I am leading the Science Team at Terradot developing a team and a MRV platform to measure & simulate C sequestration & yield compliant with global standards (e.g. Cascade, Isometric, VERRA). At Terradot, I worked in close collaboration with high-level Stanford & world-renowned Reactive Transport Modelers focused on CDR projects. I also hold an academic positions as Adjunct Senior Research Fellow at the Centre for Sustainable Agricultural Systems (University of Southern Queensland) in Australia. Currently, I built a team from scratch to >10 people in just 6 months, fostering a robust & efficient workforce. I lead my team to innovate, iterate, and excel, always keeping the end-goal in sight: delivering impactful solutions swiftly to stay ahead in the ever-evolving digital landscape. My career is anchored in a relentless pursuit of excellence, where delivering V0 products swiftly isn’t just a goal, it’s my standard.
I apply data analytics, statistics and programming to understand soil-crop-climate interactions behind coupled models. My interests are in quantifying the effects of crop management & climate/soil variability on CO2 removal and yield through mechanistic models ( APSIM & MIN3P) at different spatio-temporal scales. I have a passionate interest in disentangling sources of uncertainties/variability to better predict yield and C (organic and inorganic) sequestration trough ERW field deployments. This requires a broad understanding of how the environment influences crop growth; how rainfall/irrigation, & applied rock influence soil status; how crops obtain water/nutrients from the soil; how soil processes contribute to the loss of C, N; and how all these processes interact.
Over 15 years of experience managing large soil-crop-climate datasets (8+ years using Python and geo-statistics), developing models (APSIM, 12+ years & MIN3P) & applying data pipelining and optimization towards solving large-scale ERW and C deployments. I collected (experiments) & analyzed data including >18 crop species (annual, perennials). I have 50+ published articles in high-impact journals (13 as first author). I received 18 research grants (70% as CI) with a cash value of ~USD5.5M.
I mentored graduate/ 5 PhD students who are currently leading global research in academia/industry. As part of Regrow Ag, I led a multicultural global team (15 people including data scientists, software developers and engineers) under the Niche project (USD4M) funded by B&MGF which focuses on the analysis of GxExM in Sub-Saharan Africa.
During my career, I work with multi-cultural & multi-disciplinary teams (hybrid approach - Academia & Industry) & have ground experience working in 9 countries (Argentina, Brazil, Australia, NZ, USA, The Netherlands, Germany, Rwanda & Kenya).
PhD in Agricultural Sciences, 2017
National University of Mar del Plata, Argentina
Agricultural Engineer, 2011
National University of Entre Ríos, Argentina
National University of Mar del Plata, Argentina (duration = 5 years; 422 credit hours)
GPA: 8.8 (out of 10) No. failed: 0 (none)
Dissertation: Precipitation use efficiency in annual forage crop sequences and perennial pastures (in Spanish)
The growing demand for beef and dairy products requires technological options to improve the productivity and resource use efficiency of forages crops with less environmental impact. Livestock production systems based on forage crop sequences (FCS) could be more productive and efficient than those based on perennial pastures (PP). However, there are many questions about the FCS implementation related to the system stability in the long term and about the root-derived soil organic carbon in these systems. The main objective of this thesis was to provide original knowledge about the main ecophysiological aspects determining forage supply in livestock systems based on the use of FCS and PP.
The study was conducted in three steps (i) above-ground dry matter yield (AGDM) and precipitation use efficiency (PUE) were analyzed in Rafaela, Pergamino, General Villegas and Trenque Lauquen, (ii) in Balcarce were evaluated AGDM, below-ground dry matter yield (BGDM), PUE (i.e. water capture [WC] * water use efficiency [WUE]), radiation productivity (RP, i.e. radiation capture [RC] * radiation use efficiency [RUE]) and soil carbon (C) variations in different organic matter fractions. Finally, (iii) Agricultural Production Systems Simulator (APSIM) was calibrated and validated to analyze the accumulated annual precipitation, AGDM and PUE variability using a long-term climate database (30 years).
In general, the AGDM was higher for the FCS than the PP treatments, although more variable in the long term. Below-ground dry matter yield was similar for both treatments. Likewise, there was a greater association between the contribution of C and BGDM in sub-surface horizons below than 0,15 m soil depth. The PP treatments shown higher RC and similar WC than the FCS treatments. However, FCS shown higher RUE and WUE, which led to higher RP and PUE. In turn, the PP treatments shown lower inter-annual variability of PUE than FCS in the long term. The multi-environmental analysis on the impacts of different forage cropping systems on PUE, as well as on the soil C variations, provide key knowledge and information to develop management strategies to increase the sustainable productivity of livestock systems in the Argentinean Pampas.
