Nabid Sheikh

Environmental engineering student, CZU Prague

I dig into how systems behave. Right now that's 24 years of Czech mountain snow cover data pulled from satellites, and the infrastructure running this site.

Currently: writing my bachelor thesis on Czech mountain snow cover trends, running the cloud infrastructure (Oracle, Azure) behind it and this site, and picking up field methods in snow hydrology along the way.

About

I'm an environmental engineering student at the Czech University of Life Sciences Prague (CZU), currently in my second year. I moved from Dhaka, Bangladesh to Prague in 2024 for the programme, which has covered environmental chemistry, hydrology, GIS, air pollution, ecotoxicology, soil science, and landscape ecology.

My focus sits at the intersection of remote sensing, hydrology, and GIS: using satellite data to answer questions about how mountain environments are changing. I'm drawn to research with a clear environmental application, not research for its own sake.

I speak Bangla natively, work comfortably in English, and am building up my Czech.

Bachelor thesis

Long-term trends in snow cover duration in Czech mountain ranges and their implications for water resources
Czech University of Life Sciences Prague, Faculty of Environmental Sciences
Supervised by doc. Ing. Jan Komárek, Ph.D.
Submission due March 2027

The thesis tracks snow cover duration across three Czech mountain ranges (Krkonoše, Šumava, and Hrubý Jeseník), using 24 years of MODIS satellite data (2000–2024), split into three elevation bands to test whether temperature or precipitation drives snowpack change at different heights.

Where the decline is showing up, by elevation
Above 1,200 m
No significant trend detected
800–1,200 m
Significant decline (p < 0.05)
Below 800 m
Significant decline (p < 0.05)
Snow cover duration is declining below 1,200 m across all three ranges, statistically significant in 11 of 18 elevation–forest categories tested.
Above 1,200 m, no significant trend has turned up yet.
Temperature tracks the decline more closely than precipitation at every elevation tested, including above 1,200 m.
Cloud cover itself shows no trend over the same period, which rules out satellite cloud obscuration as the cause of the declining signal.
In progress. Built with Google Earth Engine, Python (pandas, pyMannKendall), and ArcGIS Pro.

Field work

Mar 2025
Polar Winter School — Agricultural University of Iceland, Borgarnes
Snow microstructure, impurity deposition in snow, UAV-based snow cover mapping, arctic snow hydrology.
Jan 2026
Snow Hydrology Field Course — Krkonoše Mountains, Czechia
Snow pit stratigraphy, snow water equivalent measurement, avalanche risk assessment and safety certification.

Also on this server

This site's own host, an Oracle Cloud VM, reports its live vitals on a status page I built and deploy automatically through GitHub Actions on every push.

How this site runs
01git push to main
↓
02GitHub Actions checks out the code
↓
03Deploys over SSH to an Oracle Cloud ARM VM, installs dependencies in a fresh virtualenv
↓
04systemd restarts the app; nginx reverse-proxies it and terminates TLS
↓
05Let’s Encrypt certificate, auto-renewed by certbot
My own SSH and RDP access to the VM goes over Tailscale. GitHub's runners deploy straight over the public IP.

Technical skills

Remote sensing
Google Earth Engine, MODIS (MOD10A1 / MOD10A2), Sentinel-2, UAV data collection
GIS
ArcGIS Pro, QGIS, DEM-based elevation analysis
Programming
Python (pandas, matplotlib), R, JavaScript (Earth Engine scripting)
Statistics
Mann-Kendall trend test, Sen's slope estimator, correlation analysis
Field methods
Snow pit excavation, snow water equivalent measurement, avalanche safety certified
Infrastructure
Linux administration, cloud VMs (Oracle, Azure), CI/CD (GitHub Actions), SSH/Tailscale networking

Get in touch

Open to research assistant roles, field campaigns, thesis-related collaboration, and infrastructure or DevOps work.