Portfolio

Professional Projects

  • Data Science Intern - Geospatial Data Pipeline Engineer

    Soil Capital, Louvain-la-Neuve, Belgium

    Read More

  • Geospatial Engineering Intern

    HAEDES, Ghent, Belgium

    Read More

  • Water Resources Consultant

    Read More

  • Climate Change Consultant with a focus on Hydrology

    Read More

  • Junior Water Consultant Engineer

    Read More

  • Undergraduate Research Assistant Student

    Laboratory of Hydraulics of the University Mayor de San Simón

    Read More

Data Science Intern - Geospatial Data Pipeline Engineer

Soil Capital, Louvain-la-Neuve, Belgium


Highlights


- Built a layered (Bronze-Silver-Analytics) ingestion pipeline in Microsoft Fabric for ERA5-Land reanalysis and Copernicus EDO drought data across France, Belgium, the UK and Poland, making the data centrally accessible and reusable for other teams.
- Learned Microsoft Fabric, Spark SQL and PySpark on the job and matched gridded reanalysis data to farm field coordinates via nearest-grid-cell assignment, at scale across 270,000+ field-seasons in France alone.
- Derived agrometeorological variables (vapor pressure, 2m wind speed, ET0) and agroclimatic indicators (GDD, heat/frost days, precipitation extremes) for drought risk assessment.
- Validated reanalysis-derived variables against ground-based weather station observations.

Technical skills


- Languages: Python, SQL.


- Tools: Microsoft Fabric, PySpark, Spark SQL.


- Data Architecture: Medallion Pipeline Architecture (Bronze/Silver/Analytics).

Geospatial Engineering Intern

HAEDES, Ghent, Belgium


Highlights


- Built a pipeline to acquire and gap-fill sub-hourly meteorological data (temperature, radiation, reference ET0) from the Waterinfo Vlaanderen API (KiWIS), gap-filling temperature using bias-corrected neighboring-station data and time interpolation.
- Implemented and validated an hourly evapotranspiration (ET0) function (FAO-56 Hargreaves) to feed an hourly-timestep hydrological model.
- Built reusable Python clients automating raster zonal statistics via GeoServer's WPS/WCS service, validating output consistency against QGIS and profiling processing time across dataset sizes.
- Built an automated QA pipeline to validate and curate a geospatial platform's basemap/overlay catalog, probing endpoint availability and removing broken layers.
- Built an approach to convert continuous water volume time series into discrete sluice gate operation counts, using a fixed per-event volume and hourly scheduling.

Technical skills


- Languages: Python.


- Tools: GeoServer (WMS/WCS/WPS), QGIS.


- Version Control Systems: Git and GitHub.

Water Resources Consultant

Independent


Highlights


- Collection of hydrometeorological data from observed stations and remote sensors( GMET , CHIRPS , GPM (IMERG7) , TerraClimate , etc ).
- Statistical analysis, data cleaning and data filling of hydrometeorological data.
- Elaboration of graphs and maps of seasonal variation of precipitation (Isohyets) and temperature (Isotherms).
- Water balance using hydrological models.
- Simulation Operation Reservoir.
- Analysis of hydroclimatic variation and under climate change scenarios.

Skills


- Hydrology Modeling: Proficiency in HydroBID. Training in WEAP, SWAT and HEC-RAS.


- GIS software: ArcGIS, QGIS, ang Google Earth Engine.


- Data Wrangling and Statistical Data Analysis: Python and R.


- Data Visualization: Matplotlib, Seaborn, and Plotly


- Version Control Systems: Git and GitHub.


- Development Tools: Visual Studio, GoogleColab, JupyterNotebook



dagster_deployment
Boxplot monthly average precipitation for San Ignacio de Velasco station
Made using Matplotlib.
dagster_run
Boxplot of daily precipitation for San Ignacio de Velasco station.
Made using Matplotlib.
az-rain-prediction
Relationship between average temperature (°C) and altitude.
Made using Excel
az-sd
Double cumulative mass curve to identify inconsistencies in data filling.
Made using hyfo (R library) and Matplotlib.
az-eda-2
Average annual cumulative precipitation Isolines.
Made using QGIS.
az-eda-1
Proportion of attention demand of, for indicated dam heights and Probabilities of Failure (PF).
Made using Python.

Climate Change Consultant with a focus on Hydrology

Independent


Highlights


- Assess the impact of climate change on precipitation, average temperature, maximum temperature and minimum temperature.
- Apply statistical downscaling methods: Quantile Delta Mapping (QDM) ( Cannon, A. 2015 ) and Delta Change ( Teutschbein, C., & Seibert, J. 2012 ).
- Generate ensambles of climate models using climate change scenarios of Representative Concentration Pathways (RCP) and Shared Socioeconomic Pathways (SSP) for future periods of interest.
- Impact of climate change on precipitation and temperature for the analysis of water potential and extreme events.