The image shows the logical flow of the PhD thesis including the stages, chapters, scales, spatio-temporal levels of analysis, and the measured or estimated variables. DMa, Aerial Dry Matter Yield; WP, Water Productivity ; DMr, Root Dry Matter Yield; RP, Radiation Productivity; OM, Organic Matter
National University of Entre Ríos, Argentina (duration = 7 years; 3479 credit hours)
GPA: 8.8 (out of 10) No. failed: 0 (none) Historical GPA: 6.8 (out of 10)
Dissertation: Response to plant population density in different sunflower hybrids (in Spanish)
Applied Research
Research
Teaching and Learning
U.S. Department of Energy
Basic research and methods
Applied research and industry engagement
Coaching, supervision and leadership activities
Tasmanian Institute of Agriculture (AUD $336,423)
Research
Teaching
Research Council of Argentina, CONICET (AUD $35,568)
Research
Teaching
Research Council of Argentina, CONICET (AUD $69,240)
Consultancy
Teaching
Argentinian Ministry of Education (AUD $9,471)
Ability to get own funding as chief investigator or partner to do research 💸
Responsibility: Collaborator
Funding body: University of Sydney
Partners: NSW Department of Primary Industries (Dr Gargiulo), University of Sydney (Drs Garcia and Islam).
Responsibility: Chief investigator
Funding body: Bill & Melinda Gates Foundation (4 yr).
Partners: University of Lincoln (Dr Grassini), NASA Harvest (Dr Inbal Becker-Reshef), Gates Foundation (Dr Hausmann), One Acre Fund (Dr Aston).
Responsibility: Chief investigator
Funding body: Argentinian Agency for the Promotion of Research, Technological Development and Innovation PICT-2020-SERIEA-III-A RAICES
Partners: University of Entre Rios, Argentina ( Dr Caviglia); PIRSA-SARDI, Australia ( Dr Sadras)
Responsibility: Partner
Funding body: Argentinian Agency for the Promotion of Research, Technological Development and Innovation PICT-2019-I-D
Partners: Regional Analysis and Remote Sensing Lab, University of Buenos Aires, Argentina ( Dr Texeira & Dr Oesterheld)
Responsibility: Chief investigator
Funding body: JM Roberts Charitable Trust and the University of Tasmania
Partners: Simplot, McCain, Tasmanian Department of Primary Industries, Parks, Water and Environment (DPIPWE) & University of Sydney ( Mat Webb)
Responsibility: Partner
Funding body: Argentinian Agency for the Promotion of Research, Technological Development and Innovation PICT-I-A-2018
Partners: Regional Analysis and Remote Sensing Lab, University of Buenos Aires, Argentina ( Dr Irisarri & Dr Oesterheld)
Responsibility: Chief investigator
Funding body: College of Sciences and Engineering, University of Tasmania
Partners: Simplot, McCain and The Tasmanian Department of Primary Industries, Parks, Water and Environment (DPIPWE) & University of Sydney (Mat Webb)
Responsibility: Chief investigator
Funding body: Council on Australia Latin America Relations (COALAR) Australia’s Department of Foreign Affairs & Trade, Australian Government
Partners: University of Southern Queensland ( Assoc Prof Pembleton president of the APSIM Initiative), Regional Analysis and Remote Sensing Lab (University of Buenos Aires), National Institute of Agricultural Research (INIA; Uruguay), CREA Farmer Groups (Argentina-Uruguay)
Responsibility: Chief investigator
Funding body: CSIRO-Tasmanian Institute of Agriculture
Partners: CSIRO Global Food and Nutrition Security, Australia ( Dr Katharina Waha)
Responsibility: Chief investigator
Funding body: CSIRO-Tasmanian Institute of Agriculture
Partners: CSIRO Agriculture, Australia ( Dr Neil Huth)
Responsibility: Chief investigator
Funding body: Universities Australia, German Academic Exchange Service
Partners: The University of Göttingen Prof Siebert, The Leibniz Centre for Agricultural Landscape Research (ZALF) ( Dr Rezaei, Dr Ewert), The University of Bonn ( Dr Kamali), Germany
Responsibility: Partner
Funding body: Soil CRC High-Performance Soils
Partners: University of Southern Queensland (Assoc Prof Pembleton), Federation University ( Assoc Prof Peter Dahlhaus & Dr Robinson), NSW Department of Primary Industries
Funding body: National Research Council, Argentina
Funding body: Fulbright Commission, United States
Funding body: National Research Council, Argentina
Funding body: National Research Council, Argentina
Funding body: Argentinian Ministry of Education, Argentina
A python package to automate the extraction, processing and visualisation of climate data for crop modelling
Bestiapop (a spanish word that translates to pop beast), is a Python package which allows climate and agricultural data scientists to automatically download SILO’s (Scientific Information for Land Owners) or NASAPOWER gridded climate data and convert this data to files that can be ingested by Crop Modelling Software like APSIM or DSSAT. The package offers the possibility to select a range of grids (0.05° x 0.05° for SILO and 0.5° x 0.5° for NASAPOWER) and years producing various types of output files: CSV, MET (for APSIM), WTH (for DSSAT) and soon JSON (which will become part of Bestiapop’s API in the future). Users can also visualise data statistics (mean, standard deviation, CV, etc) spatially for any selected region in the world.
