal>=3.3
   - h5py>=3.6
   - hdf5!=1.12.2 # https://github.com/SciTools/iris/issues/5187 and https://github.com/pydata/xarray/issues/7549
-  - h5netcdf>=1.0
   - numba>=0.54
   - numpy>=1.20
-  - pillow>=7.0
   - pydantic>=2.1
   - pymp-pypi>=0.4.5
   - pyproj>=3.3
diff --git a/docker/create-lockfile.sh b/docker/create-lockfile.sh
index 643a9425..a13541ed 100755
--- a/docker/create-lockfile.sh
+++ b/docker/create-lockfile.sh
@@ -5,42 +5,82 @@ set -o errexit
 set -o nounset
 set -o pipefail
 
-readonly HELP='usage: ./create-lockfile.sh ENVFILE > specfile.txt
+readonly HELP='usage: ./create-lockfile.sh --file ENVFILE [--pkgs PACKAGE ...] > specfile.txt
 
-Create a conda lockfile from an environment YAML file for reproducible
-environments.
-
-positional arguments:
-ENVFILE     a YAML file containing package specifications
+Create a conda lockfile from an environment YAML file and additional packages for reproducible environments.
 
 options:
--h, --help  show this help message and exit
+--file ENVFILE    Specify a YAML file containing package specifications.
+--pkgs PACKAGE    Specify additional packages separated by spaces. Example: --pkgs numpy scipy
+-h, --help        Show this help message and exit
 '
 
-main() {
-    # Get absolute path of input YAML file.
-    local ENVFILE
-    ENVFILE=$(realpath "$1")
+install_packages() {
+    local ENVFILE=$(realpath "$1")
+    shift
+    local PACKAGES="$@"
+
+    # Prepare arguments for the command
+    local FILE_ARG="--file /tmp/$(basename "$ENVFILE")"
+    local PKGS_ARGS=(${PACKAGES[@]})
 
     # Get concretized package list.
     local PKGLIST
-    PKGLIST=$(docker run --network=host \
-        -v "$ENVFILE:/tmp/environment.yml:ro" --rm \
-        mambaorg/micromamba:1.1.0 bash -c '\
-            micromamba install -y -n base -f /tmp/environment.yml > /dev/null && \
-            micromamba env export --explicit')
+    PKGLIST=$(docker run --rm --network=host \
+        -v "$ENVFILE:/tmp/$(basename "$ENVFILE"):ro" \
+        mambaorg/micromamba:1.1.0 bash -c "\
+            micromamba install -y -n base $FILE_ARG ${PKGS_ARGS[*]} > /dev/null && \
+            micromamba env export --explicit")
 
     # Sort packages alphabetically.
     # (The first 4 lines are assumed to be header lines and ignored.)
-    echo "$PKGLIST" | (sed -u 4q; sort)
+    echo "$PKGLIST" | (
+        sed -u 4q
+        sort
+    )
+}
+
+main() {
+    local ENVFILE=""
+    local PACKAGES=()
+
+    while [[ "$#" -gt 0 ]]; do
+        case $1 in
+        --file)
+            shift
+            if [[ -z "${1-}" ]]; then
+                echo "No file provided after --file" >&2
+                exit 1
+            fi
+            ENVFILE="$1"
+            shift
+            ;;
+        --pkgs)
+            shift
+            while [[ "$#" -gt 0 && ! "$1" =~ ^-- ]]; do
+                PACKAGES+=("$1")
+                shift
+            done
+            ;;
+        -h | --help)
+            echo "$HELP"
+            exit 0
+            ;;
+        *)
+            echo "Unknown option: $1" >&2
+            echo "$HELP"
+            exit 1
+            ;;
+        esac
+    done
+
+    if [[ -z "$ENVFILE" ]]; then
+        echo 'No environment file provided' >&2
+        echo "$HELP"
+        exit 1
+    fi
+
+    install_packages "$ENVFILE" "${PACKAGES[@]}"
 }
 
