o paper: add JOSS-style Markdown draft and bib alongside HTML .doc

Author: Rajeev Jain

paper/paper.bib

@@ -0,0 +1,11 @@
+@misc{xarray,
+  title = {xarray: N-D labeled arrays and datasets in Python},
+  howpublished = {\url{https://xarray.dev/}},
+  year = {2025}
+}
+
+@misc{ugrid,
+  title = {UGRID Conventions},
+  howpublished = {\url{https://ugrid-conventions.github.io/ugrid-conventions/}},
+  year = {2025}
+}

paper/paper.md

@@ -0,0 +1,102 @@
+---
+title: "UXarray: Xarray extensions for unstructured grids"
+authors:
+  - name: Rajeev Jain
+    orcid: "0000-0000-0000-0000"
+    affiliation: 1
+  - name: Aaron Zedwick
+    affiliation: 2
+  - name: David Ahijevych
+    affiliation: 3
+  - name: Anissa Zacharias
+    affiliation: 3
+  - name: David Galicia
+    affiliation: 4
+  - name: Orhan Eroglu
+    affiliation: 4
+  - name: Paul Ullrich
+    affiliation: 5
+  - name: Philip Chmielowiec
+    affiliation: 4
+  - name: Hongyu Chen
+    affiliation: 5
+  - name: Kristen Thyng
+    affiliation: 6
+  - name: Mario A. Rodriguez
+    affiliation: 3
+  - name: Maxime Liquet
+    affiliation: 7
+  - name: Michaela Sizemore
+    affiliation: 4
+  - name: Katelyn FitzGerald
+    affiliation: 6
+affiliations:
+  - name: Affiliation TBD (Lead)
+    index: 1
+  - name: Affiliation TBD
+    index: 2
+  - name: NCAR/UCAR (TBD)
+    index: 3
+  - name: Affiliation TBD
+    index: 4
+  - name: University of California, Davis (TBD)
+    index: 5
+  - name: Affiliation TBD
+    index: 6
+  - name: Affiliation TBD
+    index: 7
+date: 2025-08-18
+bibliography: paper.bib
+---
+
+# Summary
+
+UXarray provides an Xarray-compatible interface for loading, analyzing, and visualizing data on unstructured grids. It standardizes access to common grid representations (UGRID, MPAS, Exodus, ICON, SCRIP, ESMF, FESOM, HEALPix), pairing arrays with a Grid object that encodes topology, geometry, and conventions. Grid-aware accessors expose familiar workflows for subset, remap, cross-section extraction, and zonal statistics.
+
+![Architecture overview](images/fig1.svg)
+
+# Statement of need
+
+Modern climate and weather models increasingly use unstructured meshes. While Xarray offers labeled arrays and rich I/O, it does not natively understand unstructured grid topology. UXarray bridges this gap by enforcing a minimal UGRID-consistent contract and adding accessors that operate with mesh-aware semantics, enabling reproducible and scalable analysis pipelines.
+
+# State of the field
+
+We compare UXarray to adjacent tools and libraries (e.g., generic mesh readers, plotting stacks, and interpolation toolkits) and highlight its focus on tight Xarray integration, standardized I/O across formats, and accessor-based APIs that minimize glue code.
+
+# Implementation
+
+- Core abstractions: Grid and UxDataArray, with UncachedAccessor-based extensions.
+- I/O across multiple formats with normalization to a common topology representation.
+- Cross-sections via geodesic and constant-lat/lon sampling with nearest-neighbor face selection.
+- Remapping, subsetting, and plotting integrations.
+
+```python
+import uxarray as ux
+uxgrid = ux.open_grid("grid.ug")
+uxds = ux.open_dataset("grid.ug", "psi.nc")
+
+# Cross-section along a geodesic
+cs = uxds["psi"].cross_section(start=(-45, -45), end=(45, 45), steps=200)
+```
+
+![Cross-section sampling](images/fig2.svg)
+
+# Use cases
+
+- Cross-section analysis along great-circle arcs and along constant parallels/meridians.
+- Zonal statistics for climate diagnostics.
+- Format-agnostic ingestion and analysis across UGRID/MPAS/Exodus/ICON/SCRIP/ESMF/FESOM/HEALPix.
+- Integration with visualization stacks (Cartopy, HoloViz) for interactive exploration.
+- Project Pythia examples of UXarray usage (to be summarized and cited here).
+
+# Quality control
+
+Continuous integration, unit tests, style and linting (pre-commit, ruff), and multi-platform testing. Target Python 3.10+.
+
+# Acknowledgements
+
+Funding and institutional support to be added. We thank all contributors and users for feedback and testing.
+
+# References
+
+References will be added via paper.bib in subsequent commits (UGRID, Xarray, model format specs, Project Pythia resources).