If you would like to use Bestiapop in Jupyter Notebook, you can see here! You can also try it live in Binder Project without the need to install any software in your computer (Yes! 😄 you do not need to know about Python, Anaconda, etc. to use this tool).
Making art with Python 🌾 💻 👨🌾
Windows, Linux (Ubuntu), Unix.
Python, .NET (medium, APSIM Classic, APSIM Next Generation), C#, Markdown (advanced), R Studio, Shell.
Python (pandas, statsmodels, sqlite3, json, glob, os, functools, lxml [handling of XML and HTML files], csv).
Python (numpy, scipy, scikit-learn [e.g. KMeans used for data clustering], pandas, matplotlib, math).
Python (seaborn, dask, xarray, cartopy, pyproj, shapefile, netCDF4, geopandas, rasterio, GDAL), remote sensing imagery in vegetation and soil moisture mapping (MODIS, Sentinel2, Sentinel1-SAR), ArcGIS, QGis, netCDF file format, and relational databases. pSIMS (gridded crop model simulations), nco operators (manipulates and analyzes data stored in netCDF in Linux), FluroSense (Regrow Ag cloud-based crop management and analytics platform that drives planting and growing decisions).
Google Cloud Platform, Docker, Singularity, Amazon Web Services, GitHub repositories, APIs, Swift, SQL tools (Database Client, SQL editor, Visual Query Builder, e.g. DBeaver).
Bestiapop (Python package to automate the extraction and processing of climate data for crop modelling, >3000 downloads), Atlassian (Jira and Confluence), Buddy (The DevOps Automation Platform), GitHub operations, Jupyter Notebook/Lab, Spyder, Anaconda, PostMan, Visual Studio.
APSIM Next Generation, APSIM Classic, DSSAT, SIMPLACE, MONICA, CropWat, pSIMS, DNDC
The original Parallel System for Integrating Impact Models and Sectors (pSIMS) was developed by Elliot et al. (2014) in Python 2, we updated pSIMS to pSIMSV2 which is able to run the APSIM sorghum module at a regional scale (US-wide) using netCDF input data (climate, soil and crop management). pSIMSV2 has the ability to run APSIM using a singularity image which avoid the need to install the soft dependencies manually.
Visualisation tools to map crop features and environmental variables across regions. Main functionalities include: import shp and tif files, use Basemap, edit legend and work with iso_3 codes, plot categories by country, edit legends in the map, inset charts in the map, read netCDF using xarray, explore and plot multidimensional files using xarray, create maps using xarray and dataframes, create 2D dataframe from xarray, create multi-dimensional xarray from 2D pandas dataframe, work with NASS API for crop statistics, etc…
Tools to use remote sensing data (MODIS, Sentinel2, NASA-POWER, etc) to validate crop models. These include the data curation and data analysis of remote sensing products before being used to validate models. Examples for linking APSIM Classic and Next Generation outputs with RS products are included.
This tool allows calculating the variance contribution of several factors on different crop model outputs. The theory developed by Monod 2006 was converted to a single Jupyter Notebook through Python. This tool produces a series of plots that allow the user to see the weight of each factor on the variance of crop yield.
During my free time, I enjoy writing my own webpage (the page you are reading right now!) in Markdown using the Hugo platform. Hugo is a popular static site generator written in the Go programming language. Hugo is jam-packed with features, but one of its main selling points is speed — Hugo takes mere seconds to generate a site with thousands of pages. By default, Hugo uses the Goldmark Markdown processor which is fully CommonMark-compliant.