-if [[ "${1-}" =~ ^-*h(elp)?$ ]]; then
-    echo "$HELP"
-elif [[ "$#" -ne 1 ]]; then
-    echo 'Illegal number of parameters' >&2
-    echo "$HELP"
-    exit 1
-else
-    main "$@"
-fi
+main "$@"
diff --git a/docker/specfile.txt b/docker/specfile.txt
index 4fae4e09..6923f494 100644
--- a/docker/specfile.txt
+++ b/docker/specfile.txt
@@ -2,162 +2,168 @@
 # $ conda create --name <env> --file <this file>
 # platform: linux-64
 @EXPLICIT
-https://conda.anaconda.org/conda-forge/linux-64/blosc-1.21.3-hafa529b_0.conda#bcf0664a2dbbbb86cbd4c1e6ff10ddd6
-https://conda.anaconda.org/conda-forge/linux-64/boost-cpp-1.78.0-h5adbc97_2.conda#09be6b4c66c7881e2b24214c6f6841c9
-https://conda.anaconda.org/conda-forge/linux-64/brotlipy-0.7.0-py310h5764c6d_1005.tar.bz2#87669c3468dff637bbd0363bc0f895cf
+https://conda.anaconda.org/conda-forge/linux-64/blosc-1.21.5-h0f2a231_0.conda#009521b7ed97cca25f8f997f9e745976
+https://conda.anaconda.org/conda-forge/linux-64/boost-cpp-1.78.0-h2c5509c_4.conda#417a9d724dc4b651f4a711d3aa3694e3
+https://conda.anaconda.org/conda-forge/linux-64/brotli-python-1.1.0-py311hb755f60_0.conda#b8128d083dbf6abd472b1a3e98b0b83d
 https://conda.anaconda.org/conda-forge/linux-64/bzip2-1.0.8-h7f98852_4.tar.bz2#a1fd65c7ccbf10880423d82bca54eb54
-https://conda.anaconda.org/conda-forge/linux-64/ca-certificates-2022.12.7-ha878542_0.conda#ff9f73d45c4a07d6f424495288a26080
-https://conda.anaconda.org/conda-forge/linux-64/cairo-1.16.0-ha61ee94_1014.tar.bz2#d1a88f3ed5b52e1024b80d4bcd26a7a0
-https://conda.anaconda.org/conda-forge/linux-64/c-ares-1.18.1-h7f98852_0.tar.bz2#f26ef8098fab1f719c91eb760d63381a
-https://conda.anaconda.org/conda-forge/linux-64/cffi-1.15.1-py310h255011f_3.conda#800596144bb613cd7ac58b80900ce835
-https://conda.anaconda.org/conda-forge/linux-64/cfitsio-4.1.0-hd9d235c_0.tar.bz2#ebc04a148d7204bb428f8633b89fd3dd
-https://conda.anaconda.org/conda-forge/linux-64/cryptography-40.0.2-py310h34c0648_0.conda#991a12eccbca3c9897c62f44b1104a54
-https://conda.anaconda.org/conda-forge/linux-64/curl-8.0.1-h588be90_0.conda#69691e828381dd12df671c26b680f1b0
+https://conda.anaconda.org/conda-forge/linux-64/ca-certificates-2023.7.22-hbcca054_0.conda#a73ecd2988327ad4c8f2c331482917f2
+https://conda.anaconda.org/conda-forge/linux-64/cairo-1.16.0-h0c91306_1017.conda#3db543896d34fc6804ddfb9239dcb125
+https://conda.anaconda.org/conda-forge/linux-64/c-ares-1.19.1-hd590300_0.conda#e8c18d865be43e2fb3f7a145b6adf1f5
+https://conda.anaconda.org/conda-forge/linux-64/cfitsio-4.3.0-hbdc6101_0.conda#797554b8b7603011e8677884381fbcc5
+https://conda.anaconda.org/conda-forge/linux-64/curl-8.2.1-hca28451_0.conda#b7bf35457c5495009392c17feec4fddd
 https://conda.anaconda.org/conda-forge/linux-64/expat-2.5.0-hcb278e6_1.conda#8b9b5aca60558d02ddaa09d599e55920
-https://conda.anaconda.org/conda-forge/linux-64/fftw-3.3.10-nompi_hc118613_107.conda#28b2b46b350ddb6a01d061392f75af54
+https://conda.anaconda.org/conda-forge/linux-64/fftw-3.3.10-nompi_hc118613_108.conda#6fa90698000b05dfe8ce6515794fe71a
 https://conda.anaconda.org/conda-forge/linux-64/fontconfig-2.14.2-h14ed4e7_0.conda#0f69b688f52ff6da70bccb7ff7001d1d
 https://conda.anaconda.org/conda-forge/linux-64/freetype-2.12.1-hca18f0e_1.conda#e1232042de76d24539a436d37597eb06
 https://conda.anaconda.org/conda-forge/linux-64/freexl-1.0.6-h166bdaf_1.tar.bz2#897e772a157faf3330d72dd291486f62
-https://conda.anaconda.org/conda-forge/linux-64/gdal-3.5.1-py310h8172e47_1.tar.bz2#e3e8495e882625fc0679d321ed383134
-https://conda.anaconda.org/conda-forge/linux-64/geos-3.11.0-h27087fc_0.tar.bz2#a583d0bc9a85c48e8b07a588d1ac8a80
-https://conda.anaconda.org/conda-forge/linux-64/geotiff-1.7.1-h4fc65e6_3.tar.bz2#dc47fc3ed22615992f89effadd512988
+https://conda.anaconda.org/conda-forge/linux-64/gdal-3.7.1-py311h815a124_9.conda#e026f17deff5512eeb5119b0e6ba9103
+https://conda.anaconda.org/conda-forge/linux-64/geos-3.12.0-h59595ed_0.conda#3fdf79ef322c8379ae83be491d805369
+https://conda.anaconda.org/conda-forge/linux-64/geotiff-1.7.1-h22adcc9_11.conda#514167b60f598eaed3f7a60e1dceb9ee
 https://conda.anaconda.org/conda-forge/linux-64/gettext-0.21.1-h27087fc_0.tar.bz2#14947d8770185e5153fdd04d4673ed37
 https://conda.anaconda.org/conda-forge/linux-64/giflib-5.2.1-h0b41bf4_3.conda#96f3b11872ef6fad973eac856cd2624f
-https://conda.anaconda.org/conda-forge/linux-64/git-2.40.1-pl5321h86e50cf_0.conda#0cb5ff348eb4c201b3b920eff851675d
-https://conda.anaconda.org/conda-forge/linux-64/h5py-3.7.0-nompi_py310h06dffec_100.tar.bz2#cc5a27c23e1c2a6fd608846bb09239a8
-https://conda.anaconda.org/conda-forge/linux-64/hdf4-4.2.15-h9772cbc_5.tar.bz2#ee08782aff2ff9b3291c967fa6bc7336
-https://conda.anaconda.org/conda-forge/linux-64/hdf5-1.12.1-nompi_h4df4325_104.tar.bz2#ddecfd15c2a412c595c68ff22d3557a0
-https://conda.anaconda.org/conda-forge/linux-64/icu-70.1-h27087fc_0.tar.bz2#87473a15119779e021c314249d4b4aed
-https://conda.anaconda.org/conda-forge/linux-64/isce3-0.8.0-py310hc48c5ce_2.tar.bz2#a622fa5f725fc6e2cb762b157ba83ddb
-https://conda.anaconda.org/conda-forge/linux-64/jpeg-9e-h0b41bf4_3.conda#c7a069243e1fbe9a556ed2ec030e6407
-https://conda.anaconda.org/conda-forge/linux-64/json-c-0.16-hc379101_0.tar.bz2#0e2bca6857cb73acec30387fef7c3142
-https://conda.anaconda.org/conda-forge/linux-64/kealib-1.4.15-hfe1a663_0.tar.bz2#32fda93b8db4b6be630d1ee6a94f23b9
+https://conda.anaconda.org/conda-forge/linux-64/git-2.42.0-pl5321h86e50cf_0.conda#96ad24c67e0056d171385859c43218a2
+https://conda.anaconda.org/conda-forge/linux-64/gtest-1.14.0-h00ab1b0_1.conda#d362a81b815334cc921b9362782881f3
+https://conda.anaconda.org/conda-forge/linux-64/h5py-3.9.0-nompi_py311h3839ddf_102.conda#8d9855dc6328f3568740ee1e9414f200
+https://conda.anaconda.org/conda-forge/linux-64/hdf4-4.2.15-h501b40f_6.conda#c3e9338e15d90106f467377017352b97
+https://conda.anaconda.org/conda-forge/linux-64/hdf5-1.14.2-nompi_h4f84152_100.conda#2de6a9bc8083b49f09b2f6eb28d3ba3c
+https://conda.anaconda.org/conda-forge/linux-64/icu-73.2-h59595ed_0.conda#cc47e1facc155f91abd89b11e48e72ff
+https://conda.anaconda.org/conda-forge/linux-64/isce3-0.14.0-py311h360d1b0_1.conda#6d099fec840e5a9c37b71f38b12c27ba
+https://conda.anaconda.org/conda-forge/linux-64/json-c-0.17-h7ab15ed_0.conda#9961b1f100c3b6852bd97c9233d06979
+https://conda.anaconda.org/conda-forge/linux-64/kealib-1.5.1-hcd42e92_5.conda#d871720bf750347506062ba23a91662d
 https://conda.anaconda.org/conda-forge/linux-64/keyutils-1.6.1-h166bdaf_0.tar.bz2#30186d27e2c9fa62b45fb1476b7200e3
-https://conda.anaconda.org/conda-forge/linux-64/krb5-1.20.1-h81ceb04_0.conda#89a41adce7106749573d883b2f657d78
-https://conda.anaconda.org/conda-forge/linux-64/lcms2-2.14-h6ed2654_0.tar.bz2#dcc588839de1445d90995a0a2c4f3a39
+https://conda.anaconda.org/conda-forge/linux-64/krb5-1.21.2-h659d440_0.conda#cd95826dbd331ed1be26bdf401432844
+https://conda.anaconda.org/conda-forge/linux-64/lcms2-2.15-haa2dc70_1.conda#980d8aca0bc23ca73fa8caa3e7c84c28
 https://conda.anaconda.org/conda-forge/linux-64/ld_impl_linux-64-2.40-h41732ed_0.conda#7aca3059a1729aa76c597603f10b0dd3
 https://conda.anaconda.org/conda-forge/linux-64/lerc-4.0.0-h27087fc_0.tar.bz2#76bbff344f0134279f225174e9064c8f
-https://conda.anaconda.org/conda-forge/linux-64/libblas-3.9.0-16_linux64_openblas.tar.bz2#d9b7a8639171f6c6fa0a983edabcfe2b
-https://conda.anaconda.org/conda-forge/linux-64/libcblas-3.9.0-16_linux64_openblas.tar.bz2#20bae26d0a1db73f758fc3754cab4719
-https://conda.anaconda.org/conda-forge/linux-64/libcurl-8.0.1-h588be90_0.conda#b635278a73eb67edcfba7d01a6b48a03
-https://conda.anaconda.org/conda-forge/linux-64/libdap4-3.20.6-hd7c4107_2.tar.bz2#c265ae57e3acdc891f3e2b93cf6784f5
-https://conda.anaconda.org/conda-forge/linux-64/libdeflate-1.14-h166bdaf_0.tar.bz2#fc84a0446e4e4fb882e78d786cfb9734
+https://conda.anaconda.org/conda-forge/linux-64/libabseil-20230125.3-cxx17_h59595ed_0.conda#d1db1b8be7c3a8983dcbbbfe4f0765de
+https://conda.anaconda.org/conda-forge/linux-64/libaec-1.0.6-hcb278e6_1.conda#0f683578378cddb223e7fd24f785ab2a
+https://conda.anaconda.org/conda-forge/linux-64/libarchive-3.6.2-h039dbb9_1.conda#29cf970521d30d113f3425b84cb250f6
+https://conda.anaconda.org/conda-forge/linux-64/libblas-3.9.0-18_linux64_openblas.conda#bcddbb497582ece559465b9cd11042e7
+https://conda.anaconda.org/conda-forge/linux-64/libcblas-3.9.0-18_linux64_openblas.conda#93dd9ab275ad888ed8113953769af78c
+https://conda.anaconda.org/conda-forge/linux-64/libcrc32c-1.1.2-h9c3ff4c_0.tar.bz2#c965a5aa0d5c1c37ffc62dff36e28400
+https://conda.anaconda.org/conda-forge/linux-64/libcurl-8.2.1-hca28451_0.conda#96aec6156d58591f5a4e67056521ce1b
+https://conda.anaconda.org/conda-forge/linux-64/libdeflate-1.18-h0b41bf4_0.conda#6aa9c9de5542ecb07fdda9ca626252d8
 https://conda.anaconda.org/conda-forge/linux-64/libedit-3.1.20191231-he28a2e2_2.tar.bz2#4d331e44109e3f0e19b4cb8f9b82f3e1
 https://conda.anaconda.org/conda-forge/linux-64/libev-4.33-h516909a_1.tar.bz2#6f8720dff19e17ce5d48cfe7f3d2f0a3
 https://conda.anaconda.org/conda-forge/linux-64/libexpat-2.5.0-hcb278e6_1.conda#6305a3dd2752c76335295da4e581f2fd
 https://conda.anaconda.org/conda-forge/linux-64/libffi-3.4.2-h7f98852_5.tar.bz2#d645c6d2ac96843a2bfaccd2d62b3ac3
 https://conda.anaconda.org/conda-forge/linux-64/_libgcc_mutex-0.1-conda_forge.tar.bz2#d7c89558ba9fa0495403155b64376d81
-https://conda.anaconda.org/conda-forge/linux-64/libgcc-ng-12.2.0-h65d4601_19.tar.bz2#e4c94f80aef025c17ab0828cd85ef535
-https://conda.anaconda.org/conda-forge/linux-64/libgdal-3.5.1-h0c20829_1.tar.bz2#cdf55fbb2010f619d74a9f3e4bc79174
-https://conda.anaconda.org/conda-forge/linux-64/libgfortran5-12.2.0-h337968e_19.tar.bz2#164b4b1acaedc47ee7e658ae6b308ca3
-https://conda.anaconda.org/conda-forge/linux-64/libgfortran-ng-12.2.0-h69a702a_19.tar.bz2#cd7a806282c16e1f2d39a7e80d3a3e0d
-https://conda.anaconda.org/conda-forge/linux-64/libglib-2.76.2-hebfc3b9_0.conda#db1d4a1dfc04f3eab50d97551850759a
-https://conda.anaconda.org/conda-forge/linux-64/libgomp-12.2.0-h65d4601_19.tar.bz2#cedcee7c064c01c403f962c9e8d3c373
+https://conda.anaconda.org/conda-forge/linux-64/libgcc-ng-13.1.0-he5830b7_0.conda#cd93f779ff018dd85c7544c015c9db3c
+https://conda.anaconda.org/conda-forge/linux-64/libgdal-3.7.1-h880a63b_9.conda#6e41df426ad7c3153554297f57b9017d
+https://conda.anaconda.org/conda-forge/linux-64/libgfortran5-13.1.0-h15d22d2_0.conda#afb656a334c409dd9805508af1c89c7a
+https://conda.anaconda.org/conda-forge/linux-64/libgfortran-ng-13.1.0-h69a702a_0.conda#506dc07710dd5b0ba63cbf134897fc10
+https://conda.anaconda.org/conda-forge/linux-64/libglib-2.76.4-hebfc3b9_0.conda#c6f951789c888f7bbd2dd6858eab69de
+https://conda.anaconda.org/conda-forge/linux-64/libgomp-13.1.0-he5830b7_0.conda#56ca14d57ac29a75d23a39eb3ee0ddeb
+https://conda.anaconda.org/conda-forge/linux-64/libgoogle-cloud-2.12.0-h840a212_1.conda#03c225a73835f5aa68c13e62eb360406
+https://conda.anaconda.org/conda-forge/linux-64/libgrpc-1.56.2-h3905398_1.conda#0b01e6ff8002994bd4ddbffcdbec7856
+https://conda.anaconda.org/conda-forge/linux-64/libhwloc-2.9.2-default_h554bfaf_1009.conda#9369f407667517fe52b0e8ed6965ffeb
 https://conda.anaconda.org/conda-forge/linux-64/libiconv-1.17-h166bdaf_0.tar.bz2#b62b52da46c39ee2bc3c162ac7f1804d
+https://conda.anaconda.org/conda-forge/linux-64/libjpeg-turbo-2.1.5.1-h0b41bf4_0.conda#1edd9e67bdb90d78cea97733ff6b54e6
 https://conda.anaconda.org/conda-forge/linux-64/libkml-1.3.0-h37653c0_1015.tar.bz2#37d3747dd24d604f63d2610910576e63
-https://conda.anaconda.org/conda-forge/linux-64/liblapack-3.9.0-16_linux64_openblas.tar.bz2#955d993f41f9354bf753d29864ea20ad
-https://conda.anaconda.org/conda-forge/linux-64/libllvm11-11.1.0-he0ac6c6_5.tar.bz2#cae79c6fd61cc6823cbebdbb2c16c60e
-https://conda.anaconda.org/conda-forge/linux-64/libnetcdf-4.8.1-nompi_h329d8a1_102.tar.bz2#a857af323c5edbd8e9e45fb9facb6f5f
+https://conda.anaconda.org/conda-forge/linux-64/liblapack-3.9.0-18_linux64_openblas.conda#a1244707531e5b143c420c70573c8ec5
+https://conda.anaconda.org/conda-forge/linux-64/libllvm14-14.0.6-hcd5def8_4.conda#73301c133ded2bf71906aa2104edae8b
+https://conda.anaconda.org/conda-forge/linux-64/libnetcdf-4.9.2-nompi_h80fb2b6_112.conda#a19fa6cacf80c8a366572853d5890eb4
 https://conda.anaconda.org/conda-forge/linux-64/libnghttp2-1.52.0-h61bc06f_0.conda#613955a50485812985c059e7b269f42e
 https://conda.anaconda.org/conda-forge/linux-64/libnsl-2.0.0-h7f98852_0.tar.bz2#39b1328babf85c7c3a61636d9cd50206
-https://conda.anaconda.org/conda-forge/linux-64/libopenblas-0.3.21-pthreads_h78a6416_3.tar.bz2#8c5963a49b6035c40646a763293fbb35
+https://conda.anaconda.org/conda-forge/linux-64/libopenblas-0.3.24-pthreads_h413a1c8_0.conda#6e4ef6ca28655124dcde9bd500e44c32
 https://conda.anaconda.org/conda-forge/linux-64/libpng-1.6.39-h753d276_0.conda#e1c890aebdebbfbf87e2c917187b4416
-https://conda.anaconda.org/conda-forge/linux-64/libpq-14.5-hb675445_5.conda#2fcfbd27d237032b4d74144454b75064
-https://conda.anaconda.org/conda-forge/linux-64/librttopo-1.1.0-hf730bdb_11.tar.bz2#13b2138ccfbd9821fe13ee0cb7471c03
-https://conda.anaconda.org/conda-forge/linux-64/libspatialite-5.0.1-hd36657c_19.tar.bz2#d7f5e67dd5cd58da7b78489183e96070
-https://conda.anaconda.org/conda-forge/linux-64/libsqlite-3.40.0-h753d276_1.conda#32565f92ca100184c67509d1d91c858a
-https://conda.anaconda.org/conda-forge/linux-64/libssh2-1.10.0-hf14f497_3.tar.bz2#d85acad4b47dff4e3def14a769a97906
-https://conda.anaconda.org/conda-forge/linux-64/libstdcxx-ng-12.2.0-h46fd767_19.tar.bz2#1030b1f38c129f2634eae026f704fe60
-https://conda.anaconda.org/conda-forge/linux-64/libtiff-4.4.0-h82bc61c_5.conda#e712a63a21f9db647982971dc121cdcf
+https://conda.anaconda.org/conda-forge/linux-64/libpq-15.4-hfc447b1_0.conda#b9ce311e7aba8b5fc3122254f0a6e97e
+https://conda.anaconda.org/conda-forge/linux-64/libprotobuf-4.23.3-hd1fb520_1.conda#78c10e8637a6f8d377f9989327d0267d
+https://conda.anaconda.org/conda-forge/linux-64/librttopo-1.1.0-hb58d41b_14.conda#264f9a3a4ea52c8f4d3e8ae1213a3335
+https://conda.anaconda.org/conda-forge/linux-64/libspatialite-5.0.1-h15f6e67_28.conda#bc9758e23157cb8362e60d3de06aa6fb
+https://conda.anaconda.org/conda-forge/linux-64/libsqlite-3.43.0-h2797004_0.conda#903fa782a9067d5934210df6d79220f6
+https://conda.anaconda.org/conda-forge/linux-64/libssh2-1.11.0-h0841786_0.conda#1f5a58e686b13bcfde88b93f547d23fe
+https://conda.anaconda.org/conda-forge/linux-64/libstdcxx-ng-13.1.0-hfd8a6a1_0.conda#067bcc23164642f4c226da631f2a2e1d
+https://conda.anaconda.org/conda-forge/linux-64/libtiff-4.5.1-h8b53f26_1.conda#5b09e13d732dda1a2bc9adc711164f4d
 https://conda.anaconda.org/conda-forge/linux-64/libuuid-2.38.1-h0b41bf4_0.conda#40b61aab5c7ba9ff276c41cfffe6b80b
-https://conda.anaconda.org/conda-forge/linux-64/libwebp-base-1.3.0-h0b41bf4_0.conda#0d4a7508d8c6c65314f2b9c1f56ad408
-https://conda.anaconda.org/conda-forge/linux-64/libxcb-1.13-h7f98852_1004.tar.bz2#b3653fdc58d03face9724f602218a904
-https://conda.anaconda.org/conda-forge/linux-64/libxml2-2.10.3-hca2bb57_4.conda#bb808b654bdc3c783deaf107a2ffb503
-https://conda.anaconda.org/conda-forge/linux-64/libzip-1.9.2-hc929e4a_1.tar.bz2#5b122b50e738c4be5c3f2899f010d7cf
-https://conda.anaconda.org/conda-forge/linux-64/libzlib-1.2.13-h166bdaf_4.tar.bz2#f3f9de449d32ca9b9c66a22863c96f41
-https://conda.anaconda.org/conda-forge/linux-64/llvmlite-0.39.1-py310h58363a5_1.tar.bz2#18c28e036ae9c5366c56ecd0c875a108
+https://conda.anaconda.org/conda-forge/linux-64/libwebp-base-1.3.1-hd590300_0.conda#82bf6f63eb15ef719b556b63feec3a77
+https://conda.anaconda.org/conda-forge/linux-64/libxcb-1.15-h0b41bf4_0.conda#33277193f5b92bad9fdd230eb700929c
+https://conda.anaconda.org/conda-forge/linux-64/libxml2-2.11.5-h232c23b_1.conda#f3858448893839820d4bcfb14ad3ecdf
+https://conda.anaconda.org/conda-forge/linux-64/libzip-1.10.1-h2629f0a_2.conda#a83ad320127e83ae8a86b3db8dfeec77
+https://conda.anaconda.org/conda-forge/linux-64/libzlib-1.2.13-hd590300_5.conda#f36c115f1ee199da648e0597ec2047ad
+https://conda.anaconda.org/conda-forge/linux-64/llvmlite-0.40.1-py311ha6695c7_0.conda#7a2b62d839516ba0cf56717e902229f4
 https://conda.anaconda.org/conda-forge/linux-64/lz4-c-1.9.4-hcb278e6_0.conda#318b08df404f9c9be5712aaa5a6f0bb0
-https://conda.anaconda.org/conda-forge/linux-64/ncurses-6.3-h27087fc_1.tar.bz2#4acfc691e64342b9dae57cf2adc63238
+https://conda.anaconda.org/conda-forge/linux-64/lzo-2.10-h516909a_1000.tar.bz2#bb14fcb13341b81d5eb386423b9d2bac
+https://conda.anaconda.org/conda-forge/linux-64/ncurses-6.4-hcb278e6_0.conda#681105bccc2a3f7f1a837d47d39c9179
 https://conda.anaconda.org/conda-forge/linux-64/nspr-4.35-h27087fc_0.conda#da0ec11a6454ae19bff5b02ed881a2b1
-https://conda.anaconda.org/conda-forge/linux-64/nss-3.89-he45b914_0.conda#2745719a58eeaab6657256a3f142f099
-https://conda.anaconda.org/conda-forge/linux-64/numba-0.56.4-py310h0e39c9b_1.conda#fe57b83f32db6d332342db71571f501b
-https://conda.anaconda.org/conda-forge/linux-64/numpy-1.23.5-py310h53a5b5f_0.conda#3b114b1559def8bad228fec544ac1812
-https://conda.anaconda.org/conda-forge/linux-64/openjpeg-2.5.0-h7d73246_1.tar.bz2#a11b4df9271a8d7917686725aa04c8f2
+https://conda.anaconda.org/conda-forge/linux-64/nss-3.92-h1d7d5a4_0.conda#22c89a3d87828fe925b310b9cdf0f574
+https://conda.anaconda.org/conda-forge/linux-64/numba-0.57.1-py311h96b013e_0.conda#618010d18c4a38073a7f51d9dd3fd8a8
+https://conda.anaconda.org/conda-forge/linux-64/numpy-1.24.4-py311h64a7726_0.conda#5a03d7c75dd4a9ae9a58850860eca468
+https://conda.anaconda.org/conda-forge/linux-64/openjpeg-2.5.0-hfec8fc6_2.conda#5ce6a42505c6e9e6151c54c3ec8d68ea
 https://conda.anaconda.org/conda-forge/linux-64/_openmp_mutex-4.5-2_gnu.tar.bz2#73aaf86a425cc6e73fcf236a5a46396d
-https://conda.anaconda.org/conda-forge/linux-64/openssl-3.1.0-hd590300_3.conda#8f24d371ed9efb3f0b0de383fb81d51c
+https://conda.anaconda.org/conda-forge/linux-64/openssl-3.1.2-hd590300_0.conda#e5ac5227582d6c83ccf247288c0eb095
 https://conda.anaconda.org/conda-forge/linux-64/pcre2-10.40-hc3806b6_0.tar.bz2#69e2c796349cd9b273890bee0febfe1b
-https://conda.anaconda.org/conda-forge/linux-64/pcre-8.45-h9c3ff4c_0.tar.bz2#c05d1820a6d34ff07aaaab7a9b7eddaa
-https://conda.anaconda.org/conda-forge/linux-64/perl-5.32.1-2_h7f98852_perl5.tar.bz2#09ba115862623f00962e9809ea248f1a
