I am an Earth scientist currently pursuing my PhD. in the Remote Sensing Ecohydrology Group at Stanford University. I develop technologies to measure forest health using remote sensing and machine learning. Read more at https://krishnakrao.github.io/
Honors & Awards
NASA Earth and Space Science Fellow, NASA (2018)
J.N. Tata Fellowship, J. N. Tata endowment (2016)
K.C. Mahindra Fellowship, K.C. Mahindra Educational Trust (2016)
Institute Blues, Indian Institute of Technology Madras (2014)
Larsen & Toubro ECC Endowment Fellow, Indian Institute of Technology Madras (2014)
Professional Affiliations and Activities
Chair, Talk of the Farm Speaking Club (2020 - Present)
President, Hindu Students Association (2018 - 2019)
Vice-president of Education, Stanford Toastmasters (2017 - 2018)
Education & Certifications
Master of Science, Stanford University, Civil and Environmental Engineering (2018)
Bachelor of Technology, Indian Institute of Technology Madras, Civil Engineering (2014)
In my free time I ride my bicycle to remote areas and write about my experiences here https://medium.com/@kkrao
Current Research and Scholarly Interests
My research focuses on developing machine learning algorithms to remotely monitor vegetation health by using microwave remote sensing. I am interested in answering questions such as:
- How can we leverage the vast amounts of data our satellites are capturing to understand the health of our forests?
- How can we obtain information about hydraulic health of vegetation with little or no ground-collected data
Effects of heterogeneity on group behaviour, Ouellette Lab, Stanford University (October 10, 2016 - March 31, 2017)
In nature, group behaviour exists widely in the form of flocks, swarms, herds and so on. This groups rarely consist of purely homogenous members. As part of this project, we studied the effects of physical and relationship based heterogeneity among the members of a group. We also numerically modelled the growth disorder in a flock due to individual asymmetries.
Remotely Sensed Vegetation Optical Depth as an Indicator of Drought-driven Tree Mortality, Stanford University (April 1, 2017 - July 30, 2018)
A nonparametric approach for estimating high-resolution fuel moisture content across the Western United States using synthetic aperture radar, Stanford University (July 31, 2018 - Present)
Fuel moisture content (FMC) is a key determinant of wildfire ignition and propagation- and thus fire risk. However, FMC has high spatial heterogeneity, making it cumbersome and expensive to gather field measurements at sufficient density. Alternatively, microwave backscatter measurements from synthetic aperture radar (SAR) can be used. These measurements are sensitive to canopy moisture and thus contain information about fuel moisture. Previous efforts to estimate FMC from SAR use highly detailed radiative transfer models that rely on a number of geometric properties such as stand height, crown width, leaf density, orientation and so on. Such properties vary widely across species and forests. As a result, existing models are impossible to apply at large scales. In this project, I am developing a nonparametric model powered by machine learning - an artificial neural network (ANN) - to estimate FMC from SAR backscatter.
Wireline Field Engineer, Schlumberger (11/16/2014 - 6/1/2016)
Krishna worked in Schlumberger as a Wireline Field Engineer. After obtaining his theoretical training in geology and petrophysics and practical exposure to field scenarios, Krishna worked in North-East India for two of the biggest clients there. On the field, Krishna designed, executed, and delivered several service projects related to reservoir characterization and production.
Abu Dhabi, UAE
- SAR-enhanced mapping of live fuel moisture content REMOTE SENSING OF ENVIRONMENT 2020; 245
- Satellite-based vegetation optical depth as an indicator of drought-driven tree mortality REMOTE SENSING OF ENVIRONMENT 2019; 227: 125–36
Macro to micro: microwave remote sensing of plant water content for physiology and ecology
2019; 223: 1166-1172
Although primarily valued for their suitability for oceanographic applications and soil moisture estimation, microwave remote sensing observations are also sensitive to plant water content (Mw ). Since Mw depends on both plant water status and biomass, these observations have the potential to be useful for a range of plant drought response studies. In this paper, we introduce the principles behind microwave remote sensing observations to illustrate how they are sensitive to plant water content and discuss the relationship between landscape-scale Mw and common stand-scale metrics, including plant-scale relative water content, live fuel moisture content and leaf water potential. Lastly, we discuss how various sensor types can be leveraged for specific applications depending on the spatio-temporal resolution needed.
View details for DOI 10.1111/nph.15808