Series of tools to develop and test APSIM using Python and C# 🤓
This code is able to retrieve APSIM Next Generation outputs and carried out a variance decomposition analysis to identify the main contributors to the variance in selected model outputs (e.g. crop yield). This code calculates the main (ME) and total effect (TE) of a series of factors on the variability of a selected variable (in this example crop biomass).
ME explains the share of the components to model output variability without interactions, i.e. if ME=1, the assessed factors explain the entire proportion of model output variability, but if ME<1, residuals exist which means additional factors are required to explain this variability. TE represents the interaction of a given factor with other factors, i.e. high TE values for a given factor denote high interactions of that factor with other factors, therefore, TE does not include residuals.
Series of tools that allow users to interact between two APSIM versions (Classic and Next Generation) and Python through Jupyter Notebook. Main functionalities include: read .out files and .db files, create new variables, clean model outputs, create time series plots and XY plots, etc.
Crop models are usually developed using a test set of data and simulations representing a range of environment, soil, management and genotype combinations. Previous studies demonstrated that errors in the configuration of test simulations and aggregation of observed data sets are common and can cause major problems for model development. However, the extent and effect of such errors are not usually considered as a source of model uncertainty. This code presents a systematic method for testing simulation configuration using extensive visualisation approaches. A crop model – potato (Solanum tuberosum L.) is described to demonstrate the main sources of uncertainty from simulation configuration and data collation. A test set of 426 experiments conducted from 1970 to 2019 in 19 countries were run using the APSIM Next Generation model. Plots were made comparing simulation configuration across the entire test set . This identified a surprising number of errors and inappropriate assumptions that had been made which were influencing model predictions. The approach presented here moved the bulk of the effort from fitting model processes to setting up broad simulation configuration testing and detailed interrogation to identify current gaps for further model development.
APSIM Classic was modified so that it could accurately predict growth and yield of switchgrass and Miscanthus; two plant species that were not represented in this large, multi-species model. Two existing APSIM sub-models (lucerne, sugarcane) were altered using knowledge of species-specific differences in growth, development and agronomic practices. Large databases for soils and weather were assembled for subsequently association with site-specific yield data of both species and successfully calibration and validation. These NEW APSIM sub-models predict the yield of both species across broad geographies from the East Coast to the Great Plains of the US.
Invited talks and guest lectures around the world 🌎
TropAg International Conference, Brisbane, Queensland, Australia. 31 October-2 November 2022
20th Australian Agronomy Conference, Toowoomba, Queensland, Australia. 18-22 September 2022
20th Australian Agronomy Conference, Toowoomba, Queensland, Australia. 18-22 September 2022
5th Annual Crops in silico Symposium & Hackathon, University of Illinois, USA (online)
APSIM monthly training YouTube videos, Brisbane, Australia (scheduled for November)
at regional scales APSIM Symposium 2020, Brisbane, Australia (cancelled due to COVID19)
at regional scales iCROPM2020 International Symposium, Montpellier, France
Workshop Lucerne, Lincoln, New Zealand.
Simplot/McCain workshop, Devonport, Australia
University of Entre Rios, Oro Verde, Argentina
Data Network Hobart Teas and Workshop, University of Tasmania, Australia
University of Gottingen, Germany
and modelling platforms Soil CRC Conference, Newcastle, Australia
system at regional levels? Guest Lecture, TIA seminars, University of Tasmania, Australia
Guest Lecture, University of Tasmania, Australia
PhD Dissertation defense. National University of Mar del Plata, Argentina
Crop Sequences Workshop. INTA General Villegas, Argentina
University Seminar 2015. National University of Entre Rios, Argentina
Oral and public defense of PhD Dissertation Project 2015. National University of Mar del Plata, Argentina
Postgraduate Seminar. Faculty of Veterinary Science. The University of Sydney, Australia
Research Seminar, Instituto Nacional de Tecnologia Agropecuaria, Paraná, Argentina
Forage Workshop, National Northwest University of Buenos Aires, Argentina
Research Conference for PhD students. Instituto Nacional de Tecnologia Agropecuaria, Balcarce, Argentina
Research Workshop for PhD students. Instituto Nacional de Tecnologia Agropecuaria, Rafaela, Argentina
Impact of cover crops with different defoliation levels on soil carbon
Comparative analysis of root production and root distribution in oats (Avena sativa) and tall fescue (Festuca arundinacea Schreb.)
Effects of previous crop, additives and pre-wilted in the nutritional quality of oat silage (Avena sativa)
Comparative analysis of water productivity between oats (Avena sativa) and tall fescue (Festuca arundinacea Schreb.)
Nutritional evaluation of silage maize-soybean intercropping