-https://conda.anaconda.org/conda-forge/linux-64/pillow-9.2.0-py310h454ad03_3.tar.bz2#eb354ff791f505b1d6f13f776359d88e
+https://conda.anaconda.org/conda-forge/linux-64/perl-5.32.1-4_hd590300_perl5.conda#3e785bff761095eb7f8676f4694bd1b1
 https://conda.anaconda.org/conda-forge/linux-64/pixman-0.40.0-h36c2ea0_0.tar.bz2#660e72c82f2e75a6b3fe6a6e75c79f19
-https://conda.anaconda.org/conda-forge/linux-64/poppler-22.04.0-h0733791_3.tar.bz2#74fe2fe238e0cdb31ec61427527a80d7
-https://conda.anaconda.org/conda-forge/linux-64/postgresql-14.5-h3248436_5.conda#eed1ea87210eec1572e7415c1029f04b
-https://conda.anaconda.org/conda-forge/linux-64/proj-9.0.1-h93bde94_1.tar.bz2#8259528ea471b0963a91ce174f002e55
+https://conda.anaconda.org/conda-forge/linux-64/poppler-23.08.0-hd18248d_0.conda#59a093146aa911da2ca056c1197e3e41
+https://conda.anaconda.org/conda-forge/linux-64/postgresql-15.4-h8972f4a_0.conda#bf6169ef6f83cc04d8b2a72cd5c364bc
+https://conda.anaconda.org/conda-forge/linux-64/proj-9.2.1-ha643af7_0.conda#e992387307f4403ba0ec07d009032550
 https://conda.anaconda.org/conda-forge/linux-64/pthread-stubs-0.4-h36c2ea0_1001.tar.bz2#22dad4df6e8630e8dff2428f6f6a7036
-https://conda.anaconda.org/conda-forge/linux-64/pydantic-1.10.7-py310h1fa729e_0.conda#6306ca76bc0635d84940349cf8d96264
-https://conda.anaconda.org/conda-forge/linux-64/pyproj-3.4.0-py310hf94497c_0.tar.bz2#d034e57d18be11a94f4922e31539a4a5
-https://conda.anaconda.org/conda-forge/linux-64/pyre-1.11.2-py310hdfe83f4_1.tar.bz2#b681b5319466bb09e4a6dea69d6aea00
-https://conda.anaconda.org/conda-forge/linux-64/python-3.10.10-he550d4f_0_cpython.conda#de25afc7041c103c7f510c746bb63435
-https://conda.anaconda.org/conda-forge/linux-64/python_abi-3.10-3_cp310.conda#4eb33d14d794b0f4be116443ffed3853
-https://conda.anaconda.org/conda-forge/linux-64/pyyaml-6.0-py310h5764c6d_5.tar.bz2#9e68d2ff6d98737c855b65f48dd3c597
+https://conda.anaconda.org/conda-forge/linux-64/pydantic-core-2.6.3-py311h46250e7_0.conda#cc8b1e7eab870b5e2a01440d585d2f3d
+https://conda.anaconda.org/conda-forge/linux-64/pyproj-3.6.0-py311ha169711_1.conda#92633556d37e88ce45193374d408072c
+https://conda.anaconda.org/conda-forge/linux-64/pyre-1.12.1-py311h6b0c3e6_2.conda#917da9c839a1ddba516809a613ed19e2
+https://conda.anaconda.org/conda-forge/linux-64/python-3.11.5-hab00c5b_0_cpython.conda#f0288cb82594b1cbc71111d1cd3c5422
+https://conda.anaconda.org/conda-forge/linux-64/python_abi-3.11-3_cp311.conda#c2e2630ddb68cf52eec74dc7dfab20b5
+https://conda.anaconda.org/conda-forge/linux-64/pyyaml-6.0.1-py311h459d7ec_0.conda#30eaaf31141e785a445bf1ede6235fe3
+https://conda.anaconda.org/conda-forge/linux-64/re2-2023.03.02-h8c504da_0.conda#206f8fa808748f6e90599c3368a1114e
 https://conda.anaconda.org/conda-forge/linux-64/readline-8.2-h8228510_1.conda#47d31b792659ce70f470b5c82fdfb7a4
-https://conda.anaconda.org/conda-forge/linux-64/ruamel.yaml-0.17.22-py310h2372a71_0.conda#315a6bdce667bf9673b18c7951da15b6
-https://conda.anaconda.org/conda-forge/linux-64/ruamel.yaml.clib-0.2.7-py310h1fa729e_1.conda#2f9b517412af46255cef5e53a22c264e
-https://conda.anaconda.org/conda-forge/linux-64/scipy-1.10.1-py310h8deb116_0.conda#4c9604c5ec179c21f8f0a09e3c164480
-https://conda.anaconda.org/conda-forge/linux-64/shapely-1.8.5-py310h5e49deb_1.tar.bz2#e5519576751b59d67164b965a4eb4406
+https://conda.anaconda.org/conda-forge/linux-64/ruamel.yaml-0.17.32-py311h459d7ec_0.conda#628868dc17f9bd39a2eb77846e35980c
+https://conda.anaconda.org/conda-forge/linux-64/ruamel.yaml.clib-0.2.7-py311h2582759_1.conda#5e997292429a22ad50c11af0a2cb0f08
+https://conda.anaconda.org/conda-forge/linux-64/scipy-1.11.2-py311h64a7726_0.conda#18d094fb8e4ac52f93a4f4857a8f1e8f
+https://conda.anaconda.org/conda-forge/linux-64/shapely-2.0.1-py311he06c224_2.conda#10a1953d2f74d292b5de093ceea104b2
 https://conda.anaconda.org/conda-forge/linux-64/snappy-1.1.10-h9fff704_0.conda#e6d228cd0bb74a51dd18f5bfce0b4115
-https://conda.anaconda.org/conda-forge/linux-64/sqlite-3.40.0-h4ff8645_1.conda#a5cec05bd9e0a5843fa29de9591b93cb
-https://conda.anaconda.org/conda-forge/linux-64/tiledb-2.9.5-h3f4058f_0.tar.bz2#4128510b35ed578b909ea68234c9330f
+https://conda.anaconda.org/conda-forge/linux-64/sqlite-3.43.0-h2c6b66d_0.conda#713f9eac95d051abe14c3774376854fe
+https://conda.anaconda.org/conda-forge/linux-64/tbb-2021.10.0-h00ab1b0_0.conda#9c82b1b389e46b64ec685ec487043e70
+https://conda.anaconda.org/conda-forge/linux-64/tiledb-2.16.3-h84d19f0_1.conda#1cc4e61dc7ca15a570e733a6e20a7b33
 https://conda.anaconda.org/conda-forge/linux-64/tk-8.6.12-h27826a3_0.tar.bz2#5b8c42eb62e9fc961af70bdd6a26e168
 https://conda.anaconda.org/conda-forge/linux-64/tzcode-2023c-h0b41bf4_0.conda#0c0533894f21c3d35697cb8378d390e2
-https://conda.anaconda.org/conda-forge/linux-64/xerces-c-3.2.4-h55805fa_1.tar.bz2#d127dc8efe24033b306180939e51e6af
+https://conda.anaconda.org/conda-forge/linux-64/xerces-c-3.2.4-hac6953d_3.conda#297e6a75dc1b6a440cd341a85eab8a00
 https://conda.anaconda.org/conda-forge/linux-64/xorg-kbproto-1.0.7-h7f98852_1002.tar.bz2#4b230e8381279d76131116660f5a241a
-https://conda.anaconda.org/conda-forge/linux-64/xorg-libice-1.0.10-h7f98852_0.tar.bz2#d6b0b50b49eccfe0be0373be628be0f3
-https://conda.anaconda.org/conda-forge/linux-64/xorg-libsm-1.2.3-hd9c2040_1000.tar.bz2#9e856f78d5c80d5a78f61e72d1d473a3
-https://conda.anaconda.org/conda-forge/linux-64/xorg-libx11-1.8.4-h0b41bf4_0.conda#ea8fbfeb976ac49cbeb594e985393514
-https://conda.anaconda.org/conda-forge/linux-64/xorg-libxau-1.0.9-h7f98852_0.tar.bz2#bf6f803a544f26ebbdc3bfff272eb179
+https://conda.anaconda.org/conda-forge/linux-64/xorg-libice-1.1.1-hd590300_0.conda#b462a33c0be1421532f28bfe8f4a7514
+https://conda.anaconda.org/conda-forge/linux-64/xorg-libsm-1.2.4-h7391055_0.conda#93ee23f12bc2e684548181256edd2cf6
+https://conda.anaconda.org/conda-forge/linux-64/xorg-libx11-1.8.6-h8ee46fc_0.conda#7590b76c3d11d21caa44f3fc38ac584a
+https://conda.anaconda.org/conda-forge/linux-64/xorg-libxau-1.0.11-hd590300_0.conda#2c80dc38fface310c9bd81b17037fee5
 https://conda.anaconda.org/conda-forge/linux-64/xorg-libxdmcp-1.1.3-h7f98852_0.tar.bz2#be93aabceefa2fac576e971aef407908
 https://conda.anaconda.org/conda-forge/linux-64/xorg-libxext-1.3.4-h0b41bf4_2.conda#82b6df12252e6f32402b96dacc656fec
-https://conda.anaconda.org/conda-forge/linux-64/xorg-libxrender-0.9.10-h7f98852_1003.tar.bz2#f59c1242cc1dd93e72c2ee2b360979eb
+https://conda.anaconda.org/conda-forge/linux-64/xorg-libxrender-0.9.11-hd590300_0.conda#ed67c36f215b310412b2af935bf3e530
 https://conda.anaconda.org/conda-forge/linux-64/xorg-renderproto-0.11.1-h7f98852_1002.tar.bz2#06feff3d2634e3097ce2fe681474b534
 https://conda.anaconda.org/conda-forge/linux-64/xorg-xextproto-7.3.0-h0b41bf4_1003.conda#bce9f945da8ad2ae9b1d7165a64d0f87
 https://conda.anaconda.org/conda-forge/linux-64/xorg-xproto-7.0.31-h7f98852_1007.tar.bz2#b4a4381d54784606820704f7b5f05a15
 https://conda.anaconda.org/conda-forge/linux-64/xz-5.2.6-h166bdaf_0.tar.bz2#2161070d867d1b1204ea749c8eec4ef0
 https://conda.anaconda.org/conda-forge/linux-64/yaml-0.2.5-h7f98852_2.tar.bz2#4cb3ad778ec2d5a7acbdf254eb1c42ae
-https://conda.anaconda.org/conda-forge/linux-64/zlib-1.2.13-h166bdaf_4.tar.bz2#4b11e365c0275b808be78b30f904e295
-https://conda.anaconda.org/conda-forge/linux-64/zstd-1.5.2-h3eb15da_6.conda#6b63daed8feeca47be78f323e793d555
+https://conda.anaconda.org/conda-forge/linux-64/zlib-1.2.13-hd590300_5.conda#68c34ec6149623be41a1933ab996a209
+https://conda.anaconda.org/conda-forge/linux-64/zstd-1.5.5-hfc55251_0.conda#04b88013080254850d6c01ed54810589
+https://conda.anaconda.org/conda-forge/noarch/annotated-types-0.5.0-pyhd8ed1ab_0.conda#578ae086f225bc2380c79f3b551ff2f7
 https://conda.anaconda.org/conda-forge/noarch/backoff-2.2.1-pyhd8ed1ab_0.tar.bz2#4600709bd85664d8606ae0c76642f8db
 https://conda.anaconda.org/conda-forge/noarch/cached-property-1.5.2-hd8ed1ab_1.tar.bz2#9b347a7ec10940d3f7941ff6c460b551
 https://conda.anaconda.org/conda-forge/noarch/cached_property-1.5.2-pyha770c72_1.tar.bz2#576d629e47797577ab0f1b351297ef4a
-https://conda.anaconda.org/conda-forge/noarch/certifi-2022.12.7-pyhd8ed1ab_0.conda#fb9addc3db06e56abe03e0e9f21a63e6
-https://conda.anaconda.org/conda-forge/noarch/charset-normalizer-3.1.0-pyhd8ed1ab_0.conda#7fcff9f6f123696e940bda77bd4d6551
+https://conda.anaconda.org/conda-forge/noarch/certifi-2023.7.22-pyhd8ed1ab_0.conda#7f3dbc9179b4dde7da98dfb151d0ad22
+https://conda.anaconda.org/conda-forge/noarch/charset-normalizer-3.2.0-pyhd8ed1ab_0.conda#313516e9a4b08b12dfb1e1cd390a96e3
 https://conda.anaconda.org/conda-forge/noarch/fonts-conda-ecosystem-1-0.tar.bz2#fee5683a3f04bd15cbd8318b096a27ab
 https://conda.anaconda.org/conda-forge/noarch/fonts-conda-forge-1-0.tar.bz2#f766549260d6815b0c52253f1fb1bb29
 https://conda.anaconda.org/conda-forge/noarch/font-ttf-dejavu-sans-mono-2.37-hab24e00_0.tar.bz2#0c96522c6bdaed4b1566d11387caaf45
 https://conda.anaconda.org/conda-forge/noarch/font-ttf-inconsolata-3.000-h77eed37_0.tar.bz2#34893075a5c9e55cdafac56607368fc6
 https://conda.anaconda.org/conda-forge/noarch/font-ttf-source-code-pro-2.038-h77eed37_0.tar.bz2#4d59c254e01d9cde7957100457e2d5fb
 https://conda.anaconda.org/conda-forge/noarch/font-ttf-ubuntu-0.83-hab24e00_0.tar.bz2#19410c3df09dfb12d1206132a1d357c5
-https://conda.anaconda.org/conda-forge/noarch/h5netcdf-1.1.0-pyhd8ed1ab_1.conda#1e64e16b588612c98491cfb09bb24c35
 https://conda.anaconda.org/conda-forge/noarch/idna-3.4-pyhd8ed1ab_0.tar.bz2#34272b248891bddccc64479f9a7fffed
-https://conda.anaconda.org/conda-forge/noarch/markdown-it-py-2.2.0-pyhd8ed1ab_0.conda#b2928a6c6d52d7e3562b4a59c3214e3a
+https://conda.anaconda.org/conda-forge/noarch/markdown-it-py-3.0.0-pyhd8ed1ab_0.conda#93a8e71256479c62074356ef6ebf501b
 https://conda.anaconda.org/conda-forge/noarch/mdurl-0.1.0-pyhd8ed1ab_0.tar.bz2#f8dab71fdc13b1bf29a01248b156d268
 https://conda.anaconda.org/conda-forge/noarch/packaging-23.1-pyhd8ed1ab_0.conda#91cda59e66e1e4afe9476f8ef98f5c30
-https://conda.anaconda.org/conda-forge/noarch/pip-23.1.2-pyhd8ed1ab_0.conda#7288da0d36821349cf1126e8670292df
-https://conda.anaconda.org/conda-forge/noarch/platformdirs-3.5.0-pyhd8ed1ab_0.conda#6c36f1c42dd0069b7f23acc74f19be46
+https://conda.anaconda.org/conda-forge/noarch/pip-23.2.1-pyhd8ed1ab_0.conda#e2783aa3f9235225eec92f9081c5b801
+https://conda.anaconda.org/conda-forge/noarch/platformdirs-3.10.0-pyhd8ed1ab_0.conda#0809187ef9b89a3d94a5c24d13936236
 https://conda.anaconda.org/conda-forge/noarch/pooch-1.7.0-pyha770c72_3.conda#5936894aade8240c867d292aa0d980c6
 https://conda.anaconda.org/conda-forge/noarch/poppler-data-0.4.12-hd8ed1ab_0.conda#d8d7293c5b37f39b2ac32940621c6592
-https://conda.anaconda.org/conda-forge/noarch/pycparser-2.21-pyhd8ed1ab_0.tar.bz2#076becd9e05608f8dc72757d5f3a91ff
-https://conda.anaconda.org/conda-forge/noarch/pygments-2.15.1-pyhd8ed1ab_0.conda#d316679235612869eba305aa7d41d9bf
+https://conda.anaconda.org/conda-forge/noarch/pydantic-2.3.0-pyhd8ed1ab_0.conda#55aaca64695fcebdfa8057c87ed180e7
+https://conda.anaconda.org/conda-forge/noarch/pygments-2.16.1-pyhd8ed1ab_0.conda#40e5cb18165466773619e5c963f00a7b
 https://conda.anaconda.org/conda-forge/noarch/pymp-pypi-0.4.5-pyhd8ed1ab_0.tar.bz2#3d56b4c5a162223b9f20cef1fdc8a89a
-https://conda.anaconda.org/conda-forge/noarch/pyopenssl-23.1.1-pyhd8ed1ab_0.conda#0b34aa3ab7e7ccb1765a03dd9ed29938
 https://conda.anaconda.org/conda-forge/noarch/pysocks-1.7.1-pyha2e5f31_6.tar.bz2#2a7de29fb590ca14b5243c4c812c8025
-https://conda.anaconda.org/conda-forge/noarch/requests-2.29.0-pyhd8ed1ab_0.conda#5fa992d972fbccfc069161805122cb8d
-https://conda.anaconda.org/conda-forge/noarch/rich-13.3.5-pyhd8ed1ab_0.conda#2e40a02ad28e34f26cee2a72042843db
-https://conda.anaconda.org/conda-forge/noarch/setuptools-67.7.2-pyhd8ed1ab_0.conda#3b68bc43ec6baa48f7354a446267eefe
-https://conda.anaconda.org/conda-forge/noarch/threadpoolctl-3.1.0-pyh8a188c0_0.tar.bz2#a2995ee828f65687ac5b1e71a2ab1e0c
-https://conda.anaconda.org/conda-forge/noarch/typing-extensions-4.5.0-hd8ed1ab_0.conda#b3c594fde1a80a1fc3eb9cc4a5dfe392
-https://conda.anaconda.org/conda-forge/noarch/typing_extensions-4.5.0-pyha770c72_0.conda#43e7d9e50261fb11deb76e17d8431aac
+https://conda.anaconda.org/conda-forge/noarch/requests-2.31.0-pyhd8ed1ab_0.conda#a30144e4156cdbb236f99ebb49828f8b
+https://conda.anaconda.org/conda-forge/noarch/rich-13.5.1-pyhd8ed1ab_0.conda#38e7446efa3c8b8a770a0fff862935c0
+https://conda.anaconda.org/conda-forge/noarch/setuptools-68.1.2-pyhd8ed1ab_0.conda#4fe12573bf499ff85a0a364e00cc5c53
+https://conda.anaconda.org/conda-forge/noarch/threadpoolctl-3.2.0-pyha21a80b_0.conda#978d03388b62173b8e6f79162cf52b86
+https://conda.anaconda.org/conda-forge/noarch/typing-extensions-4.7.1-hd8ed1ab_0.conda#f96688577f1faa58096d06a45136afa2
+https://conda.anaconda.org/conda-forge/noarch/typing_extensions-4.7.1-pyha770c72_0.conda#c39d6a09fe819de4951c2642629d9115
 https://conda.anaconda.org/conda-forge/noarch/tzdata-2023c-h71feb2d_0.conda#939e3e74d8be4dac89ce83b20de2492a
-https://conda.anaconda.org/conda-forge/noarch/urllib3-1.26.15-pyhd8ed1ab_0.conda#27db656619a55d727eaf5a6ece3d2fd6
-https://conda.anaconda.org/conda-forge/noarch/wheel-0.40.0-pyhd8ed1ab_0.conda#49bb0d9e60ce1db25e151780331bb5f3
+https://conda.anaconda.org/conda-forge/noarch/urllib3-2.0.4-pyhd8ed1ab_0.conda#18badd8fa3648d1beb1fcc7f2e0f756e
+https://conda.anaconda.org/conda-forge/noarch/wheel-0.41.2-pyhd8ed1ab_0.conda#1ccd092478b3e0ee10d7a891adbf8a4f
 https://conda.anaconda.org/conda-forge/noarch/yamale-4.0.4-pyh6c4a22f_0.tar.bz2#cc9f59f147740d88679bf1bd94dbe588
diff --git a/requirements.txt b/requirements.txt
index 70e56090..8a8b08cc 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,9 +1,7 @@
 # These are the packages easily-installable through pip
-h5netcdf>=1.0
 h5py>=3.6
 numba>=0.54
 numpy>=1.20
-pillow>=7.0
 pydantic>=2.1
 pymp-pypi>=0.4.5
 pyproj>=3.3
diff --git a/scripts/benchmarking/benchmark-local.sh b/scripts/benchmarking/benchmark-local.sh
deleted file mode 100755
index 812b9bec..00000000
--- a/scripts/benchmarking/benchmark-local.sh
+++ /dev/null
@@ -1,91 +0,0 @@
-#!/bin/bash
-
-# Enable common error handling options.
-set -o errexit
-set -o nounset
-set -o pipefail
-
-USAGE="Usage: $0 -b|--benchmark-dir [-i|--image <docker image>]"
-
-# Parse input arguments
-while [[ $# -gt 0 ]]; do
-    key="$1"
-    case $key in
-    -i | --image)
-        IMAGE="$2"
-        shift
-        shift
-        ;;
-    -b | --benchmark-dir)
-        BENCH_DIR="$2"
-        shift
-        shift
-        ;;
-    -h | --help)
-        echo "$USAGE"
-        exit 0
-        ;;
-    *)
-        echo "Unknown option: $key"
-        exit 1
-        ;;
-    esac
-done
-
-if [ -z "${BENCH_DIR+x}" ]; then
-    echo "Please specify a benchmark directory."
-    echo "$USAGE"
-    exit 1
-fi
-
-if [ -z "${IMAGE+x}" ]; then
-    IMAGE="opera-adt/dolphin:latest"
-    echo "Using default docker image: $IMAGE"
-fi
-
-# Run the SAS workflow.
-DATA_DIR="$BENCH_DIR/data"
-
-# Check that the data directory exists.
-if [ ! -d "$DATA_DIR" ]; then
-    echo "Data directory does not exist: $DATA_DIR"
-    exit 1
-fi
-# Check we have input_slcs/ and dynamic_ancillary/ config_files/ and
-# config_files/dolphin_config.yaml
-if [ ! -d "$DATA_DIR/input_slcs" ]; then
-    echo "Input SLCs directory does not exist: $DATA_DIR/input_slcs"
-    exit 1
-fi
-if [ ! -d "$DATA_DIR/dynamic_ancillary" ]; then
-    echo "Dynamic ancillary directory does not exist: $DATA_DIR/dynamic_ancillary"
-    exit 1
-fi
-if [ ! -d "$DATA_DIR/config_files" ]; then
-    echo "Config files directory does not exist: $DATA_DIR/config_files"
-    exit 1
-fi
-
-cd $DATA_DIR
-WORKDIR="/tmp"
-mkdir -p scratch output
-rm -rf scratch/* output/*
-docker run --rm --user=$(id -u):$(id -g) \
-    --volume="$(realpath input_slcs):$WORKDIR/input_slcs:ro" \
-    --volume="$(realpath dynamic_ancillary):$WORKDIR/dynamic_ancillary:ro" \
-    --volume="$(realpath config_files/dolphin_config.yaml):$WORKDIR/dolphin_config.yaml:ro" \
-    --volume="$(realpath scratch):$WORKDIR/scratch" \
-    --volume="$(realpath output):$WORKDIR/output" \
-    --workdir="$WORKDIR" \
-    "$IMAGE" dolphin run dolphin_config.yaml
-
-LOG_DIR="$BENCH_DIR/logs"
-mkdir -p $LOG_DIR
-
-# TODO: run this in a docker container.
-eval "$(conda shell.bash hook)"
-source ~/anaconda3/etc/profile.d/conda.sh
-conda activate mapping
-python /u/aurora-r0/staniewi/repos/dolphin/scripts/benchmarking/parse_benchmark_logs.py \
-    --config-files config_files/dolphin_config.yaml \
-    --outfile "$LOG_DIR/benchmark_results_$(date +%Y%m%d).csv"
diff --git a/scripts/benchmarking/parse_benchmark_logs.py b/scripts/benchmarking/parse_benchmark_logs.py
deleted file mode 100755
index 601ac9d6..00000000
--- a/scripts/benchmarking/parse_benchmark_logs.py
+++ /dev/null
@@ -1,134 +0,0 @@
-#!/usr/bin/env python
-from __future__ import annotations
-
-import argparse
-import re
-import sys
-from collections import defaultdict
-from datetime import datetime
-from pathlib import Path
-
-import pandas as pd
-
-from dolphin._types import Filename
-from dolphin.workflows._pge_runconfig import RunConfig
-from dolphin.workflows.config import Workflow
-
-
-def get_df(dolphin_config_file: Filename):
-    """Create a dataframe from a directory of log files."""
-    if Path(dolphin_config_file).name == "runconfig.yaml":
-        rc = RunConfig.from_yaml(dolphin_config_file)
-        alg_file = rc.dynamic_ancillary_file_group.algorithm_parameters_file
-        if not alg_file.is_absolute():
-            alg_file = (
-                Path(dolphin_config_file).parent
-                / rc.dynamic_ancillary_file_group.algorithm_parameters_file
-            )
-        w = rc.to_workflow()
-    else:
-        w = Workflow.from_yaml(dolphin_config_file)
-
-    log_file = Path(w.log_file)
-
-    result = _parse_logfile(log_file)
-    cfg_data = _parse_config(w)
-
-    result.update(cfg_data)
-    return pd.DataFrame([result])
-
-
-def _parse_logfile(logfile: Path) -> dict:
-    out: dict = defaultdict(list)
-    out["file"] = str(Path(logfile).resolve())