To achieve the goals, I updated the tool:
- dClimate an application to automate the process of statistical downscaling regional climate models CORDEX with QDM and Delta Change.
- SD_DC_NEX-GDDP and SD_QDM_NEX-GDDP application to automate the process of statistical downscaling of NEX-GDDP-CMIP6 models.

Technical skills


- Languages: Python and R.


- Python Libraries: NumPy, Pandas, Xarray, NetCDF, ee (API Google Earth Engine), Matplotlib, Plotly, Tkinter


- Version Control Systems: Git and GitHub.


- Development Tools: Visual Studio Code, Google Colab, Spyder and Rstudio



dagster_deployment
Statistical graphical analysis of precipitation change under climate change scenarios.
Made using Python.
dagster_run
Statistical graphical analysis of minimum temperature change under climate change scenarios.
Made using Python.
az-rain-prediction
Statistical graphical analysis of monthly precipitation change under climate change scenarios.
Made using Python.
az-sd
Probability of occurrence of maximum daily precipitation under climate change scenarios.
Made using Matplotlib.
az-eda-2
Probability of occurrence of maximum daily precipitation under climate change scenarios.
Made using Matplotlib.
az-eda-1
Precipitation-Duration-Frequency curves under climate change effect.
Made using Python.
az-eda-1
Location of Angostura Reservoir (Santa Cruz, Bolivia).
Made using QGIS.
az-eda-1
Reservoir height vs. demand service ratio period 2080s.
Made using Python, and Matplotlib.

Junior Water Resources Consultant

Hydraulics Laboratory of Universidad Mayor de San Simón


Highlights


- Compilation and analysis of the temporal and spatial availability of water level and flow series.
- Analysis and visualization of hydrometric data.
- Verify and/or updating of rating curves. Adjustment of rating curves using the potential equation, logarithmic transformation, and the match
point.
- Analysis, critique, and treatment of water level data using the Modi coefficient..
- Fill of hydrometric information, generating level series using water level data from different sources, and also using transfer functions that relate precipitation-level, precipitation-flow, and flow-flow.
- Preparation of reports and files.

Skills


- GIS software: ArcGIS, and QGIS.


- Data Wrangling and Statistical Data Analysis: Python and Excel.


- Data Visualization: Matplotlib.


- Development Tools: Visual Studio, GoogleColab, JupyterNotebook



dagster_deployment
Discharge curves and confidence intervals
Made using Excel SpreadSheet and Python.
dagster_run
Discharge curves
Made using Excel SpreadSheet and Python.
az-eda-2
Corrected level series and flow rate at Boopi station.
Made using Excel SpreadSheet and Python.
az-eda-1
Corrected level series and flow rate at El Sena station.
Made using Excel SpreadSheet and Python.
az-eda-1
Spatial correlation Riberalta station.
Made using Excel SpreadSheet and Python.
az-eda-1
Cumulative consistency of floods at Riberalta station.
Made using Excel SpreadSheet and Python.

Undergraduate Research Assistant

Hydraulics Laboratory of Universidad Mayor de San Simón


Highlights


• Collaboration in the monitoring of inter-Andean basins
- Static and dynamic calibration of rain gauges.
- Installation of new measurement equipment, rain gauges, and pressure sensors, as well as its maintenance.
- Data download for precipitation, temperature, and pressure sensor data for water level recording.
- Cleaning and post-processing of recorded precipitation and temperature data, implementing Python.
• Collaboration in service provision projects.
- Topographic survey.
- Generate Maps.
- Post-processing of hydrometeorological data.
• Collaboration in research dissemination events: "LHUMSS Research Dissemination Webinars"
- Creation and design of visual content or brochures to promote
- Development of registration forms, questions, and attendance tracking.

Skills


- GIS software: ArcGIS, QGIS, ang Google Earth Engine.


- Data Wrangling and Statistical Data Analysis: Python and R.


- Data Visualization: Matplotlib, Seaborn, and Plotly


- Development Tools: GoogleColab, and JupyterNotebook



dagster_deployment
Download data from rain gauge precipitation.
Location Titiri, Cochabamba.
dagster_run
Topobatimetry of reservoir 2 in Country Club.
Location Cercado, Cochabamba.
az-eda-2
Contour lines of reservoir 2 in Country Club.
Made using ArcGIS.
az-eda-1
LHUMSS Research Dissemination Webinars.
Made using Canvas.



Contact


Feel free to contact me for any question. For open source projects, please open an issue or pull request on Github . If you want to follow my work, reach me on Linkedin. Otherwise, send me an email at caldunate.lp@gmail.com .
web seo icon