Arif in a paddock, wearing a straw hat and working on a laptop connected to a solar-powered field station

Environmental informatics

Dr Mohammad Arifur Rahman (Arif)

Remote sensing, spatial data science and scientific software

Analyst Programmer, Wimmera Catchment Management Authority

Horsham, Victoria, Australia / +61 406 155 493

arifcs@gmail.com / Google Scholar / GitHub / LinkedIn

About

I came to Wimmera CMA in November 2022 as a PhD student, on a scholarship it funded jointly with the CRC for High Performance Soils, to estimate soil organic carbon from hyperspectral data. In June 2025 I stayed on full time, and the questions widened from soil to the whole catchment: where water sits in the landscape, how vegetation is changing, and what condition the land is in. Satellite imagery, LiDAR and aerial photography supply the evidence, and more than sixteen years of commercial software development turn it into tools people use.

Current work

Wetlands. Water and vegetation histories for about 30,000 mapped wetlands, from Sentinel-2 and long-term aerial imagery.

Vegetation. Mapping native against commercial vegetation, and following paddock-scale vegetation through time.

Water and land. Waterways, flood behaviour and environmental water, worked through with ecologists and water specialists.

Career

Environmental informaticsSince June 2025

Analyst Programmer (GIS & Remote Sensing), Wimmera CMA, Horsham, Victoria.

Software development2008 to 2025

Machine Learning Engineer at WebAlive, Melbourne. Earlier, Engineering Manager at Bit Mascot, Bangladesh.

Education

PhD (Engineering)December 2025

Federation University Australia. Data-efficient hyperspectral band selection for soil organic carbon estimation. Examiner: “an outstanding, and highly complete piece of doctoral research.”

BSc, Computer ScienceJanuary 2007

American International University-Bangladesh. Summa cum laude, GPA 3.97 of 4.

Publications

Deep Learning-based Adaptive Downsampling of Hyperspectral Bands for Soil Organic Carbon Estimation. IEEE Access, 2025.

BSDR: A Data-Efficient Deep Learning-Based Hyperspectral Band Selection Algorithm Using Discrete Relaxation. Sensors, 2024.

Addressing Limitations of Common Methods in Attention-based Hyperspectral Band Selection Algorithms. DICTA (IEEE), 2024.

Fieldwork

CASA-accredited drone (RPA) operator. Took part in Monash University’s airborne P- and L-band microwave remote-sensing campaign.