-    mempat = r"Maximum memory usage: (\d\.\d{2}) GB"
-    timepat = (
-        r"Total elapsed time for dolphin.workflows.s1_disp.run : (\d*\.\d{2}) minutes"
-        r" \((\d*\.\d{2}) seconds\)"
-    )
-    wrapped_phase_timepat = (
-        r"Total elapsed time for dolphin.workflows.wrapped_phase.run : (\d*\.\d{2})"
-        r" minutes"
-        r" \((\d*\.\d{2}) seconds\)"
-    )
-    cfg_version_pat = r"Config file dolphin version: (.*)"
-    run_version_pat = r"Current running dolphin version: (.*)"
-    # lines start with datetimes like 2023-04-19 17:13:02
-    # so at first line, parse and store the time of execution
-    for i, line in enumerate(open(logfile).readlines()):
-        if i == 0:
-            datetime_str = line[:19]
-            out["execution_time"] = datetime.strptime(datetime_str, "%Y-%m-%d %H:%M:%S")
-        if m := re.search(mempat, line):
-            out["memory"] = float(m.groups()[0])
-            continue
-        if m := re.search(timepat, line):
-            out["runtime"] = float(m.groups()[1])
-            continue
-        if m := re.search(wrapped_phase_timepat, line):
-            out["wrapped_phase_runtimes"].append(float(m.groups()[1]))
-            continue
-        if m := re.search(cfg_version_pat, line):
-            out["config_version"] = m.groups()[0]
-            continue
-        if m := re.search(run_version_pat, line):
-            out["run_version"] = m.groups()[0]
-            continue
-    return out
-
-
-def _parse_config(workflow: Workflow):
-    """Grab the relevant parameters from the config file."""
-    return {
-        "block": workflow.worker_settings.block_shape,
-        "strides": workflow.output_options.strides,
-        "threads_per_worker": workflow.worker_settings.threads_per_worker,
-        "n_slc": len(workflow.cslc_file_list),
-        "n_workers": workflow.worker_settings.n_workers,
-        "creation_time": workflow.creation_time_utc,
-    }
-
-
-def _get_cli_args():
-    parser = argparse.ArgumentParser()
-    parser.add_argument(
-        "--config-files",
-        nargs="*",
-        help="Path(s) to config file",
-    )
-    parser.add_argument(
-        "-o",
-        "--outfile",
-        help="Output file (CSV). If None, outputs as `log_file`.csv",
-        default=None,
-    )
-    return parser.parse_args()
-
-
-def main():
-    """Run main entry point."""
-    args = _get_cli_args()
-    dfs = []
-    for config_file in args.config_files:
-        dfs.append(get_df(config_file))
-
-    df = pd.concat(dfs)
-
-    if not args.outfile:
-        # Save as csv file with same directory as log_file
-        outfile = (
-            Path(args.config_files[0]).parent
-            / f"benchmarks_{datetime.now().strftime('%Y%m%d_%H%M%S')}.csv"
-        )
-    else:
-        outfile = args.outfile
-
-    if not outfile.endswith(".csv"):
-        outfile = outfile + ".csv"
-        print(f"Output file must be csv. Writing to {outfile}", file=sys.stderr)
-
-    df.to_csv(outfile, index=False)
-
-
-if __name__ == "__main__":
-    main()
diff --git a/scripts/benchmarking/zenodo.py b/scripts/benchmarking/zenodo.py
deleted file mode 100644
index df11dd99..00000000
--- a/scripts/benchmarking/zenodo.py
+++ /dev/null
@@ -1,91 +0,0 @@
-from __future__ import annotations
-
-from pathlib import Path
-from typing import Optional, Union
-
-import pooch
-import requests
-
-from dolphin._log import get_log
-from dolphin._types import Filename
-
-logger = get_log(__name__)
-
-RECORD_ID = "1171149"
-
-
-def make_pooch(
-    record: Union[int, str], sandbox: bool = True, path: Optional[Filename] = None
-):
-    """Create a Pooch instance for a Zenodo record.
-
-    Parameters
-    ----------
-    record : Union[int, str]
-        The Zenodo record ID.
-    sandbox : bool, optional
-        Whether to use the sandbox (default) or the real Zenodo.
-    path : Optional[Filename], optional
-        The path to the cache folder. If None (default), use the default
-        cache folder for the operating system.
-
-    Returns
-    -------
-    pooch.Pooch
-        The Pooch instance.
-    """
-    base_url, registry = get_zenodo_links(record, sandbox)
-    dog = pooch.create(
-        # Use the default cache folder for the operating system
-        path=path or pooch.os_cache("dolphin"),
-        base_url=base_url,
-        # The registry specifies the files that can be fetched
-        registry=registry,
-    )
-    return dog
-
-
-def get_zenodo_links(
-    record: Union[int, str], sandbox=True
-) -> tuple[str, dict[str, str]]:
-    """Get the urls and MD5 checksums for the files in a Zenodo record."""
-    # Get the record metadata
-    if sandbox:
-        url = f"https://sandbox.zenodo.org/api/records/{record}"
-    else:
-        url = f"https://zenodo.org/api/records/{record}"
-    r = requests.get(url)
-    r.raise_for_status()
-    data = r.json()
-
-    # Get the files
-    files = data["files"]
-    # print the total size of the files
-    total_size = sum(f["size"] for f in files)
-    logger.info(f"Total size of {len(files)} files: {total_size / 1e6:.1f} MB")
-
-    # Extract the urls and checksums
-    urls = [f["links"]["self"] for f in files]
-    # Get the base url
-    base_url = "/".join(urls[0].split("/")[:-1]) + "/"
-
-    filenames = [Path(fn).name for fn in urls]
-    checksums = [f["checksum"] for f in files]
-
-    # Return a dict compatible with the Pooch registry
-    return base_url, dict(zip(filenames, checksums))
-
-
-POOCH = make_pooch(RECORD_ID)
-
-
-def get_all_files(
-    record: Union[int, str], sandbox: bool = True, path: Optional[Filename] = None
-):
-    """Download all the files in a Zenodo record."""
-    dog = make_pooch(record, sandbox, path)
-    # Fetch all the files
-    for fn in dog.registry_files:
-        logger.info("Fetching " + fn)
-        dog.fetch(fn)
-    return dog
diff --git a/scripts/release/generate_product_docx_table.py b/scripts/release/generate_product_docx_table.py
deleted file mode 100755
index b17b2812..00000000
--- a/scripts/release/generate_product_docx_table.py
+++ /dev/null
@@ -1,120 +0,0 @@
-#!/usr/bin/env python
-from __future__ import annotations
-
-import argparse
-from itertools import groupby
-
-import h5py
-
-from dolphin._types import Filename
-
-
-def generate_docx_table(hdf5_path: Filename, output_path: Filename):
-    """Create a Word document with a table of HDF5 datasets."""
-    # https://python-docx.readthedocs.io/en/latest/user/quickstart.html#adding-a-table
-    from docx import Document
-    from docx.enum.table import WD_ROW_HEIGHT_RULE, WD_TABLE_ALIGNMENT
-    from docx.oxml import parse_xml
-    from docx.oxml.ns import nsdecls
-    from docx.shared import Pt
-
-    def _add_row(table, text, height=15, shade=False, bold=False):
-        # _tc.get_or_add_tcPr().append(shading_elm)
-        row = table.add_row()
-        row.height_rule = WD_ROW_HEIGHT_RULE.AT_LEAST
-        row.height = Pt(height)
-        if isinstance(text, list):
-            for i in range(len(text)):
-                row.cells[i].text = text[i]
-        else:
-            row.cells[1].merge(row.cells[0])
-            row.cells[1].merge(row.cells[2])
-            row.cells[1].text = text
-        # https://stackoverflow.com/questions/26752856/python-docx-set-table-cell-background-and-text-color  # noqa
-        if shade:
-            shading_elm = parse_xml(r'<w:shd {} w:fill="D9D9D9"/>'.format(nsdecls("w")))
-            row.cells[0]._tc.get_or_add_tcPr().append(shading_elm)
-        # Set the text color to black and remove bold
-        run = row.cells[0].paragraphs[0].runs[0]
-        run.font.color.rgb = None
-        if not bold:
-            run.font.bold = False
-
-    document = Document()
-    # Set the default document font to Arial
-    style = document.styles["Normal"]
-    font = style.font
-    font.name = "Arial"
-
-    for group_name, rows in _get_hdf5_attributes_by_group(hdf5_path).items():
-        document.add_heading(f"Group: {group_name}", level=2)
-        table = document.add_table(cols=3, rows=0)
-        table.style = "Table Grid"  # Use the "Table Grid" style to get borders
-        table.alignment = WD_TABLE_ALIGNMENT.LEFT
-
-        for row in rows:
-            name = row.pop("Name")
-            desc = row.pop("Description")
-
-            _add_row(table, f"Name: {name}", shade=True)
-
-            row_text = [f"{k}: {v or 'scalar'}" for k, v in row.items()]
-            _add_row(table, row_text)
-            _add_row(table, f"Description: {desc}")
-
-    print(f"Saving to {output_path}")
-    document.save(output_path)
-
-
-def _get_hdf5_attributes(hdf5_path: Filename) -> list:
-    table_data = []
-
-    def append_dset_to_table(name, item):
-        """Add all dataset's metadata using `visititems`."""
-        if not isinstance(item, h5py.Dataset):
-            return None
-        data_type = item.dtype
-        shape = item.shape
-        description = item.attrs.get("long_name", "")
-        units = item.attrs.get("units", "")
-        table_data.append(
-            dict(
-                Name=name,
-                Type=data_type,
-                Shape=shape,
-                Units=units,
-                Description=description,
-            )
-        )
-
-    with h5py.File(hdf5_path, "r") as hf:
-        hf.visititems(append_dset_to_table)
-    return table_data
-
-
-def _get_hdf5_attributes_by_group(hdf5_path: Filename) -> dict[str, list]:
-    def get_group(name):
-        try:
-            return name.split("/")[-2]
-        except IndexError:
-            return "root"
-
-    table_data = _get_hdf5_attributes(hdf5_path)
-
-    group_sorted_rows = sorted(table_data, key=lambda row: get_group(row["Name"]))
-    # Make a dict, where keys are group name, value is the list of rows
-    # e.g.:  { 'DISP': [ {'Name': ,....], 'corrections': [{'Name':...}]
-    return {
-        k: list(v)
-        for k, v in groupby(group_sorted_rows, key=lambda row: get_group(row["Name"]))
-    }
-
-
-if __name__ == "__main__":
-    parser = argparse.ArgumentParser()
-    parser.add_argument("hdf5_path", type=str, help="Path to the HDF5 file")
-    parser.add_argument(
-        "output_path", type=str, help="Path to the output Word docx file"
-    )
-    args = parser.parse_args()
-    generate_docx_table(args.hdf5_path, args.output_path)
diff --git a/scripts/release/list_packages.py b/scripts/release/list_packages.py
deleted file mode 100755
index 4b76a4fe..00000000
--- a/scripts/release/list_packages.py
+++ /dev/null
@@ -1,240 +0,0 @@
-#!/usr/bin/env python
-"""list packages installed with conda & yum in CSV format.
-
-Optionally list packages installed inside a container image instead.
-"""
-from __future__ import annotations
-
-import argparse
-import csv
-import functools
-import json
-import os
-import subprocess
-import sys
-from dataclasses import astuple, dataclass
-from itertools import dropwhile
-from pathlib import Path
-from typing import Any, Iterator, Optional, TextIO
-
-DESCRIPTION = __doc__
-
-
-@dataclass
-class Package:
-    """An installed package."""
-
-    name: str
-    version: str
-    package_manager: str
-    channel: str
-
-
-class CommandNotFoundError(Exception):
-    """Raised when a required Unix shell command was not found."""
-
-    pass
-
-
-class YumListIsAnnoyingError(Exception):
-    """Raised when 'yum list' does something annoying."""
-
-    pass
-
-
-def check_command(cmd: str) -> bool:
-    """Check if a Unix shell command is available."""
-    # Check if `cmd` is available on the PATH or as a builtin command.
-    args = ["command", "-v", cmd]
-    try:
-        subprocess.run(args, check=True, stdout=subprocess.DEVNULL)
-    except subprocess.CalledProcessError as exc:
-        # The exit status is 0 if the command was found, and 1 if not.
-        # Any other status is unexpected, so re-raise the error.
-        if exc.returncode == 1:
-            return False
-        else:
-            raise
-    else:
-        return True
-
-
-@functools.lru_cache(maxsize=None)
-def conda_command() -> str:
-    """Get the name of the conda executable."""
-    # A list of possible "conda-like" commands to try.
-    possible_commands = [
-        "conda",
-        "mamba",
-        "micromamba",
-    ]
-
-    # Try each possible command. Return the first one that's available.
-    for cmd in possible_commands:
-        if check_command(cmd):
-            return cmd
-    else:
-        errmsg = f"conda command not found -- tried {possible_commands}"
-        raise CommandNotFoundError(errmsg)
-
-
-def list_conda_packages(env: str = "base") -> list[Package]:
-    """List conda packages installed in the specified environment."""
-    # Get the list of conda packages as a JSON-formatted string.
-    args = [conda_command(), "list", "--name", env, "--json"]
-    res = subprocess.run(args, capture_output=True, check=True, text=True)
-
-    # Parse the JSON string.
-    package_list = json.loads(res.stdout)
-
-    def as_package(package_info: dict) -> Package:
-        name = package_info["name"]
-        version = package_info["version"]
-        channel = package_info["channel"]
-        return Package(name, version, "conda", channel)
-
-    # Convert to `Package` objects.
-    packages = map(as_package, package_list)
-
-    # Sort alphabetically by name.
-    return sorted(packages, key=lambda package: package.name)
-
-
-def list_yum_packages() -> list[Package]:
-    """List system packages installed with yum."""
-    # Check if 'yum' is available.
-    if not check_command("yum"):
-        errmsg = "'yum' command not found"
-        raise CommandNotFoundError(errmsg)
-
-    # Get the list of yum packages.
-    args = ["yum", "list", "installed"]
-    res = subprocess.run(args, capture_output=True, check=True, text=True)
-
-    # Split the output into a sequence of lines.
-    # Skip past the first few lines until we get to the actual list of packages.
-    lines = res.stdout.splitlines()
-    package_list = dropwhile(lambda line: line != "Installed Packages", lines)
-    try:
-        next(package_list)  # type: ignore[call-overload]
-    except StopIteration:
-        cmd = subprocess.list2cmdline(args)
-        errmsg = f"unexpected output from {cmd!r}"
-        raise RuntimeError(errmsg)
-
-    def as_package(line: str) -> Package:
-        try:
-            name, version, channel = line.split()
-        except ValueError as exc:
-            raise YumListIsAnnoyingError from exc
-
-        # Package names are in '{name}.{arch}' format.
-        # Strip arch info from package name.
-        name = ".".join(name.split(".")[:-1])
-
-        # If the repo cannot be determined, yum prints "installed" by default.
-        channel = channel.lstrip("@") if (channel != "installed") else ""
-
-        return Package(name, version, "yum", channel)
-
-    def parse_lines(lines: Iterator[str]) -> Iterator[Package]:
-        for line in lines:
-            # Sometimes package info in the output is inexplicably broken up
-            # across two lines...
-            try:
-                package = as_package(line)
-            except YumListIsAnnoyingError:
-                line += next(lines)
-                package = as_package(line)
-
-            yield package
-
-    # Parse each line as a `Package` object.
-    packages = parse_lines(package_list)
-
-    # Sort alphabetically by name.
-    return sorted(packages, key=lambda package: package.name)
-
-
-def list_packages() -> list[Package]:
-    """List installed packages."""
-    return list_conda_packages() + list_yum_packages()
-
-
-def write_package_csv(csvfile: TextIO = sys.stdout) -> None:
-    """Write a list of installed packages to file in CSV format."""
-    packages = list_packages()
-    csvwriter = csv.writer(csvfile)
-    csvwriter.writerow(["Name", "Version", "Package Manager", "Channel"])
-    csvwriter.writerows(map(astuple, packages))
-
-
-def setup_parser() -> argparse.ArgumentParser:
-    """Set up the command-line argument parser."""
-    parser = argparse.ArgumentParser(description=DESCRIPTION)
-
-    docker_group = parser.add_argument_group("docker options")
-    docker_group.add_argument(
-        "image", nargs="?", type=str, help="a docker image tag or ID"
-    )
-    docker_group.add_argument(
-        "-e",
-        "--entrypoint",
-        type=str,
-        help="overwrite the default ENTRYPOINT of the image",
-    )
-
-    return parser
-
-
-def parse_args(args: Optional[list[str]] = None) -> dict[str, Any]:
-    """Parse command-line arguments."""
-    parser = setup_parser()
-    params = parser.parse_args(args)
-
-    if (params.image is None) and (params.entrypoint is not None):
-        parser.error(
-            "entrypoint should not be specified unless an image tag or ID is provided"
-        )
-
-    return vars(params)
-
-
-def run_in_container(image: str, entrypoint: Optional[str] = None) -> None:
-    """Run this module inside a container derived from the specified image."""
-    # Check if 'docker' is available.
-    if not check_command("docker"):
-        errmsg = "'docker' command not found"
-        raise CommandNotFoundError(errmsg)
-
-    this_file = Path(__file__).resolve()
-    workdir = Path("/tmp")
-
-    args = [
-        "docker",
-        "run",
-        "--rm",
-        f"--user={os.getuid()}:{os.getgid()}",
-        f"--volume={this_file}:{workdir / 'list_packages.py'}:ro",
-        f"--workdir={workdir}",
-    ]
-
-    if entrypoint is not None:
-        args.append(f"--entrypoint={entrypoint}")
-
-    args += [image, "python", "-m", "list_packages"]
-
-    # Mount this script inside the container and run it without arguments.
-    subprocess.run(args, check=True)
-
-
-def main(args: Optional[list[str]] = None) -> None:  # noqa: D103
-    kwargs = parse_args(args)
-    if kwargs.get("image") is None:
-        write_package_csv()
-    else:
-        run_in_container(**kwargs)
-
-
-if __name__ == "__main__":
-    main()
