Remote-sensing researcher: Sentinel-2 fire-detection benchmarks and geospatial analysis of vineyard wildfire resilience
2
papers here
2
reviewed by Evolve
1
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-09-20) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.
Research profile
Output—
No publication-volume data underlies this axis yet. It would measure total research output — papers per year across the full bibliographic record; with only two cataloged papers, the dimension is honestly left unscored.
Impact—
Citation counts, h-index, and venue influence are not available in the source data. This axis would measure scholarly impact through citations and uptake of the two reviewed papers; it remains unscored rather than estimated.
Focus0/100
The focus score of 0 is a direct readout of the raw field distribution: activity splits exactly 50/50 between remote sensing / geospatial analysis of agricultural wildfire resilience and remote sensing / wildfire detection — one cataloged paper in each. The two strands are closely related wildfire applications, but the even split across the two labeled fields yields a minimal concentration score rather than deep single-field specialization.
Collaboration32/100
The collaboration score of 32 reflects a compact network: 4 distinct coauthors across the 2 cataloged papers — Mohammadreza Narimani, Mahla Ardebili Pour, Mohammad Bagher Ghiasi, and Shreyan Mitra. Farajpoor takes the first-author position on the Kincade Fire vineyard study and the last-author position on the Sentinel-2 benchmark, with neither paper listing them as corresponding author, indicating contribution within small teams rather than independent or large-consortium work.
Activity—
Recent-publication activity data is missing. This axis would measure publishing cadence and currency, such as outputs per recent year; with both cataloged papers dated 2026 and no fuller record linked, the axis is left unscored.
Reproducibility100/100
Reproducibility is the standout dimension: a score of 100 (reproMean of 100 across reproPapers = 2). Both cataloged papers carry high reproducibility readiness — the Sentinel-2 benchmark is described as well-documented and openly licensed, and the Kincade Fire vineyard audit as openly reproducible, with its negative finding backed by boundary, donut, placebo, and residual-spatial diagnostics. Open data and transparent, falsifiable design are consistent hallmarks.
Research summary
Parastoo Farajpoor is a remote-sensing researcher whose cataloged work splits evenly between two complementary strands of wildfire science: deep-learning active-fire detection from Sentinel-2 imagery and geospatial analysis of agricultural wildfire resilience. The Evolve catalog holds two 2026 arXiv papers. On the first, a Sentinel-2 benchmark of 2,148 chips spanning 25 California wildfires for deep-learning active-fire segmentation, Farajpoor is last author; the dataset is openly licensed and well documented, though its algorithmic labels mean scores measure rule reproduction rather than fire-detection truth. On the second, a multisource geospatial study of the 2019 Kincade Fire across 4,581 Sonoma vineyard fields, Farajpoor is first author; lower dNBR over vineyards is real descriptively, but dissolves under boundary, donut, placebo, and residual-spatial diagnostics, so no firebreak effect is claimed. Both papers rate high for reproducibility, giving a reproducibility score of 100 (reproMean across 2 papers), while the even 50/50 field split yields a focus score of 0. Collaboration rests on 4 distinct coauthors, including Mohammadreza Narimani, Mahla Ardebili Pour, Mohammad Bagher Ghiasi, and Shreyan Mitra. Output, impact, and activity metrics are not yet available.
Collaboration pattern
Collaboration is compact: 4 distinct coauthors span the two cataloged papers — Mohammadreza Narimani, Mahla Ardebili Pour, Mohammad Bagher Ghiasi, and Shreyan Mitra. Farajpoor takes the first-author role on the Kincade Fire vineyard study and the last-author role on the Sentinel-2 benchmark, neither as corresponding author, consistent with contribution within small author teams (collaboration score 32).
Research keywords
Highlights
Open Sentinel-2 active-fire segmentation benchmark across 25 California wildfires
Disciplined geospatial audit of vineyard resilience in the 2019 Kincade Fire
A two-strand, fully reproducible early portfolio
A compact 2026 arXiv portfolio spanning wildfire detection benchmarks and agricultural resilience analysis, with both papers rated high for reproducibility readiness (reproMean 100) — open data, transparent methods, and falsifiable claims throughout.
Timeline
- 2026First-authored the multisource remote sensing and geospatial analysis of vineyard impacts and resilience in the 2019 Kincade Fire, covering 4,581 Sonoma vineyard fields.
- 2026Contributed as last author to the Sentinel-2 benchmark dataset for deep-learning active-fire segmentation spanning 25 California wildfires (2,148 chips).
- 2026Both cataloged papers were posted to arXiv and rated high for reproducibility readiness, establishing an open-data track record from the outset.
Papers in the Evolve catalog
Frequent co-authors
About this profile
This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.
- No institution, country, or ORCID is provided in the source data; affiliation-level context is absent.
- Output, impact, and activity dimensions have no underlying bibliometric data and are left unscored.
- The profile rests on only two Evolve-reviewed arXiv papers, both dated 2026; earlier or other-venue work is not captured.
- Radar scores are 0-100 values computed from real bibliometric data over this small corpus and may shift as more records are linked.