A Hybrid Bias Correction Framework for Daily Satellite Precipitation

Master's thesis (Bogor Agricultural University, 2026) on the hybrid LSEQM+DL bias correction framework for daily Integrated Multi-satellitE Retrievals for GPM Late Run (IMERG-L) precipitation over Indonesia, with gauge validation against 172 independent BMKG stations.

Status: final draft submitted to the IPB Applied Climatology programme. See the parent code repository for the implementation and the data archive.

Author

Benny Istanto (G2501222008)
Applied Climatology Study Program
Faculty of Mathematics and Natural Sciences
Bogor Agricultural University
Bogor, Indonesia

Supervised by Prof. Dr. Ir. Rizaldi Boer, M.S. and Dr. I Putu Santikayasa, S.Si., M.Sc.

What this repository contains

This repository holds the LaTeX sources, figures, and figure-generator scripts for the thesis. The corresponding code, data pipeline, and validation framework live in the parent repository:

• Parent code repository: https://github.com/bennyistanto/hybrid-bias-correction
• Data archive: deposited on Zenodo (DOI in Appendix B of the thesis)
• Companion paper (peer-reviewed): Istanto, B.; Boer, R.; Santikayasa, I.P. A Modular and Transferable Framework for Enhancing Satellite-Derived Daily Precipitation: Adjusting Values, Aligning Distributions, and Preserving Extremes. Remote Sens. 2026, 18, 2298. https://doi.org/10.3390/rs18142298

.
├── main_thesis.tex                  # entry point
├── frontmatter.tex                  # cover, summary, signatures, ToC, LoT, LoF, LoA
├── chapter_01_introduction.tex      # chapter 1
├── chapter_02_literature_review.tex # chapter 2
├── chapter_03_methods.tex           # chapter 3
├── chapter_04_results.tex           # chapter 4
├── chapter_05_discussion.tex        # chapter 5
├── chapter_06_conclusion.tex        # chapter 6
├── chapter_99_appendices.tex        # appendices A through H
├── biography.tex                    # author biography page
├── references.bib                   # bibliography (BibTeX)
├── ipb.bst                          # IPB PPKI bibliography style
├── build.sh                         # local build script (pdflatex x bibtex x pdflatex x pdflatex)
├── figures/                         # PNGs included by the .tex sources
└── scripts/                         # one Python script per figure (see scripts/README.md)

How to build the PDF

Prerequisites

• TeX Live 2023 or newer (with pdflatex, bibtex, xspace, needspace, pdflscape, tocloft, titlesec, natbib, geometry, caption, booktabs, enumitem, hyperref)
• GNU sed (for the et al. italic post-processing in build.sh)

Build

./build.sh

This runs the canonical three-pass cycle:

1. pdflatex (skeleton, generates .aux/.toc)
2. bibtex (resolves citations) + post-processing sed for italic et al.
3. pdflatex (populates ToC / LoF / LoT)
4. pdflatex (stabilises cross-references)

The output is main_thesis.pdf (~ 132 pages, ~ 12 MB).

Overleaf

This repository is Overleaf-compatible: upload the folder, set main_thesis.tex as the main document, and use the pdfLaTeX compiler. The et al. italic post-processing step does not run on Overleaf, so a small number of et al. strings in the bibliography will render upright instead of italic; the rest of the formatting is identical to the local build.

How to rebuild the figures

The figure generators live in scripts/. See scripts/README.md for the one-script-per-figure inventory and dependencies. Each script is runnable in isolation; collectively they require a scientific Python stack (numpy / pandas / xarray / matplotlib / geopandas) plus access to the data archive cited in Appendix B.

Citation

If you reference this thesis, please cite it as recorded in CITATION.cff. After publication on Zenodo a DOI will be added.

License

The thesis text and figures are released under Creative Commons Attribution 4.0 (CC-BY-4.0). The build scripts and figure-generator Python code are released under the MIT License.

See LICENSE for the full text of both.

Acknowledgments

Built on the IPB PPKI thesis format, customised for English-language submission. And the modified IPB CSL for English version.

Supervisors and data providers are credited in the front matter of the thesis.