diff --git a/scripts/release/validate_product.py b/scripts/release/validate_product.py
deleted file mode 100755
index 39eab7d2..00000000
--- a/scripts/release/validate_product.py
+++ /dev/null
@@ -1,520 +0,0 @@
-#!/usr/bin/env python
-import argparse
-from pathlib import Path
-from typing import TYPE_CHECKING, Any, Optional
-
-import h5py
-import numpy as np
-from numpy.typing import ArrayLike
-
-from dolphin import io
-from dolphin._log import get_log
-from dolphin._types import Filename
-
-logger = get_log()
-
-if TYPE_CHECKING:
-    _SubparserType = argparse._SubParsersAction[argparse.ArgumentParser]
-else:
-    _SubparserType = Any
-
-DSET_DEFAULT = "unwrapped_phase"
-
-
-class ValidationError(Exception):
-    """Raised when a product fails a validation check."""
-
-    pass
-
-
-class ComparisonError(ValidationError):
-    """Exception raised when two datasets do not match."""
-
-    pass
-
-
-def compare_groups(
-    golden_group: h5py.Group,
-    test_group: h5py.Group,
-    pixels_failed_threshold: float = 0.01,
-    diff_threshold: float = 1e-5,
-) -> None:
-    """Compare all datasets in two HDF5 files.
-
-    Parameters
-    ----------
-    golden_group : h5py.Group
-        Path to the golden file.
-    test_group : h5py.Group
-        Path to the test file to be compared.
-    pixels_failed_threshold : float, optional
-        The threshold of the percentage of pixels that can fail the comparison.
-    diff_threshold : float, optional
-        The abs. difference threshold between pixels to consider failing.
-
-    Raises
-    ------
-    ComparisonError
-        If the two files do not match in all datasets.
-    """
-    # Check if group names match
-    if set(golden_group.keys()) != set(test_group.keys()):
-        raise ComparisonError(
-            f"Group keys do not match: {set(golden_group.keys())} vs"
-            f" {set(test_group.keys())}"
-        )
-
-    for key in golden_group.keys():
-        if isinstance(golden_group[key], h5py.Group):
-            compare_groups(
-                golden_group[key],
-                test_group[key],
-                pixels_failed_threshold,
-                diff_threshold,
-            )
-        else:
-            test_dataset = test_group[key]
-            golden_dataset = golden_group[key]
-            _compare_datasets_attr(golden_dataset, test_dataset)
-
-            if key == "connected_component_labels":
-                _validate_conncomp_labels(test_dataset, golden_dataset)
-            elif key == "unwrapped_phase":
-                test_conncomps = test_group["connected_component_labels"]
-                golden_conncomps = golden_group["connected_component_labels"]
-                _validate_unwrapped_phase(
-                    test_dataset,
-                    golden_dataset,
-                    test_conncomps,
-                    golden_conncomps,
-                )
-            else:
-                _validate_dataset(
-                    test_dataset,
-                    golden_dataset,
-                    pixels_failed_threshold,
-                    diff_threshold,
-                )
-
-
-def _compare_datasets_attr(
-    golden_dataset: h5py.Dataset, test_dataset: h5py.Dataset
-) -> None:
-    if golden_dataset.name != test_dataset.name:
-        raise ComparisonError(
-            f"Dataset names do not match: {golden_dataset.name} vs {test_dataset.name}"
-        )
-    name = golden_dataset.name
-
-    if golden_dataset.shape != test_dataset.shape:
-        raise ComparisonError(
-            f"{name} shapes do not match: {golden_dataset.shape} vs"
-            f" {test_dataset.shape}"
-        )
-
-    if golden_dataset.dtype != test_dataset.dtype:
-        raise ComparisonError(
-            f"{name} dtypes do not match: {golden_dataset.dtype} vs"
-            f" {test_dataset.dtype}"
-        )
-
-    if golden_dataset.attrs.keys() != test_dataset.attrs.keys():
-        raise ComparisonError(
-            f"{name} attribute keys do not match: {golden_dataset.attrs.keys()} vs"
-            f" {test_dataset.attrs.keys()}"
-        )
-
-    for attr_key in golden_dataset.attrs.keys():
-        if attr_key in ("REFERENCE_LIST", "DIMENSION_LIST"):
-            continue
-        val1, val2 = golden_dataset.attrs[attr_key], test_dataset.attrs[attr_key]
-        if isinstance(val1, np.ndarray):
-            is_equal = np.allclose(val1, val2, equal_nan=True)
-        elif isinstance(val1, np.floating) and np.isnan(val1) and np.isnan(val2):
-            is_equal = True
-        else:
-            is_equal = val1 == val2
-        if not is_equal:
-            raise ComparisonError(
-                f"{name} attribute values for key '{attr_key}' do not match: "
-                f"{golden_dataset.attrs[attr_key]} vs {test_dataset.attrs[attr_key]}"
-            )
-
-
-def _fmt_ratio(num: int, den: int, digits: int = 3) -> str:
-    """Get a string representation of a rational number as a fraction and percent.
-
-    Parameters
-    ----------
-    num : int
-        The numerator.
-    den : int
-        The denominator.
-    digits : int, optional
-        Number of decimal digits to use. Defaults to 3.
-
-    Returns
-    -------
-    str
-        A string representation of the input.
-    """
-    return f"{num}/{den} ({100.0 * num / den:.{digits}f}%)"
-
-
-def _validate_conncomp_labels(
-    test_dataset: h5py.Dataset,
-    ref_dataset: h5py.Dataset,
-    threshold: float = 0.9,
-) -> None:
-    """Validate connected component labels from unwrapping.
-
-    Computes a binary mask of nonzero-valued labels in the test and reference datasets,
-    and checks the intersection between the two masks. The dataset fails validation if
-    the ratio of the intersection area to the reference mask area is below a
-    predetermined minimum threshold.
-
-    Parameters
-    ----------
-    test_dataset : h5py.Dataset
-        HDF5 dataset containing connected component labels to be validated.
-    ref_dataset : h5py.Dataset
-        HDF5 dataset containing connected component labels to use as reference. Must
-        have the same shape as `test_dataset`.
-    threshold : float, optional
-        Minimum allowable intersection area between nonzero-labeled regions in the test
-        and reference dataset, as a fraction of the total nonzero-labeled area in the
-        reference dataset. Must be in the interval [0, 1]. Defaults to 0.9.
-
-    Raises
-    ------
-    ComparisonError
-        If the intersecting area between the two masks was below the threshold.
-    """
-    logger.info("Checking connected component labels...")
-
-    if test_dataset.shape != ref_dataset.shape:
-        errmsg = (
-            "shape mismatch: test dataset and reference dataset must have the same"
-            f" shape, got {test_dataset.shape} vs {ref_dataset.shape}"
-        )
-        raise ComparisonError(errmsg)
-
-    if not (0.0 <= threshold <= 1.0):
-        errmsg = f"threshold must be between 0 and 1, got {threshold}"
-        raise ValueError(errmsg)
-
-    # Total size of each dataset.
-    size = ref_dataset.size
-
-    # Compute binary masks of pixels with nonzero labels in each dataset.
-    test_nonzero = np.not_equal(test_dataset, 0)
-    ref_nonzero = np.not_equal(ref_dataset, 0)
-
-    # Compute the intersection & union of both masks.
-    intersect = test_nonzero & ref_nonzero
-    union = test_nonzero | ref_nonzero
-
-    # Compute the total area of each mask.
-    test_area = np.sum(test_nonzero)
-    ref_area = np.sum(ref_nonzero)
-    intersect_area = np.sum(intersect)
-    union_area = np.sum(union)
-
-    # Log some statistics about the unwrapped area.
-    logger.info(f"Test unwrapped area: {_fmt_ratio(test_area, size)}")
-    logger.info(f"Reference unwrapped area: {_fmt_ratio(ref_area, size)}")
-    logger.info(f"Intersection/Reference: {_fmt_ratio(intersect_area, ref_area)}")
-    logger.info(f"Intersection/Union: {_fmt_ratio(intersect_area, union_area)}")
-
-    # Compute the ratio of intersection area to area in the reference mask.
-    ratio = intersect_area / ref_area
-
-    if ratio < threshold:
-        errmsg = (
-            f"connected component labels dataset {test_dataset.name!r} failed"
-            " validation: insufficient area of overlap between test and reference"
-            f" nonzero labels ({ratio} < {threshold})"
-        )
-        raise ComparisonError(errmsg)
-
-
-def _validate_unwrapped_phase(
-    test_dataset: h5py.Dataset,
-    ref_dataset: h5py.Dataset,
-    test_conncomps: ArrayLike,
-    ref_conncomps: ArrayLike,
-    nan_threshold: float = 0.01,
-    atol: float = 1e-6,
-) -> None:
-    """Validate unwrapped phase values against a reference dataset.
-
-    Checks that the phase values in the test dataset are congruent with the reference
-    dataset -- that is, their values are approximately the same modulo 2pi.
-
-    Parameters
-    ----------
-    test_dataset : h5py.Dataset
-        HDF5 dataset containing unwrapped phase values to be validated.
-    ref_dataset : h5py.Dataset
-        HDF5 dataset containing unwrapped phase values to use as reference. Must have
-        the same shape as `test_dataset`.
-    test_conncomps : array_like
-        Connected component labels associated with `test_dataset`.
-    ref_conncomps : array_like
-        Connected component labels associated with `ref_dataset`.
-    nan_threshold : float
-        Maximum allowable fraction of NaN values among valid pixels (pixels with nonzero
-        connected component label). Must be in the interval [0, 1]. Defaults to 0.01.
-    atol : float, optional
-        Maximum allowable absolute error between the re-wrapped reference and test
-        values, in radians. Must be nonnegative. Defaults to 1e-6.
-
-    Raises
-    ------
-    ValidationError
-        If the NaN value count exceeded the specified threshold.
-    ComparisonError
-        If the two datasets were not congruent within the specified error tolerance.
-    """
-    logger.info("Checking unwrapped phase...")
-
-    if test_dataset.shape != ref_dataset.shape:
-        errmsg = (
-            "shape mismatch: test dataset and reference dataset must have the same"
-            f" shape, got {test_dataset.shape} vs {ref_dataset.shape}"
-        )
-        raise ComparisonError(errmsg)
-
-    if (test_dataset.shape != test_conncomps.shape) or (
-        ref_dataset.shape != ref_conncomps.shape
-    ):
-        errmsg = (
-            "shape mismatch: unwrapped phase and connected component labels must have"
-            " the same shape"
-        )
-        raise ValidationError(errmsg)
-
-    if not (0.0 <= nan_threshold <= 1.0):
-        errmsg = f"nan_threshold must be between 0 and 1, got {nan_threshold}"
-        raise ValueError(errmsg)
-
-    if atol < 0.0:
-        errmsg = f"atol must be >= 0, got {atol}"
-        raise ValueError(errmsg)
-
-    # Get a mask of valid pixels (pixels that had nonzero connected component label) in
-    # both the test & reference data.
-    test_valid_mask = np.not_equal(test_conncomps, 0)
-    ref_valid_mask = np.not_equal(ref_conncomps, 0)
-    valid_mask = test_valid_mask & ref_valid_mask
-
-    # Get the total valid area in both datasets.
-    test_valid_area = np.sum(test_valid_mask)
-    ref_valid_area = np.sum(ref_valid_mask)
-
-    # Get a mask of NaN values in either dataset.
-    test_nan_mask = np.isnan(test_dataset)
-    ref_nan_mask = np.isnan(ref_dataset)
-    nan_mask = test_nan_mask | ref_nan_mask
-
-    # Get the total number of NaN values in the valid regions of each dataset.
-    test_nan_count = np.sum(test_nan_mask & test_valid_mask)
-    ref_nan_count = np.sum(ref_nan_mask & ref_valid_mask)
-
-    # Log some info about the NaN values.
-    logger.info(f"Test nan count: {_fmt_ratio(test_nan_count, test_valid_area)}")
-    logger.info(f"Reference nan count: {_fmt_ratio(ref_nan_count, ref_valid_area)}")
-
-    # Compute the fraction of NaN values in the valid region.
-    test_nan_frac = test_nan_count / test_valid_area
-
-    if test_nan_frac > nan_threshold:
-        errmsg = (
-            f"unwrapped phase dataset {test_dataset.name!r} failed validation: too"
-            f" many nan values ({test_nan_frac} > {nan_threshold})"
-        )
-        raise ValidationError(errmsg)
-
-    def rewrap(phi: np.ndarray) -> np.ndarray:
-        tau = 2.0 * np.pi
-        return phi - tau * np.ceil((phi - np.pi) / tau)
-
-    # Compute the difference between the test & reference values and wrap it to the
-    # interval (-pi, pi].
-    diff = np.subtract(ref_dataset, test_dataset)
-    wrapped_diff = rewrap(diff)
-
-    # Mask out invalid pixels and NaN-valued pixels.
-    wrapped_diff = wrapped_diff[valid_mask & ~nan_mask]
-
-    # Log some statistics about the deviation between the test & reference phase.
-    abs_wrapped_diff = np.abs(wrapped_diff)
-    mean_abs_err = np.mean(abs_wrapped_diff)
-    max_abs_err = np.max(abs_wrapped_diff)
-    logger.info(f"Mean absolute re-wrapped phase error: {mean_abs_err:.5f} rad")
-    logger.info(f"Max absolute re-wrapped phase error: {max_abs_err:.5f} rad")
-
-    noncongruent_count = np.sum(abs_wrapped_diff > atol)
-    logger.info(
-        "Non-congruent pixel count:"
-        f" {_fmt_ratio(noncongruent_count, wrapped_diff.size)}"
-    )
-
-    if noncongruent_count != 0:
-        errmsg = (
-            f"unwrapped phase dataset {test_dataset.name!r} failed validation: phase"
-            " values were not congruent with reference dataset"
-        )
-        raise ComparisonError(errmsg)
-
-
-def _validate_dataset(
-    test_dataset: h5py.Dataset,
-    golden_dataset: h5py.Dataset,
-    pixels_failed_threshold: float = 0.01,
-    diff_threshold: float = 1e-5,
-) -> None:
-    """Validate a generic dataset.
-
-    Parameters
-    ----------
-    test_dataset : h5py.Dataset
-        HDF5 dataset to be validated.
-    golden_dataset : h5py.Dataset
-        HDF5 dataset to use as reference.
-    pixels_failed_threshold : float, optional
-        The threshold of the percentage of pixels that can fail the comparison. Defaults
-        to 0.01.
-    diff_threshold : float, optional
-        The abs. difference threshold between pixels to consider failing. Defaults to
-        1e-5.
-
-    Raises
-    ------
-    ComparisonError
-        If the two datasets do not match.
-    """
-    golden = golden_dataset[()]
-    test = test_dataset[()]
-    if golden.dtype.kind == "S":
-        if not np.array_equal(golden, test):
-            raise ComparisonError(f"Dataset {golden_dataset.name} values do not match")
-        return
-
-    img_gold = np.ma.masked_invalid(golden)
-    img_test = np.ma.masked_invalid(test)
-    abs_diff = np.abs((img_gold.filled(0) - img_test.filled(0)))
-    num_failed = np.count_nonzero(abs_diff > diff_threshold)
-    # num_pixels = np.count_nonzero(~np.isnan(img_gold))  # do i want this?
-    num_pixels = img_gold.size
-    if num_failed / num_pixels > pixels_failed_threshold:
-        raise ComparisonError(
-            f"Dataset {golden_dataset.name} values do not match: Number of"
-            f" pixels failed: {num_failed} / {num_pixels} ="
-            f" {100*num_failed / num_pixels:.2f}%"
-        )
-
-
-def _check_raster_geometadata(golden_file: Filename, test_file: Filename) -> None:
-    """Check if the raster metadata (bounds, CRS, and GT) match.
-
-    Parameters
-    ----------
-    golden_file : Filename
-        Path to the golden file.
-    test_file : Filename
-        Path to the test file to be compared.
-
-    Raises
-    ------
-    ComparisonError
-        If the two files do not match in their metadata
-    """
-    funcs = [io.get_raster_bounds, io.get_raster_crs, io.get_raster_gt]
-    for func in funcs:
-        val_golden = func(golden_file)  # type: ignore
-        val_test = func(test_file)  # type: ignore
-        if val_golden != val_test:
-            raise ComparisonError(f"{func} does not match: {val_golden} vs {val_test}")
-
-
-def _check_compressed_slc_dirs(golden: Filename, test: Filename) -> None:
-    """Check if the compressed SLC directories match.
-
-    Assumes that the compressed SLC directories are in the same directory as the
-    `golden` and `test` product files, with the directory name `compressed_slcs`.
-
-    Parameters
-    ----------
-    golden : Filename
-        Path to the golden file.
-    test : Filename
-        Path to the test file to be compared.
-
-    Raises
-    ------
-    ComparisonError
-        If file names do not match in their compressed SLC directories
-    """
-    golden_slc_dir = Path(golden).parent / "compressed_slcs"
-    test_slc_dir = Path(test).parent / "compressed_slcs"
-
-    if not golden_slc_dir.exists():
-        logger.info("No compressed SLC directory found in golden product.")
-        return
-    if not test_slc_dir.exists():
-        raise ComparisonError(
-            f"{test_slc_dir} does not exist, but {golden_slc_dir} exists."
-        )
-
-    golden_slc_names = [p.name for p in golden_slc_dir.iterdir()]
-    test_slc_names = [p.name for p in test_slc_dir.iterdir()]
-
-    if set(golden_slc_names) != set(test_slc_names):
-        raise ComparisonError(
-            f"Compressed SLC directories do not match: {golden_slc_names} vs"
-            f" {test_slc_names}"
-        )
-
-
-def compare(golden: Filename, test: Filename, data_dset: str = DSET_DEFAULT) -> None:
-    """Compare two HDF5 files for consistency."""
-    logger.info("Comparing HDF5 contents...")
-    with h5py.File(golden, "r") as hf_g, h5py.File(test, "r") as hf_t:
-        compare_groups(hf_g, hf_t)
-
-    logger.info("Checking geospatial metadata...")
-    _check_raster_geometadata(
-        io.format_nc_filename(golden, data_dset),
-        io.format_nc_filename(test, data_dset),
-    )
-
-    logger.info(f"Files {golden} and {test} match.")
-    _check_compressed_slc_dirs(golden, test)
-
-
-def get_parser(
-    subparser: Optional[_SubparserType] = None, subcommand_name: str = "run"
-) -> argparse.ArgumentParser:
-    """Set up the command line interface."""
-    metadata = dict(
-        description="Compare two HDF5 files for consistency.",
-        formatter_class=argparse.ArgumentDefaultsHelpFormatter,
-    )
-    if subparser:
-        # Used by the subparser to make a nested command line interface
-        parser = subparser.add_parser(subcommand_name, **metadata)  # type: ignore
-    else:
-        parser = argparse.ArgumentParser(**metadata)  # type: ignore
-
-    parser.add_argument("golden", help="The golden HDF5 file.")
-    parser.add_argument("test", help="The test HDF5 file to be compared.")
-    parser.add_argument("--data-dset", default=DSET_DEFAULT)
-    parser.set_defaults(run_func=compare)
-    return parser
-
-
-if __name__ == "__main__":
-    parser = get_parser()
-    args = parser.parse_args()
-    compare(args.golden, args.test, args.data_dset)
diff --git a/scripts/run_repeated_nrt.py b/scripts/run_repeated_nrt.py
deleted file mode 100755
index a329fa13..00000000
--- a/scripts/run_repeated_nrt.py
+++ /dev/null
@@ -1,455 +0,0 @@
-#!/usr/bin/env python
-from __future__ import annotations
-
-import argparse
-from concurrent.futures import ThreadPoolExecutor
-from itertools import chain
-from pathlib import Path
-from typing import Any, Mapping, Sequence
-
-from dolphin import io, ps, stack, utils
-from dolphin._log import get_log, log_runtime
-from dolphin._types import Filename
-from dolphin.workflows import s1_disp
-from dolphin.workflows._utils import group_by_burst, make_nodata_mask
-from dolphin.workflows.config import (
-    OPERA_DATASET_NAME,
-    InterferogramNetworkType,
-    ShpMethod,
-    Workflow,
-    WorkflowName,
-)
-
-logger = get_log("dolphin.run_repeated_nrt")
-
-
-def _create_cfg(
-    *,
-    slc_files: Sequence[Filename],
-    half_window_size: tuple[int, int] = (11, 5),
-    first_ministack: bool = False,
-    run_unwrap: bool = False,
-    shp_method: ShpMethod = ShpMethod.GLRT,
-    amplitude_mean_files: Sequence[Filename] = [],
-    amplitude_dispersion_files: Sequence[Filename] = [],
-    strides: Mapping[str, int] = {"x": 6, "y": 3},
-    work_dir: Path = Path("."),
-    n_parallel_bursts: int = 1,
-):
-    # strides = {"x": 1, "y": 1}
-    interferogram_network: dict[str, Any]
-    if first_ministack:
-        interferogram_network = dict(
-            network_type=InterferogramNetworkType.SINGLE_REFERENCE
-        )
-        workflow_name = WorkflowName.STACK
-    else:
-        interferogram_network = dict(
-            network_type=InterferogramNetworkType.MANUAL_INDEX,
-            indexes=[(0, -1)],
-        )
-        workflow_name = WorkflowName.SINGLE
-
-    cfg = Workflow(
-        # Things that change with each workflow run
-        cslc_file_list=slc_files,
-        interferogram_network=interferogram_network,
-        amplitude_mean_files=amplitude_mean_files,
-        amplitude_dispersion_files=amplitude_dispersion_files,
-        # Configurable from CLI inputs:
-        output_options=dict(
-            strides=strides,
-        ),
-        phase_linking=dict(
-            ministack_size=1000,  # for single update, process in one ministack
-            half_window={"x": half_window_size[0], "y": half_window_size[1]},
-            shp_method=shp_method,
-        ),
-        scratch_directory=work_dir / "scratch",
-        output_directory=work_dir / "output",
-        worker_settings=dict(
-            #     block_size_gb=block_size_gb,
-            n_parallel_bursts=n_parallel_bursts,
-            n_workers=4,
-            threads_per_worker=8,
-        ),
-        #     ps_options=dict(
-        #         amp_dispersion_threshold=amp_dispersion_threshold,
-        #     ),
-        #     log_file=log_file,
-        # )
-        # Definite hard coded things
-        unwrap_options=dict(
-            unwrap_method="snaphu",
-            run_unwrap=run_unwrap,
-            # CHANGEME: or else run in background somehow?
-        ),
-        save_compressed_slc=True,  # always save, and only sometimes will we grab it
-        workflow_name=workflow_name,
-    )
-    return cfg
-
-
-def get_cli_args() -> argparse.Namespace:
-    """Set up the command line interface."""
-    parser = argparse.ArgumentParser(
-        description="Repeatedly run the dolphin single update mode.",
-        formatter_class=argparse.ArgumentDefaultsHelpFormatter,
-        # https://docs.python.org/3/library/argparse.html#fromfile-prefix-chars
-        fromfile_prefix_chars="@",
-    )
-    parser.add_argument(
-        "--slc-files",
-        nargs=argparse.ONE_OR_MORE,
-        help="List the paths of all SLC files to include.",
-        required=True,
-    )
-    parser.add_argument(
-        "-ms",
-        "--ministack-size",
-        type=int,
-        default=10,
-        help="Strides/decimation factor (x, y) (in pixels) to use when determining",
-    )
-    parser.add_argument(
-        "-hw",
-        "--half-window-size",
-        type=int,
-        nargs=2,
-        default=(11, 5),
-        metavar=("X", "Y"),
-        help="Half window size for the phase linking algorithm",
-    )
-    parser.add_argument(
-        "--shp-method",
-        type=ShpMethod,
-        choices=[s.value for s in ShpMethod],
-        default=ShpMethod.GLRT,
-        help="Method used to calculate the SHP.",
-    )
-    parser.add_argument(
-        "--run-unwrap",
-        action="store_true",
-        help="Run the unwrapping stack after phase linking.",
-    )
-    parser.add_argument(
-        "-j",
-        "--n-parallel-bursts",
-        type=int,
-        default=1,
-        help="Number of parallel bursts to process.",
-    )
-    parser.add_argument(
-        "--pre-compute",
-        action="store_true",
-        help=(
-            "Run the amplitude mean/dispersion pre-compute step (not the main"
-            " workflow)."
-        ),
-    )
-    return parser.parse_args()
-
-
-@log_runtime
-def compute_ps_files(
-    burst_grouped_slc_files: Mapping[str, Sequence[Filename]],
-    burst_to_nodata_mask: Mapping[str, Filename],
-    # max_workers: int = 3,
-    ps_stack_size: int = 60,
-    output_folder: Path = Path("precomputed_ps"),
-):
-    """Compute the mean/DA/PS files for each burst group."""
-    all_amp_files, all_disp_files, all_ps_files = [], [], []
-    # future_burst_dict = {}
-    # with ThreadPoolExecutor(max_workers=max_workers) as exc:
-    for burst, file_list in burst_grouped_slc_files.items():
-        nodata_mask_file = burst_to_nodata_mask[burst]
-        # fut = exc.submit(
-        # _compute_burst_ps_files,
-        amp_files, disp_files, ps_files = _compute_burst_ps_files(
-            burst,
-            file_list,
-            nodata_mask_file=nodata_mask_file,
-            ps_stack_size=ps_stack_size,
-            output_folder=output_folder,
-            # show_progress=False,
-        )
-        # future_burst_dict[fut] = burst
-
-        # for future in as_completed(future_burst_dict.keys()):
-        # burst = future_burst_dict[future]
-        # amp_files, disp_files, ps_files = future.result()
-
-        all_amp_files.extend(amp_files)
-        all_disp_files.extend(disp_files)
-        all_ps_files.extend(ps_files)
-
-        logger.info(f"Done with {burst}")
-
-    return all_amp_files, all_disp_files, all_ps_files
-
-
-@log_runtime
-def _compute_burst_ps_files(
-    burst: str,
-    file_list_all: Sequence[Filename],
-    nodata_mask_file: Filename,
-    ps_stack_size: int = 60,
-    output_folder: Path = Path("precomputed_ps"),
-) -> tuple[list[Path], list[Path], list[Path]]:
-    """Pre-compute the PS files (mean / amp. dispersion) for one burst."""
-    logger.info(f"Computing PS files for {burst} into {output_folder}")
-    vrt_all = stack.VRTStack(
-        file_list_all, subdataset=OPERA_DATASET_NAME, write_file=False
-    )
-    # logger.info("Created total vrt")
-    date_list_all = vrt_all.dates
-
-    nodata_mask = io.load_gdal(nodata_mask_file, masked=True).astype(bool).filled(False)
-    # invert the mask so 1s are the missing data pixels
-    nodata_mask = ~nodata_mask
-
-    # TODO: fixed number of PS files? fixed time window?
-    amp_files, disp_files, ps_files = [], [], []
-    for full_stack_idx in range(0, len(file_list_all), ps_stack_size):
-        cur_slice = slice(full_stack_idx, full_stack_idx + ps_stack_size)
-        cur_files = file_list_all[cur_slice]
-        cur_dates = date_list_all[cur_slice]
-
-        # Make the current ministack output folder using the start/end dates
-        d0 = cur_dates[0][0]
-        d1 = cur_dates[-1][0]
-        start_end = io._format_date_pair(d0, d1)
-        basename = f"{burst}_{start_end}"
-
-        # output_folder = output_folder / start_end
-        output_folder.mkdir(parents=True, exist_ok=True)
-        cur_vrt = stack.VRTStack(
-            cur_files,
-            outfile=output_folder / f"{basename}.vrt",
-            subdataset=OPERA_DATASET_NAME,
-        )
-        cur_ps_file = (output_folder / f"{basename}_ps_pixels.tif").resolve()
-        cur_amp_mean = (output_folder / f"{basename}_amp_mean.tif").resolve()
-        cur_amp_dispersion = (
-            output_folder / f"{basename}_amp_dispersion.tif"
-        ).resolve()
-        if not all(f.exists() for f in [cur_ps_file, cur_amp_mean, cur_amp_dispersion]):
-            ps.create_ps(
-                slc_vrt_file=cur_vrt,
-                output_amp_mean_file=cur_amp_mean,
-                output_amp_dispersion_file=cur_amp_dispersion,
-                output_file=cur_ps_file,
-                nodata_mask=nodata_mask,
-            )
-        else:
-            logger.info(f"Skipping existing {basename} files in {output_folder}")
-        amp_files.append(cur_amp_mean)
-        disp_files.append(cur_amp_dispersion)
-        ps_files.append(cur_ps_file)
-        logger.info(f"Finished with PS processing for {burst}")
-    return amp_files, disp_files, ps_files
-
-
-def create_nodata_masks(
-    # date_grouped_slc_files: dict[tuple[datetime.date], list[Filename]],
-    burst_grouped_slc_files: Mapping[str, Sequence[Filename]],
-    buffer_pixels: int = 30,
-    max_workers: int = 3,
-    output_folder: Path = Path("nodata_masks"),
-):
-    """Create the nodata binary masks for each burst."""
-    output_folder.mkdir(exist_ok=True, parents=True)
-    futures = []
-    out_burst_to_file = {}
-    with ThreadPoolExecutor(max_workers=max_workers) as exc:
-        for burst, file_list in burst_grouped_slc_files.items():
-            outfile = output_folder / f"{burst}.tif"
-            fut = exc.submit(
-                make_nodata_mask,
-                file_list,
-                outfile,
-                buffer_pixels=buffer_pixels,
-            )
-            futures.append(fut)
-            out_burst_to_file[burst] = outfile
-        for fut in futures:
-            fut.result()
-    return out_burst_to_file
-
-
-def _form_burst_vrt_stacks(
-    burst_grouped_slc_files: Mapping[str, Sequence[Filename]]
-) -> dict[str, stack.VRTStack]:
-    logger.info("For each burst, creating a VRTStack...")
-    # Each burst needs to be the same size
-    burst_to_vrt_stack = {}
-    for b, file_list in burst_grouped_slc_files.items():
-        logger.info(f"Checking {len(file_list)} files for {b}")
-        outfile = Path(f"slc_stack_{b}.vrt")
-        if not outfile.exists():
-            vrt = stack.VRTStack(
-                file_list, subdataset=OPERA_DATASET_NAME, outfile=outfile
-            )
-        else:
-            vrt = stack.VRTStack.from_vrt_file(outfile, skip_size_check=True)
-        burst_to_vrt_stack[b] = vrt
-    logger.info("Done.")
-    return burst_to_vrt_stack
-
-
-@log_runtime
-def precompute_ps_files(arg_dict) -> None:
-    """Run the pre-compute step to get means/amp. dispersion for each burst."""
-    all_slc_files = arg_dict.pop("slc_files")
-
-    burst_grouped_slc_files = group_by_burst(all_slc_files)
-    #  {'t173_370312_iw2': [PosixPath('t173_370312_iw2_20170203.h5'),... ] }
-    date_grouped_slc_files = utils.group_by_date(all_slc_files)
-    #  { (datetime.date(2017, 5, 22),) : [PosixPath('t173_370311_iw1_20170522.h5'), ] }
-    logger.info(f"Found {len(all_slc_files)} total SLC files")
-    logger.info(f"  {len(date_grouped_slc_files)} unique dates,")
-    logger.info(f"  {len(burst_grouped_slc_files)} unique bursts.")
-
-    burst_to_nodata_mask = create_nodata_masks(burst_grouped_slc_files)
-
-    all_amp_files, all_disp_files, all_ps_files = compute_ps_files(
-        burst_grouped_slc_files, burst_to_nodata_mask
-    )
-
-
-def _get_all_slc_files(
-    burst_to_file_list: Mapping[str, Sequence[Filename]], start_idx: int, end_idx: int
-) -> list[Filename]:
-    return list(
-        chain.from_iterable(
-            [file_list[start_idx:end_idx] for file_list in burst_to_file_list.values()]
-        )
-    )
-
-
-def _run_one_stack(
-    slc_idx_start: int,
-    slc_idx_end: int,
-    ministack_size: int,
-    burst_to_file_list: Mapping[str, Sequence[Filename]],
-    comp_slc_files: Sequence[Filename],
-    all_amp_files: Sequence[Filename],
-    all_disp_files: Sequence[Filename],
-):
-    cur_path = Path(f"stack_{slc_idx_start}_{slc_idx_end}")
-    cur_path.mkdir(exist_ok=True)
-
-    logger.info(f"***** START: {cur_path} *****")
-    # Get the nearest amplitude mean/dispersion files
-    cur_slc_files = _get_all_slc_files(burst_to_file_list, slc_idx_start, slc_idx_end)
-    cfg = _create_cfg(
-        slc_files=list(comp_slc_files) + cur_slc_files,
-        amplitude_mean_files=all_amp_files,
-        amplitude_dispersion_files=all_disp_files,
-        work_dir=cur_path,
-        **arg_dict,
-    )
-    cfg.to_yaml(cur_path / "dolphin_config.yaml")
-    s1_disp.run(cfg)
-
-    # On the step before we hit double `ministack_size`,
-    # archive, shrink, and pull another compressed SLC to replace.
-    stack_size = slc_idx_end - slc_idx_start
-    max_stack_size = 2 * ministack_size - 1  # Size at which we archive/shrink
-    if stack_size == max_stack_size:
-        # time to shrink!
-        # Get the compressed SLC that was output
-        comp_slc_path = (cur_path / "output/compressed_slcs/").resolve()
-        new_comp_slcs = list(comp_slc_path.glob("*.h5"))
-    else:
-        new_comp_slcs = []
-
-    logger.info(f"***** END: {cur_path} *****")
-    return new_comp_slcs
-
-
-@log_runtime
-def main(arg_dict: dict) -> None:
-    """Get the command line arguments and run the workflow."""
-    arg_dict = vars(args)
-    ministack_size = arg_dict.pop("ministack_size")
-    # TODO: verify this is fine to sort them by date?
-    all_slc_files = sorted(arg_dict.pop("slc_files"))
-    logger.info(f"Found {len(all_slc_files)} total SLC files")
-
-    # format of `group_by_burst`:
-    #   {'t173_370312_iw2': [PosixPath('t173_370312_iw2_20170203.h5'),... ] }
-    burst_grouped_slc_files = group_by_burst(all_slc_files)
-    num_bursts = len(burst_grouped_slc_files)
-    logger.info(f"  {num_bursts} unique bursts.")
-    # format of `group_by_date`:
-    #  { (datetime.date(2017, 5, 22),) : [PosixPath('t173_370311_iw1_20170522.h5'), ] }
-    date_grouped_slc_files = utils.group_by_date(all_slc_files)
-    num_dates = len(date_grouped_slc_files)
-    logger.info(f"  {num_dates} unique dates,")
-
-    burst_to_vrt_stack = _form_burst_vrt_stacks(
-        burst_grouped_slc_files=burst_grouped_slc_files
-    )
-    burst_to_file_list = {b: v.file_list for b, v in burst_to_vrt_stack.items()}
-
-    # Get the pre-compted PS files (assuming --pre-compute has been run)
-    all_amp_files = sorted(Path("precomputed_ps/").resolve().glob("*_amp_mean.tif"))
-    all_disp_files = sorted(
-        Path("precomputed_ps/").resolve().glob("*_amp_dispersion.tif")
-    )
-
-    slc_idx_start = 0
-    slc_idx_end = ministack_size
-    # max_stack_size = 2 * ministack_size - 1  # Size at which we archive/shrink
-    cur_path = Path(f"stack_{slc_idx_start}_{slc_idx_end}")
-    cur_path.mkdir(exist_ok=True)
-
-    # TODO: how to make it shift when the year changes for PS files
-
-    # Make the first ministack
-    cur_slc_files = _get_all_slc_files(burst_to_file_list, slc_idx_start, slc_idx_end)
-    cfg = _create_cfg(
-        slc_files=cur_slc_files,
-        first_ministack=True,
-        amplitude_mean_files=all_amp_files,
-        amplitude_dispersion_files=all_disp_files,
-        work_dir=cur_path,
-        **arg_dict,
-    )
-    cfg.to_yaml(cur_path / "dolphin_config.yaml")
-    s1_disp.run(cfg)
-
-    # Rest of mini stacks in incremental-mode
-    comp_slc_files: list[Path] = []
-    slc_idx_end = ministack_size + 1
-
-    while slc_idx_end <= num_dates:
-        # we have to wait for the shrink-and-archive jobs before continuing
-        new_comp_slcs = _run_one_stack(
-            slc_idx_start,
-            slc_idx_end,
-            ministack_size,
-            burst_to_file_list,
-            comp_slc_files,
-            all_amp_files,
-            all_disp_files,
-        )
-        logger.info(
-            f"{len(new_comp_slcs)} comp slcs from stack_{slc_idx_start}_{slc_idx_end}"
-        )
-        comp_slc_files.extend(new_comp_slcs)
-        slc_idx_end += 1
-        if len(new_comp_slcs) > 0:
-            # Move the front idx up by one ministack
-            slc_idx_start += ministack_size
-
-
-if __name__ == "__main__":
-    args = get_cli_args()
-    arg_dict = vars(args)
-    if arg_dict.pop("pre_compute"):
-        precompute_ps_files(arg_dict)
-    else:
-        main(arg_dict)
diff --git a/src/dolphin/py.typed b/src/dolphin/py.typed
new file mode 100644
index 00000000..e69de29b
diff --git a/src/dolphin/workflows/_cli_run.py b/src/dolphin/workflows/_cli_run.py
index c7944980..c271ba12 100644
--- a/src/dolphin/workflows/_cli_run.py
+++ b/src/dolphin/workflows/_cli_run.py
@@ -11,7 +11,6 @@
 def run(
     config_file: str,
     debug: bool = False,
-    pge_format: bool = False,
 ) -> None:
     """Run the displacement workflow.
 
@@ -21,24 +20,14 @@ def run(
         YAML file containing the workflow options.
     debug : bool, optional
         Enable debug logging, by default False.
-    pge_format : bool, optional
-        If True, the config file is a runconfig in the PGE-expected format.
-        By default False.
     """
     # rest of imports here so --help doesn't take forever
 
     from . import s1_disp
-    from ._pge_runconfig import RunConfig
     from .config import Workflow
 
-    if pge_format:
-        pge_rc = RunConfig.from_yaml(config_file)
-        cfg = pge_rc.to_workflow()
-    else:
-        cfg = Workflow.from_yaml(config_file)
-        pge_rc = None
-
-    s1_disp.run(cfg, debug=debug, pge_runconfig=pge_rc)
+    cfg = Workflow.from_yaml(config_file)
+    s1_disp.run(cfg, debug=debug)
 
 
 def get_parser(
@@ -64,11 +53,6 @@ def get_parser(
         action="store_true",
         help="Print debug messages to the log.",
     )
-    parser.add_argument(
-        "--pge-format",
-        action="store_true",
-        help="Indicate that `config_file` is in the PGE `RunConfig` format.",
-    )
     parser.set_defaults(run_func=run)
     return parser
 
diff --git a/src/dolphin/workflows/_pge_runconfig.py b/src/dolphin/workflows/_pge_runconfig.py
deleted file mode 100644
index 73bd851e..00000000
--- a/src/dolphin/workflows/_pge_runconfig.py
+++ /dev/null
@@ -1,285 +0,0 @@
-"""Module for creating PGE-compatible run configuration files."""
-from __future__ import annotations
-
-from pathlib import Path
-from typing import ClassVar, List, Optional
-
-from pydantic import ConfigDict, Field
-
-from ._yaml_model import YamlModel
-from .config import (
-    OPERA_DATASET_NAME,
-    InterferogramNetwork,
-    OutputOptions,
-    PhaseLinkingOptions,
-    PsOptions,
-    UnwrapOptions,
-    WorkerSettings,
-    Workflow,
-)
-
-
-class InputFileGroup(YamlModel):
-    """A group of input files."""
-
-    cslc_file_list: List[Path] = Field(
-        default_factory=list,
-        description="list of paths to CSLC files.",
-    )
-
-    frame_id: int = Field(
-        ...,
-        description="Frame ID of the bursts contained in `cslc_file_list`.",
-    )
-    model_config = ConfigDict(
-        extra="forbid", json_schema_extra={"required": ["cslc_file_list", "frame_id"]}
-    )
-
-
-class DynamicAncillaryFileGroup(YamlModel, extra="forbid"):
-    """A group of dynamic ancillary files."""
-
-    algorithm_parameters_file: Path = Field(  # type: ignore
-        default=...,
-        description="Path to file containing SAS algorithm parameters.",
-    )
-    amplitude_dispersion_files: List[Path] = Field(
-        default_factory=list,
-        description=(
-            "Paths to existing Amplitude Dispersion files (1 per burst) for PS update"
-            " calculation. If none provided, computed using the input SLC stack."
-        ),
-    )
-    amplitude_mean_files: List[Path] = Field(
-        default_factory=list,
-        description=(
-            "Paths to an existing Amplitude Mean files (1 per burst) for PS update"
-            " calculation. If none provided, computed using the input SLC stack."
-        ),
-    )
-    geometry_files: List[Path] = Field(
-        default_factory=list,
-        description=(
-            "Paths to the incidence/azimuth-angle files (1 per burst). If none"
-            " provided, corrections using incidence/azimuth-angle are skipped."
-        ),
-    )
-    mask_file: Optional[Path] = Field(
-        None,
-        description=(
-            "Optional Byte mask file used to ignore low correlation/bad data (e.g water"
-            " mask). Convention is 0 for no data/invalid, and 1 for good data. Dtype"
-            " must be uint8."
-        ),
-    )
-    dem_file: Optional[Path] = Field(
-        default=None,
-        description=(
-            "Path to the DEM file covering full frame. If none provided, corrections"
-            " using DEM are skipped."
-        ),
-    )
-    # TEC file in IONEX format for ionosphere correction
-    tec_files: Optional[List[Path]] = Field(
-        default=None,
-        description=(
-            "List of paths to TEC files (1 per date) in IONEX format for ionosphere"
-            " correction. If none provided, ionosphere corrections are skipped."
-        ),
-    )
-
-    # Troposphere weather model
-    weather_model_files: Optional[List[Path]] = Field(
-        default=None,
-        description=(
-            "List of paths to troposphere weather model files (1 per date). If none"
-            " provided, troposphere corrections are skipped."
-        ),
-    )
-
-
-class PrimaryExecutable(YamlModel, extra="forbid"):
-    """Group describing the primary executable."""
-
-    product_type: str = Field(
-        default="DISP_S1_SINGLE",
-        description="Product type of the PGE.",
-    )
-
-
-class ProductPathGroup(YamlModel, extra="forbid"):
-    """Group describing the product paths."""
-
-    product_path: Path = Field(  # type: ignore
-        default=...,
-        description="Directory where PGE will place results",
-    )
-    scratch_path: Path = Field(
-        default=Path("./scratch"),
-        description="Path to the scratch directory.",
-    )
-    output_directory: Path = Field(
-        default=Path("./output"),
-        description="Path to the SAS output directory.",
-        # The alias means that in the YAML file, the key will be "sas_output_path"
-        # instead of "output_directory", but the python instance attribute is
-        # "output_directory" (to match Workflow)
-        alias="sas_output_path",
-    )
-    product_version: str = Field(
-        default="0.1",
-        description="Version of the product, in <major>.<minor> format.",
-    )
-    save_compressed_slc: bool = Field(
-        default=False,
-        description=(
-            "Whether the SAS should output and save the Compressed SLCs in addition to"
-            " the standard product output."
-        ),
-    )
-
-
-class AlgorithmParameters(YamlModel, extra="forbid"):
-    """Class containing all the other [`Workflow`][dolphin.workflows.config] classes."""
-
-    # Options for each step in the workflow
-    ps_options: PsOptions = Field(default_factory=PsOptions)
-    phase_linking: PhaseLinkingOptions = Field(default_factory=PhaseLinkingOptions)
-    interferogram_network: InterferogramNetwork = Field(
-        default_factory=InterferogramNetwork
-    )
-    unwrap_options: UnwrapOptions = Field(default_factory=UnwrapOptions)
-    output_options: OutputOptions = Field(default_factory=OutputOptions)
-    subdataset: str = Field(
-        default=OPERA_DATASET_NAME,
-        description="Name of the subdataset to use in the input NetCDF files.",
-    )
-
-
-class RunConfig(YamlModel, extra="forbid"):
-    """A PGE run configuration."""
-
-    # Used for the top-level key
-    name: ClassVar[str] = "disp_s1_workflow"
-
-    input_file_group: InputFileGroup
-    dynamic_ancillary_file_group: DynamicAncillaryFileGroup
-    primary_executable: PrimaryExecutable = Field(default_factory=PrimaryExecutable)
-    product_path_group: ProductPathGroup
-
-    # General workflow metadata
-    worker_settings: WorkerSettings = Field(default_factory=WorkerSettings)
-
-    log_file: Optional[Path] = Field(
-        default=Path("output/disp_s1_workflow.log"),
-        description="Path to the output log file in addition to logging to stderr.",
-    )
-
-    # Override the constructor to allow recursively model_construct without validation
-    @classmethod
-    def model_construct(cls, **kwargs):
-        if "input_file_group" not in kwargs:
-            kwargs["input_file_group"] = InputFileGroup._construct_empty()
-        if "dynamic_ancillary_file_group" not in kwargs:
-            kwargs["dynamic_ancillary_file_group"] = (
-                DynamicAncillaryFileGroup._construct_empty()
-            )
-        if "product_path_group" not in kwargs:
-            kwargs["product_path_group"] = ProductPathGroup._construct_empty()
-        return super().model_construct(
-            **kwargs,
-        )
-
-    def to_workflow(self):
-        """Convert to a [`Workflow`][dolphin.workflows.config.Workflow] object."""
-        # We need to go to/from the PGE format to our internal Workflow object:
-        # Note that the top two levels of nesting can be accomplished by wrapping
-        # the normal model export in a dict.
-        #
-        # The things from the RunConfig that are used in the
-        # Workflow are the input files, PS amp mean/disp files,
-        # the output directory, and the scratch directory.
-        # All the other things come from the AlgorithmParameters.
-
-        workflow_name = self.primary_executable.product_type.replace(
-            "DISP_S1_", ""
-        ).lower()
-        cslc_file_list = self.input_file_group.cslc_file_list
-        output_directory = self.product_path_group.output_directory
-        scratch_directory = self.product_path_group.scratch_path
-        mask_file = self.dynamic_ancillary_file_group.mask_file
-        amplitude_mean_files = self.dynamic_ancillary_file_group.amplitude_mean_files
-        amplitude_dispersion_files = (
-            self.dynamic_ancillary_file_group.amplitude_dispersion_files
-        )
-
-        # Load the algorithm parameters from the file
-        algorithm_parameters = AlgorithmParameters.from_yaml(
-            self.dynamic_ancillary_file_group.algorithm_parameters_file
-        )
-        param_dict = algorithm_parameters.model_dump()
-        input_options = dict(subdataset=param_dict.pop("subdataset"))
-
-        # This get's unpacked to load the rest of the parameters for the Workflow
-        return Workflow(
-            workflow_name=workflow_name,
-            cslc_file_list=cslc_file_list,
-            input_options=input_options,
-            mask_file=mask_file,
-            output_directory=output_directory,
-            scratch_directory=scratch_directory,
-            save_compressed_slc=self.product_path_group.save_compressed_slc,
-            amplitude_mean_files=amplitude_mean_files,
-            amplitude_dispersion_files=amplitude_dispersion_files,
-            # These ones directly translate
-            worker_settings=self.worker_settings,
-            log_file=self.log_file,
-            # Finally, the rest of the parameters are in the algorithm parameters
-            **param_dict,
-        )
-
-    @classmethod
-    def from_workflow(
-        cls, workflow: Workflow, frame_id: int, algorithm_parameters_file: Path
-    ):
-        """Convert from a [`Workflow`][dolphin.workflows.config.Workflow] object.
-
-        This is the inverse of the to_workflow method, although there are more
-        fields in the PGE version, so it's not a 1-1 mapping.
-
-        Since there's no `frame_id` or `algorithm_parameters_file` in the
-        [`Workflow`][dolphin.workflows.config.Workflow] object, we need to pass
-        those in as arguments.
-
-        This is mostly used as preliminary setup to further edit the fields.
-        """
-        # Load the algorithm parameters from the file
-        algo_keys = set(AlgorithmParameters.model_fields.keys())
-        alg_param_dict = workflow.model_dump(include=algo_keys)
-        AlgorithmParameters(**alg_param_dict).to_yaml(algorithm_parameters_file)
-        # This get's unpacked to load the rest of the parameters for the Workflow
-
-        return cls(
-            input_file_group=InputFileGroup(
-                cslc_file_list=workflow.cslc_file_list,
-                frame_id=frame_id,
-            ),
-            dynamic_ancillary_file_group=DynamicAncillaryFileGroup(
-                algorithm_parameters_file=algorithm_parameters_file,
-                # amplitude_dispersion_files=workflow.amplitude_dispersion_files,
-                # amplitude_mean_files=workflow.amplitude_mean_files,
-                mask_file=workflow.mask_file,
-                # tec_file=workflow.tec_file,
-                # weather_model_file=workflow.weather_model_file,
-            ),
-            primary_executable=PrimaryExecutable(
-                product_type=f"DISP_S1_{str(workflow.workflow_name.upper())}",
-            ),
-            product_path_group=ProductPathGroup(
-                product_path=workflow.output_directory,
-                scratch_path=workflow.scratch_directory,
-                sas_output_path=workflow.output_directory,
-            ),
-            worker_settings=workflow.worker_settings,
-            log_file=workflow.log_file,
-        )
diff --git a/src/dolphin/workflows/_product.py b/src/dolphin/workflows/_product.py
deleted file mode 100644
index 73cae207..00000000
--- a/src/dolphin/workflows/_product.py
+++ /dev/null
@@ -1,482 +0,0 @@
-"""Module for creating the OPERA output product in NetCDF format."""
-from __future__ import annotations
-
-from io import StringIO
-from pathlib import Path
-from typing import Any, Optional, Sequence, Union
-
-import h5netcdf
-import h5py
-import numpy as np
-import pyproj
-from numpy.typing import ArrayLike, DTypeLike
-from PIL import Image
-
-from dolphin import __version__ as dolphin_version
-from dolphin import io
-from dolphin._log import get_log
-from dolphin._types import Filename
-from dolphin.utils import get_dates
-
-from ._pge_runconfig import RunConfig
-from .config import OPERA_DATASET_NAME
-
-logger = get_log(__name__)
-
-
-CORRECTIONS_GROUP_NAME = "corrections"
-IDENTIFICATION_GROUP_NAME = "identification"
-GLOBAL_ATTRS = dict(
-    Conventions="CF-1.8",
-    contact="operaops@jpl.nasa.gov",
-    institution="NASA JPL",
-    mission_name="OPERA",
-    reference_document="TBD",
-    title="OPERA L3_DISP_S1 Product",
-)
-
-# Convert chunks to a tuple or h5py errors
-HDF5_OPTS = io.DEFAULT_HDF5_OPTIONS.copy()
-HDF5_OPTS["chunks"] = tuple(HDF5_OPTS["chunks"])  # type: ignore
-# The GRID_MAPPING_DSET variable is used to store the name of the dataset containing
-# the grid mapping information, which includes the coordinate reference system (CRS)
-# and the GeoTransform. This is in accordance with the CF 1.8 conventions for adding
-# geospatial metadata to NetCDF files.
-# http://cfconventions.org/cf-conventions/cf-conventions.html#grid-mappings-and-projections
-# Note that the name "spatial_ref" used here is arbitrary, but it follows the default
-# used by other libraries, such as rioxarray:
-# https://github.com/corteva/rioxarray/blob/5783693895b4b055909c5758a72a5d40a365ef11/rioxarray/rioxarray.py#L34 # noqa
-GRID_MAPPING_DSET = "spatial_ref"
-
-
-def create_output_product(
-    unw_filename: Filename,
-    conncomp_filename: Filename,
-    tcorr_filename: Filename,
-    spatial_corr_filename: Filename,
-    output_name: Filename,
-    corrections: dict[str, ArrayLike] = {},
-    pge_runconfig: Optional[RunConfig] = None,
-    create_browse_image: bool = True,
-):
-    """Create the OPERA output product in NetCDF format.
-
-    Parameters
-    ----------
-    unw_filename : Filename
-        The path to the input unwrapped phase image.
-    conncomp_filename : Filename
-        The path to the input connected components image.
-    tcorr_filename : Filename
-        The path to the input temporal correlation image.
-    spatial_corr_filename : Filename
-        The path to the input spatial correlation image.
-    output_name : Filename, optional
-        The path to the output NetCDF file, by default "output.nc"
-    corrections : dict[str, ArrayLike], optional
-        A dictionary of corrections to write to the output file, by default None
-    pge_runconfig : Optional[RunConfig], optional
-        The PGE run configuration, by default None
-        Used to add extra metadata to the output file.
-    create_browse_image : bool
-        If true, creates a PNG browse image of the unwrapped phase
-        with filename as `output_name`.png
-    """
-    # Read the Geotiff file and its metadata
-    crs = io.get_raster_crs(unw_filename)
-    gt = io.get_raster_gt(unw_filename)
-    unw_arr = io.load_gdal(unw_filename)
-
-    conncomp_arr = io.load_gdal(conncomp_filename)
-    tcorr_arr = _zero_mantissa(io.load_gdal(tcorr_filename))
-    # TODO: add spatial correlation, pass through to function
-    spatial_corr_arr = _zero_mantissa(io.load_gdal(spatial_corr_filename))
-
-    # Get the nodata mask (which for snaphu is 0)
-    mask = unw_arr == 0
-    # Set to NaN for final output
-    unw_arr[mask] = np.nan
-
-    assert unw_arr.shape == conncomp_arr.shape == tcorr_arr.shape
-
-    if create_browse_image:
-        make_browse_image(Path(output_name).with_suffix(".png"), unw_arr)
-
-    with h5netcdf.File(output_name, "w") as f:
-        # Create the NetCDF file
-        f.attrs.update(GLOBAL_ATTRS)
-
-        # Set up the grid mapping variable for each group with rasters
-        _create_grid_mapping(group=f, crs=crs, gt=gt)
-
-        # Set up the X/Y variables for each group
-        _create_yx_dsets(group=f, gt=gt, shape=unw_arr.shape)
-
-        # Write the displacement array / conncomp arrays
-        _create_geo_dataset(
-            group=f,
-            name="unwrapped_phase",
-            data=unw_arr,
-            description="Unwrapped phase",
-            fillvalue=np.nan,
-            attrs=dict(units="radians"),
-        )
-        _create_geo_dataset(
-            group=f,
-            name="connected_component_labels",
-            data=conncomp_arr,
-            description="Connected component labels of the unwrapped phase",
-            fillvalue=0,
-            attrs=dict(units="unitless"),
-        )
-        _create_geo_dataset(
-            group=f,
-            name="temporal_correlation",
-            data=tcorr_arr,
-            description="Temporal correlation of phase inversion",
-            fillvalue=np.nan,
-            attrs=dict(units="unitless"),
-        )
-        _create_geo_dataset(
-            group=f,
-            name="spatial_correlation",
-            data=spatial_corr_arr,
-            description="Multilooked sample interferometric correlation",
-            fillvalue=np.nan,
-            attrs=dict(units="unitless"),
-        )
-
-        # Create the group holding phase corrections that were used on the unwrapped phase
-        corrections_group = f.create_group(CORRECTIONS_GROUP_NAME)
-        corrections_group.attrs["description"] = (
-            "Phase corrections applied to the unwrapped_phase"
-        )
-
-        # TODO: Are we going to downsample these for space?
-        # if so, they need they're own X/Y variables and GeoTransform
-        _create_grid_mapping(group=corrections_group, crs=crs, gt=gt)
-        _create_yx_dsets(group=corrections_group, gt=gt, shape=unw_arr.shape)
-        troposphere = corrections.get("troposphere", np.zeros_like(unw_arr))
-        _create_geo_dataset(
-            group=corrections_group,
-            name="tropospheric_delay",
-            data=troposphere,
-            description="Tropospheric phase delay used to correct the unwrapped phase",
-            fillvalue=np.nan,
-            attrs=dict(units="radians"),
-        )
-        ionosphere = corrections.get("ionosphere", np.zeros_like(unw_arr))
-        _create_geo_dataset(
-            group=corrections_group,
-            name="ionospheric_delay",
-            data=ionosphere,
-            description="Ionospheric phase delay used to correct the unwrapped phase",
-            fillvalue=np.nan,
-            attrs=dict(units="radians"),
-        )
-        solid_earth = corrections.get("solid_earth", np.zeros_like(unw_arr))
-        _create_geo_dataset(
-            group=corrections_group,
-            name="solid_earth_tide",
-            data=solid_earth,
-            description="Solid Earth tide used to correct the unwrapped phase",
-            fillvalue=np.nan,
-            attrs=dict(units="radians"),
-        )
-        plate_motion = corrections.get("plate_motion", np.zeros_like(unw_arr))
-        _create_geo_dataset(
-            group=corrections_group,
-            name="plate_motion",
-            data=plate_motion,
-            description="Phase ramp caused by tectonic plate motion",
-            fillvalue=np.nan,
-            attrs=dict(units="radians"),
-        )
-        # Make a scalar dataset for the reference point
-        reference_point = corrections.get("reference_point", 0.0)
-        _create_dataset(
-            group=corrections_group,
-            name="reference_point",
-            dimensions=(),
-            data=reference_point,
-            fillvalue=0,
-            description=(
-                "Dummy dataset containing attributes with the locations where the"
-                " reference phase was taken."
-            ),
-            dtype=int,
-            # Note: the dataset contains attributes with lists, since the reference
-            # could have come from multiple points (e.g. some boxcar average of an area).
-            attrs=dict(units="unitless", rows=[], cols=[], latitudes=[], longitudes=[]),
-        )
-
-    # End of the product for non-PGE users
-    if pge_runconfig is None:
-        return
-
-    # Add the PGE metadata to the file
-    with h5netcdf.File(output_name, "a") as f:
-        identification_group = f.create_group(IDENTIFICATION_GROUP_NAME)
-        _create_dataset(
-            group=identification_group,
-            name="frame_id",
-            dimensions=(),
-            data=pge_runconfig.input_file_group.frame_id,
-            fillvalue=None,
-            description="ID number of the processed frame.",
-            attrs=dict(units="unitless"),
-        )
-        # product_version
-        _create_dataset(
-            group=identification_group,
-            name="product_version",
-            dimensions=(),
-            data=pge_runconfig.product_path_group.product_version,
-            fillvalue=None,
-            description="Version of the product.",
-            attrs=dict(units="unitless"),
-        )
-        # software_version
-        _create_dataset(
-            group=identification_group,
-            name="software_version",
-            dimensions=(),
-            data=dolphin_version,
-            fillvalue=None,
-            description="Version of the Dolphin software used to generate the product.",
-            attrs=dict(units="unitless"),
-        )
-
-        # TODO: prob should just make a _to_string method?
-        ss = StringIO()
-        pge_runconfig.to_yaml(ss)
-        runconfig_str = ss.getvalue()
-        _create_dataset(
-            group=identification_group,
-            name="pge_runconfig",
-            dimensions=(),
-            data=runconfig_str,
-            fillvalue=None,
-            description=(
-                "The full PGE runconfig YAML file used to generate the product."
-            ),
-            attrs=dict(units="unitless"),
-        )
-
-
-def _create_dataset(
-    *,
-    group: h5netcdf.Group,
-    name: str,
-    dimensions: Optional[Sequence[str]],
-    data: Union[np.ndarray, str],
-    description: str,
-    fillvalue: Optional[float],
-    attrs: Optional[dict[str, Any]] = None,
-    dtype: Optional[DTypeLike] = None,
-) -> h5netcdf.Variable:
-    if attrs is None:
-        attrs = {}
-    attrs.update(long_name=description)
-
-    options = HDF5_OPTS
-    if isinstance(data, str):
-        options = {}
-        # This is a string, so we need to convert it to bytes or it will fail
-        data = np.string_(data)
-    elif np.array(data).size <= 1:
-        # Scalars don't need chunks/compression
-        options = {}
-    dset = group.create_variable(
-        name,
-        dimensions=dimensions,
-        data=data,
-        dtype=dtype,
-        fillvalue=fillvalue,
-        **options,
-    )
-    dset.attrs.update(attrs)
-    return dset
-
-
-def _create_geo_dataset(
-    *,
-    group: h5netcdf.Group,
-    name: str,
-    data: np.ndarray,
-    description: str,
-    fillvalue: float,
-    attrs: Optional[dict[str, Any]],
-) -> h5netcdf.Variable:
-    dimensions = ["y", "x"]
-    dset = _create_dataset(
-        group=group,
-        name=name,
-        dimensions=dimensions,
-        data=data,
-        description=description,
-        fillvalue=fillvalue,
-        attrs=attrs,
-    )
-    dset.attrs["grid_mapping"] = GRID_MAPPING_DSET
-    return dset
-
-
-def _create_yx_arrays(
-    gt: list[float], shape: tuple[int, int]
-) -> tuple[np.ndarray, np.ndarray]:
-    """Create the x and y coordinate datasets."""
-    ysize, xsize = shape
-    # Parse the geotransform
-    x_origin, x_res, _, y_origin, _, y_res = gt
-
-    # Make the x/y arrays
-    # Note that these are the center of the pixels, whereas the GeoTransform
-    # is the upper left corner of the top left pixel.
-    y = np.arange(y_origin + y_res / 2, y_origin + y_res * ysize, y_res)
-    x = np.arange(x_origin + x_res / 2, x_origin + x_res * xsize, x_res)
-    return y, x
-
-
-def _create_yx_dsets(
-    group: h5netcdf.Group,
-    gt: list[float],
-    shape: tuple[int, int],
-) -> tuple[h5netcdf.Variable, h5netcdf.Variable]:
-    """Create the x and y coordinate datasets."""
-    y, x = _create_yx_arrays(gt, shape)
-
-    if not group.dimensions:
-        group.dimensions = dict(y=y.size, x=x.size)
-    # Create the datasets
-    y_ds = group.create_variable("y", ("y",), data=y, dtype=float)
-    x_ds = group.create_variable("x", ("x",), data=x, dtype=float)
-
-    for name, ds in zip(["y", "x"], [y_ds, x_ds]):
-        ds.attrs["standard_name"] = f"projection_{name}_coordinate"
-        ds.attrs["long_name"] = f"{name.replace('_', ' ')} coordinate of projection"
-        ds.attrs["units"] = "m"
-
-    return y_ds, x_ds
-
-
-def _create_grid_mapping(group, crs: pyproj.CRS, gt: list[float]) -> h5netcdf.Variable:
-    """Set up the grid mapping variable."""
-    # https://github.com/corteva/rioxarray/blob/21284f67db536d9c104aa872ab0bbc261259e59e/rioxarray/rioxarray.py#L34
-    dset = group.create_variable(GRID_MAPPING_DSET, (), data=0, dtype=int)
-
-    dset.attrs.update(crs.to_cf())
-    # Also add the GeoTransform
-    gt_string = " ".join([str(x) for x in gt])
-    dset.attrs.update(
-        dict(
-            GeoTransform=gt_string,
-            units="unitless",
-            long_name=(
-                "Dummy variable containing geo-referencing metadata in attributes"
-            ),
-        )
-    )
-
-    return dset
-
-
-def create_compressed_products(comp_slc_dict: dict[str, Path], output_dir: Filename):
-    """Make the compressed SLC output product."""
-
-    def form_name(filename: Path, burst: str):
-        # filename: compressed_20180222_20180716.tif
-        date_str = io._format_date_pair(*get_dates(filename.stem))
-        return f"compressed_slc_{burst}_{date_str}.h5"
-
-    attrs = GLOBAL_ATTRS.copy()
-    attrs["title"] = "Compressed SLC"
-    *parts, dset_name = OPERA_DATASET_NAME.split("/")
-    group_name = "/".join(parts)
-
-    for burst, comp_slc_file in comp_slc_dict.items():
-        outname = Path(output_dir) / form_name(comp_slc_file, burst)
-        if outname.exists():
-            logger.info(f"Skipping existing {outname}")
-            continue
-
-        crs = io.get_raster_crs(comp_slc_file)
-        gt = io.get_raster_gt(comp_slc_file)
-        data = _zero_mantissa(io.load_gdal(comp_slc_file))
-
-        logger.info(f"Writing {outname}")
-        with h5py.File(outname, "w") as hf:
-            # add type to root for GDAL recognition of complex datasets in NetCDF
-            ctype = h5py.h5t.py_create(np.complex64)
-            ctype.commit(hf["/"].id, np.string_("complex64"))
-
-        with h5netcdf.File(outname, mode="a", invalid_netcdf=True) as f:
-            f.attrs.update(attrs)
-
-            data_group = f.create_group(group_name)
-            _create_grid_mapping(group=data_group, crs=crs, gt=gt)
-            _create_yx_dsets(group=data_group, gt=gt, shape=data.shape)
-            _create_geo_dataset(
-                group=data_group,
-                name=dset_name,
-                data=data,
-                description="Compressed SLC product",
-                fillvalue=np.nan + 0j,
-                attrs=dict(units="unitless"),
-            )
-
-
-def _zero_mantissa(data: np.ndarray, bits_to_keep: int = 10):
-    """Zero out 23-`bits_to_keep` bits of the mantissa of a float32 array.
-
-    This is used to make the data more compressible when we don't need the
-    full precision (e.g. for correlation estimates).
-
-    By default, this will zero out 13 bits, which (for data between 0 and 1)
-    is `1 / 2**13 ~= 0.0001` of precision.
-    """
-    float32_mantissa_bits = 23
-    nzero = float32_mantissa_bits - bits_to_keep
-
-    # Start with 0b11111111111111111111111111111111
-    allbits = (1 << 32) - 1
-    # Shift it to the left by `nzero` bits
-    bitmask = (allbits << nzero) & allbits
-    # Mask out the least significant `nzero` bits
-    if np.iscomplexobj(data):
-        dr = data.real.view(np.uint32)
-        dr &= bitmask
-        di = data.imag.view(np.uint32)
-        di &= bitmask
-    else:
-        dr = data.view(np.uint32)
-        dr &= bitmask
-    return data
-
-
-def make_browse_image(
-    output_filename: Filename,
-    arr: ArrayLike,
-    max_dim_allowed: int = 2048,
-) -> None:
-    """Create a PNG browse image for the output product.
-
-    Parameters
-    ----------
-    output_filename : Filename
-        Name of output PNG
-    arr : ArrayLike
-        input 2D image array
-    max_dim_allowed : int, default = 2048
-        Size (in pixels) of the maximum allowed dimension of output image.
-        Image gets rescaled with same aspect ratio.
-    """
-    orig_shape = arr.shape
-    scaling_ratio = max([s / max_dim_allowed for s in orig_shape])
-    # scale original shape by scaling ratio
-    scaled_shape = [int(np.ceil(s / scaling_ratio)) for s in orig_shape]
-
-    # TODO: Make actual browse image
-    dummy = np.zeros(scaled_shape, dtype="uint8")
-    img = Image.fromarray(dummy, mode="L")
-    img.save(output_filename, transparency=0)
diff --git a/src/dolphin/workflows/_utils.py b/src/dolphin/workflows/_utils.py
index 7d2df5ef..cc91aa71 100644
--- a/src/dolphin/workflows/_utils.py
+++ b/src/dolphin/workflows/_utils.py
@@ -15,7 +15,7 @@
 from dolphin._log import get_log
 from dolphin._types import Filename
 
-from .config import OPERA_BURST_RE, OPERA_DATASET_NAME, OPERA_IDENTIFICATION
+from .config import OPERA_BURST_RE, OPERA_DATASET_NAME, OPERA_IDENTIFICATION, Workflow
 
 logger = get_log(__name__)
 
@@ -207,3 +207,28 @@ def make_nodata_mask(
         cmd = f"gdal_rasterize -q -burn 1 {temp_vector_file} {out_file}"
         logger.info(cmd)
         subprocess.check_call(cmd, shell=True)
+
+
+def _create_burst_cfg(
+    cfg: Workflow,
+    burst_id: str,
+    grouped_slc_files: dict[str, list[Path]],
+    grouped_amp_mean_files: dict[str, list[Path]],
+    grouped_amp_dispersion_files: dict[str, list[Path]],
+) -> Workflow:
+    cfg_temp_dict = cfg.model_dump(exclude={"cslc_file_list"})
+
+    # Just update the inputs and the work directory
+    top_level_work = cfg_temp_dict["work_directory"]
+    cfg_temp_dict.update({"work_directory": top_level_work / burst_id})
+    cfg_temp_dict["cslc_file_list"] = grouped_slc_files[burst_id]
+    cfg_temp_dict["amplitude_mean_files"] = grouped_amp_mean_files[burst_id]
+    cfg_temp_dict["amplitude_dispersion_files"] = grouped_amp_dispersion_files[burst_id]
+    return Workflow(**cfg_temp_dict)
+
+
+def _remove_dir_if_empty(d: Path) -> None:
+    try:
+        d.rmdir()
+    except OSError:
+        pass
diff --git a/src/dolphin/workflows/config.py b/src/dolphin/workflows/config.py
index e8e742a7..a79d0f04 100644
--- a/src/dolphin/workflows/config.py
+++ b/src/dolphin/workflows/config.py
@@ -326,14 +326,9 @@ class Workflow(YamlModel):
             " uint8."
         ),
     )
-    scratch_directory: Path = Field(
-        Path("scratch"),
-        description="Name of sub-directory to use for scratch files",
-        validate_default=True,
-    )
-    output_directory: Path = Field(
-        Path("output"),
-        description="Name of sub-directory to use for output files",
+    work_directory: Path = Field(
+        Path("."),
+        description="Name of sub-directory to use for writing output files",
         validate_default=True,
     )
 
@@ -387,7 +382,7 @@ class Workflow(YamlModel):
         extra="forbid", json_schema_extra={"required": ["cslc_file_list"]}
     )
 
-    @field_validator("output_directory", "scratch_directory")
+    @field_validator("work_directory")
     @classmethod
     def _make_dir_absolute(cls, v: Path):
         return v.resolve()
@@ -467,13 +462,13 @@ def __init__(self, *args: Any, **kwargs: Any) -> None:
         """After validation, set up properties for use during workflow run."""
         super().__init__(*args, **kwargs)
 
-        # Ensure outputs from workflow steps are within scratch directory.
-        scratch_dir = self.scratch_directory
+        # Ensure outputs from workflow steps are within work directory.
+        work_dir = self.work_directory
         # Save all directories as absolute paths
-        scratch_dir = scratch_dir.resolve(strict=False)
+        work_dir = work_dir.resolve(strict=False)
 
         # For each workflow step that has an output folder, move it inside
-        # the scratch directory (if it's not already inside).
+        # the work directory (if it's not already inside).
         # They may already be inside if we're loading from a json/yaml file.
         for step in [
             "ps_options",
@@ -482,25 +477,24 @@ def __init__(self, *args: Any, **kwargs: Any) -> None:
             "unwrap_options",
         ]:
             opts = getattr(self, step)
-            if not opts._directory.parent == scratch_dir:
-                opts._directory = scratch_dir / opts._directory
+            if not opts._directory.parent == work_dir:
+                opts._directory = work_dir / opts._directory
             opts._directory = opts._directory.resolve(strict=False)
 
         # Track the directories that need to be created at start of workflow
         self._directory_list = [
-            scratch_dir,
-            self.output_directory,
+            work_dir,
             self.ps_options._directory,
             self.phase_linking._directory,
             self.interferogram_network._directory,
             self.unwrap_options._directory,
         ]
         # Add the output PS files we'll create to the `PS` directory, making
-        # sure they're inside the scratch directory
+        # sure they're inside the work directory
         ps_opts = self.ps_options
-        ps_opts._amp_dispersion_file = scratch_dir / ps_opts._amp_dispersion_file
-        ps_opts._amp_mean_file = scratch_dir / ps_opts._amp_mean_file
-        ps_opts._output_file = scratch_dir / ps_opts._output_file
+        ps_opts._amp_dispersion_file = work_dir / ps_opts._amp_dispersion_file
+        ps_opts._amp_mean_file = work_dir / ps_opts._amp_mean_file
+        ps_opts._output_file = work_dir / ps_opts._output_file
 
     def create_dir_tree(self, debug=False):
         """Create the directory tree for the workflow."""
diff --git a/src/dolphin/workflows/s1_disp.py b/src/dolphin/workflows/s1_disp.py
index a7e37942..0d11b559 100755
--- a/src/dolphin/workflows/s1_disp.py
+++ b/src/dolphin/workflows/s1_disp.py
@@ -5,16 +5,14 @@
 from concurrent.futures import ProcessPoolExecutor
 from pathlib import Path
 from pprint import pformat
-from typing import Optional
 
 from dolphin import __version__
 from dolphin._background import DummyProcessPoolExecutor
 from dolphin._log import get_log, log_runtime
 from dolphin.utils import get_max_memory_usage, set_num_threads
 
-from . import _product, stitch_and_unwrap, wrapped_phase
-from ._pge_runconfig import RunConfig
-from ._utils import group_by_burst
+from . import stitch_and_unwrap, wrapped_phase
+from ._utils import _create_burst_cfg, _remove_dir_if_empty, group_by_burst
 from .config import Workflow
 
 
@@ -22,7 +20,6 @@
 def run(
     cfg: Workflow,
     debug: bool = False,
-    pge_runconfig: Optional[RunConfig] = None,
 ):
     """Run the displacement workflow on a stack of SLCs.
 
@@ -33,9 +30,6 @@ def run(
         workflow.
     debug : bool, optional
         Enable debug logging, by default False.
-    pge_runconfig : RunConfig, optional
-        If provided, adds PGE-specific metadata to the output product.
-        Not used by the workflow itself, only for extra metadata.
     """
     # Set the logging level for all `dolphin.` modules
     logger = get_log(name="dolphin", debug=debug, filename=cfg.log_file)
@@ -139,62 +133,8 @@ def run(
         )
     )
 
-    # ######################################
-    # 3. Finalize the output as an HDF5 product
-    # ######################################
-    logger.info(f"Creating {len(unwrapped_paths)} outputs in {cfg.output_directory}")
-    for unw_p, cc_p, s_corr_p in zip(
-        unwrapped_paths,
-        conncomp_paths,
-        spatial_corr_paths,
-    ):
-        output_name = cfg.output_directory / unw_p.with_suffix(".nc").name
-        _product.create_output_product(
-            unw_filename=unw_p,
-            conncomp_filename=cc_p,
-            tcorr_filename=stitched_tcorr_file,
-            spatial_corr_filename=s_corr_p,
-            output_name=output_name,
-            corrections={},
-            pge_runconfig=pge_runconfig,
-        )
-
-    if cfg.save_compressed_slc:
-        logger.info(f"Saving {len(comp_slc_dict.items())} compressed SLCs")
-        output_dir = cfg.output_directory / "compressed_slcs"
-        output_dir.mkdir(exist_ok=True)
-        _product.create_compressed_products(
-            comp_slc_dict=comp_slc_dict,
-            output_dir=output_dir,
-        )
-
     # Print the maximum memory usage for each worker
     max_mem = get_max_memory_usage(units="GB")
     logger.info(f"Maximum memory usage: {max_mem:.2f} GB")
     logger.info(f"Config file dolphin version: {cfg._dolphin_version}")
     logger.info(f"Current running dolphin version: {__version__}")
-
-
-def _create_burst_cfg(
-    cfg: Workflow,
-    burst_id: str,
-    grouped_slc_files: dict[str, list[Path]],
-    grouped_amp_mean_files: dict[str, list[Path]],
-    grouped_amp_dispersion_files: dict[str, list[Path]],
-) -> Workflow:
-    cfg_temp_dict = cfg.model_dump(exclude={"cslc_file_list"})
-
-    # Just update the inputs and the scratch directory
-    top_level_scratch = cfg_temp_dict["scratch_directory"]
-    cfg_temp_dict.update({"scratch_directory": top_level_scratch / burst_id})
-    cfg_temp_dict["cslc_file_list"] = grouped_slc_files[burst_id]
-    cfg_temp_dict["amplitude_mean_files"] = grouped_amp_mean_files[burst_id]
-    cfg_temp_dict["amplitude_dispersion_files"] = grouped_amp_dispersion_files[burst_id]
-    return Workflow(**cfg_temp_dict)
-
-
-def _remove_dir_if_empty(d: Path) -> None:
-    try:
-        d.rmdir()
-    except OSError:
-        pass
diff --git a/src/dolphin/workflows/stitch_and_unwrap.py b/src/dolphin/workflows/stitch_and_unwrap.py
index c00c11e1..98b06ff3 100644
--- a/src/dolphin/workflows/stitch_and_unwrap.py
+++ b/src/dolphin/workflows/stitch_and_unwrap.py
@@ -4,10 +4,20 @@
 from pathlib import Path
 from typing import Sequence
 
-from dolphin import io, stitching, unwrap
+from dolphin import io, stitching
 from dolphin._log import get_log, log_runtime
 from dolphin.interferogram import estimate_correlation_from_phase
 
+try:
+    from dolphin import unwrap
+
+    UNWRAP_INSTALLED = False
+except ImportError as e:
+    logger = get_log(__name__)
+    logger.info("Unwrapping dependencies not installed: %s", e)
+    UNWRAP_INSTALLED = False
+
+
 from .config import UnwrapMethod, Workflow
 
 
@@ -132,24 +142,23 @@ def run(
     ifg_filenames = sorted(Path(stitched_ifg_dir).glob("*.int"))  # type: ignore
     if not ifg_filenames:
         raise FileNotFoundError(f"No interferograms found in {stitched_ifg_dir}")
-    unwrapped_paths, conncomp_paths = unwrap.run(
-        ifg_filenames=ifg_filenames,
-        cor_filenames=spatial_corr_paths,
-        output_path=cfg.unwrap_options._directory,
-        nlooks=nlooks,
-        mask_file=output_mask,
-        # mask_file: Optional[Filename] = None,
-        # TODO: max jobs based on the CPUs and the available RAM? use dask?
-        max_jobs=unwrap_jobs,
-        # overwrite: bool = False,
-        no_tile=True,
-        use_icu=use_icu,
-    )
-
-    # ####################
-    # 3. Phase Corrections
-    # ####################
-    # TODO: Determine format for the tropospheric/ionospheric phase correction
+    if UNWRAP_INSTALLED:
+        unwrapped_paths, conncomp_paths = unwrap.run(
+            ifg_filenames=ifg_filenames,
+            cor_filenames=spatial_corr_paths,
+            output_path=cfg.unwrap_options._directory,
+            nlooks=nlooks,
+            mask_file=output_mask,
+            # mask_file: Optional[Filename] = None,
+            # TODO: max jobs based on the CPUs and the available RAM? use dask?
+            max_jobs=unwrap_jobs,
+            # overwrite: bool = False,
+            no_tile=True,
+            use_icu=use_icu,
+        )
+    else:
+        logger.info("Unwrapping dependencies not installed, skipping.")
+        unwrapped_paths, conncomp_paths = [], []
 
     return unwrapped_paths, conncomp_paths, spatial_corr_paths, stitched_tcorr_file
 
diff --git a/src/dolphin/workflows/wrapped_phase.py b/src/dolphin/workflows/wrapped_phase.py
index 28979658..cdcbeaf4 100644
--- a/src/dolphin/workflows/wrapped_phase.py
+++ b/src/dolphin/workflows/wrapped_phase.py
@@ -35,7 +35,7 @@ def run(cfg: Workflow, debug: bool = False) -> tuple[list[Path], Path, Path, Pat
         In the case of sequential phase linking, this is the average tcorr file.
     """
     logger = get_log(debug=debug)
-    logger.info(f"Running wrapped phase estimation in {cfg.scratch_directory}")
+    logger.info(f"Running wrapped phase estimation in {cfg.work_directory}")
 
     input_file_list = cfg.cslc_file_list
     if not input_file_list:
@@ -48,12 +48,12 @@ def run(cfg: Workflow, debug: bool = False) -> tuple[list[Path], Path, Path, Pat
     vrt_stack = stack.VRTStack(
         input_file_list,
         subdataset=subdataset,
-        outfile=cfg.scratch_directory / "slc_stack.vrt",
+        outfile=cfg.work_directory / "slc_stack.vrt",
     )
 
     # Make the nodata mask from the polygons, if we're using OPERA CSLCs
     try:
-        nodata_mask_file = cfg.scratch_directory / "nodata_mask.tif"
+        nodata_mask_file = cfg.work_directory / "nodata_mask.tif"
         _utils.make_nodata_mask(
             vrt_stack.file_list, out_file=nodata_mask_file, buffer_pixels=200
         )
diff --git a/tests/test_workflows_config.py b/tests/test_workflows_config.py
index 83abfdde..e450e85b 100644
--- a/tests/test_workflows_config.py
+++ b/tests/test_workflows_config.py
@@ -280,22 +280,19 @@ def test_config_defaults(dir_with_1_slc):
     assert c.output_options == config.OutputOptions()
     assert c.worker_settings == config.WorkerSettings()
     assert c.input_options == config.InputOptions(subdataset="data")
-    assert c.output_directory == Path("output").resolve()
-    assert c.scratch_directory == Path("scratch").resolve()
+    assert c.work_directory == Path(".").resolve()
 
     # Check the defaults for the sub-configs, where the folders
-    # should have been moved to the scratch directory
-    assert c.ps_options._directory == Path("scratch/PS").resolve()
-    assert c.ps_options._amp_mean_file == Path("scratch/PS/amp_mean.tif").resolve()
+    # should have been moved to the working directory
+    assert c.ps_options._directory == Path("PS").resolve()
+    assert c.ps_options._amp_mean_file == Path("PS/amp_mean.tif").resolve()
 
-    p = Path("scratch/PS/amp_dispersion.tif")
+    p = Path("PS/amp_dispersion.tif")
     assert c.ps_options._amp_dispersion_file == p.resolve()
 
-    assert c.phase_linking._directory == Path("scratch/linked_phase").resolve()
+    assert c.phase_linking._directory == Path("linked_phase").resolve()
 
-    assert (
-        c.interferogram_network._directory == Path("scratch/interferograms").resolve()
-    )
+    assert c.interferogram_network._directory == Path("interferograms").resolve()
     assert c.interferogram_network.reference_idx == 0
     assert (
         c.interferogram_network.network_type
@@ -305,7 +302,7 @@ def test_config_defaults(dir_with_1_slc):
     assert c.interferogram_network.max_bandwidth is None
     assert c.interferogram_network.max_temporal_baseline is None
 
-    assert c.unwrap_options._directory == Path("scratch/unwrapped").resolve()
+    assert c.unwrap_options._directory == Path("unwrapped").resolve()
 
     now = datetime.utcnow()
     assert (now - c.creation_time_utc).seconds == 0
@@ -325,8 +322,8 @@ def test_config_create_dir_tree(tmpdir, slc_file_list_nc):
         assert c.phase_linking._directory.exists()
         assert c.unwrap_options._directory.exists()
 
-        # Check that the scratch directory is created
-        assert Path("scratch").exists()
+        # Check that the working directory is created
+        assert Path(".").exists()
 
         for d in c._directory_list:
             assert d.exists()
diff --git a/tests/test_workflows_pge_runconfig.py b/tests/test_workflows_pge_runconfig.py
deleted file mode 100644
index 9a6b62b7..00000000
--- a/tests/test_workflows_pge_runconfig.py
+++ /dev/null
@@ -1,92 +0,0 @@
-import sys
-import warnings
-
-import pytest
-
-from dolphin.workflows._pge_runconfig import (
-    AlgorithmParameters,
-    DynamicAncillaryFileGroup,
-    InputFileGroup,
-    PrimaryExecutable,
-    ProductPathGroup,
-    RunConfig,
-)
-
-
-def test_algorithm_parameters_schema():
-    with warnings.catch_warnings():
-        warnings.filterwarnings("ignore", category=UserWarning)
-        AlgorithmParameters.print_yaml_schema()
-
-
-def test_run_config_schema():
-    with warnings.catch_warnings():
-        warnings.filterwarnings("ignore", category=UserWarning)
-        RunConfig.print_yaml_schema()
-
-
-@pytest.fixture
-def input_file_group(slc_file_list_nc_with_sds):
-    return InputFileGroup(cslc_file_list=slc_file_list_nc_with_sds, frame_id=10)
-
-
-@pytest.fixture
-def algorithm_parameters_file(tmp_path):
-    f = tmp_path / "test.yaml"
-    AlgorithmParameters().to_yaml(f)
-    return f
-
-
-@pytest.fixture
-def dynamic_ancillary_file_group(algorithm_parameters_file):
-    return DynamicAncillaryFileGroup(
-        algorithm_parameters_file=algorithm_parameters_file
-    )
-
-
-@pytest.fixture
-def product_path_group(tmp_path):
-    product_path = tmp_path / "product_path"
-    product_path.mkdir()
-    return ProductPathGroup(product_path=product_path)
-
-
-@pytest.fixture
-def runconfig_minimum(
-    input_file_group,
-    dynamic_ancillary_file_group,
-    product_path_group,
-):
-    c = RunConfig(
-        input_file_group=input_file_group,
-        primary_executable=PrimaryExecutable(),
-        dynamic_ancillary_file_group=dynamic_ancillary_file_group,
-        product_path_group=product_path_group,
-    )
-    return c
-
-
-def test_runconfig_to_yaml(runconfig_minimum):
-    print(runconfig_minimum.to_yaml(sys.stdout))
-
-
-def test_runconfig_to_workflow(runconfig_minimum):
-    print(runconfig_minimum.to_workflow())
-
-
-def test_runconfig_from_workflow(tmp_path, runconfig_minimum):
-    w = runconfig_minimum.to_workflow()
-    frame_id = runconfig_minimum.input_file_group.frame_id
-    algo_file = tmp_path / "algo_params.yaml"
-    w2 = RunConfig.from_workflow(w, frame_id, algo_file).to_workflow()
-
-    # these will be slightly different
-    w2.creation_time_utc = w.creation_time_utc
-    assert w == w2
-
-
-def test_runconfig_yaml_rountrip(tmp_path, runconfig_minimum):
-    f = tmp_path / "test.yaml"
-    runconfig_minimum.to_yaml(f)
-    c = RunConfig.from_yaml(f)
-    assert c == runconfig_minimum
diff --git a/tests/test_workflows_product.py b/tests/test_workflows_product.py
deleted file mode 100644
index f42461c3..00000000
--- a/tests/test_workflows_product.py
+++ /dev/null
@@ -1,69 +0,0 @@
-import numpy as np
-import pytest
-
-from dolphin import io
-from dolphin.workflows._product import create_output_product
-
-# random place in hawaii
-GEOTRANSFORM = [204500.0, 5.0, 0.0, 2151300.0, 0.0, -10.0]
-SRS = "EPSG:32605"
-SHAPE = (256, 256)
-
-
-@pytest.fixture
-def unw_filename(tmp_path) -> str:
-    data = np.random.randn(*SHAPE).astype(np.float32)
-    filename = tmp_path / "unw.tif"
-    io.write_arr(
-        arr=data, output_name=filename, geotransform=GEOTRANSFORM, projection=SRS
-    )
-    return filename
-
-
-@pytest.fixture
-def conncomp_filename(tmp_path) -> str:
-    data = np.random.randn(*SHAPE).astype(np.uint32)
-    filename = tmp_path / "conncomp.tif"
-    io.write_arr(
-        arr=data, output_name=filename, geotransform=GEOTRANSFORM, projection=SRS
-    )
-    return filename
-
-
-@pytest.fixture
-def tcorr_filename(tmp_path) -> str:
-    data = np.random.randn(*SHAPE).astype(np.float32)
-    filename = tmp_path / "tcorr.tif"
-    io.write_arr(
-        arr=data, output_name=filename, geotransform=GEOTRANSFORM, projection=SRS
-    )
-    return filename
-
-
-@pytest.fixture
-def spatial_corr_filename(tmp_path) -> str:
-    data = np.random.randn(*SHAPE).astype(np.float32)
-    filename = tmp_path / "spatial_corr.tif"
-    io.write_arr(
-        arr=data, output_name=filename, geotransform=GEOTRANSFORM, projection=SRS
-    )
-    return filename
-
-
-def test_create_output_product(
-    tmp_path,
-    unw_filename,
-    conncomp_filename,
-    tcorr_filename,
-    spatial_corr_filename,
-):
-    output_name = tmp_path / "output_product.nc"
-
-    create_output_product(
-        unw_filename=unw_filename,
-        conncomp_filename=conncomp_filename,
-        tcorr_filename=tcorr_filename,
-        spatial_corr_filename=spatial_corr_filename,
-        output_name=output_name,
-        corrections={},
-    )