An open science approach for evaluating the Vegetation Optimality Model along the North-Australian Tropical Transect

An open science approach for evaluating the Vegetation Optimality Model along the North-Australian Tropical Transect Different state-of-the-art land surface models produce a large variety in model outcomes (Whitley et al., 2016), despite their extensive use of local observations such as rooting depths and leaf area index dynamics. Our understanding of differences between models and their underlying assumptions is greatly hampered by our inability to reproduce results generated by different scientists, as model setup routines and exact simulation conditions are not documented in a complete and homogenous fashion. Here we present an open science approach to reproducible modelling, using the Vegetation Optimality Model (VOM; Schymanski et al., 2009) as an example. The VOM optimizes vegetation properties and dynamics to maximise its Net Carbon Profit, i.e. the difference in total carbon taken up by photosysthesis minus the carbon costs. In this way, vegetation properties are not prescribed, but optimized for maximizing the Net Carbon Profit. The model was applied over six study sites along the North-Australian Tropical Transect. Open Science is facilitated by the command line tool Renku (renkulab.io), developed by the Swiss Data Science Center and based on existing technologies such as gitlab. Renku tracks all scientific steps from input data processing to the generation of the final figures in order to generate a tree containing the entire workflow, i.e. a knowledge graph. Renku enables easy reproduction of results including all preparatory steps and can be used to easily repeat given steps on updated or new data sets. Previously, it was already suggested to document the scientific process in more qualitative protocols (Ceola et al., 2015), and to make more more code openly available and formalize workflows (Hutton et al., 2016). With Renku, we take it one step further, also in line with Hut et al., (2017), and document the entire scientific process automatically by using data science, assuring openness, reproducibility and repeatability at the same time. References Ceola, S., Arheimer, B., Baratti, E., Blöschl, G., Capell, R., Castellarin, A., Freer, J., Han, D., Hrachowitz, M., Hundecha, Y., Hutton, C., Lindström, G., Montanari, A., Nijzink, R., Parajka, J., Toth, E., Viglione, A., Wagener, T., 2015. Virtual laboratories: new opportunities for collaborative water science. Hydrology and Earth System Sciences 19, 2101–2117. https://doi.org/10.5194/hess-19-2101-2015 Hut, R.W., Giesen, N.C. van de, Drost, N., 2017. Comment on “Most computational hydrology is not reproducible, so is it really science?” by Christopher Hutton et al.: Let hydrologists learn the latest computer science by working with Research Software Engineers (RSEs) and not reinvent the waterwheel ourselves. Water Resources Research 53, 4524–4526. https://doi.org/10.1002/2017WR020665 Hutton, C., Wagener, T., Freer, J., Han, D., Duffy, C., Arheimer, B., 2016. Most computational hydrology is not reproducible, so is it really science? Water Resources Research 52, 7548–7555. https://doi.org/10.1002/2016WR019285 Schymanski, S.J., Sivapalan, M., Roderick, M.L., Hutley, L.B., Beringer, J., 2009. An optimality‐based model of the dynamic feedbacks between natural vegetation and the water balance. Water Resources Research 45. https://doi.org/10.1029/2008WR006841 Whitley, R., Beringer, J., Hutley, L.B., Abramowitz, G., De Kauwe, M.G., Duursma, R., Evans, B., Haverd, V., Li, L., Ryu, Y., Smith, B., Wang, Y.-P., Williams, M., Yu, Q., 2016. A model inter-comparison study to examine limiting factors in modelling Australian tropical savannas. Biogeosciences 13, 3245–3265. https://doi.org/10.5194/bg-13-3245-2016Supported by the Luxembourg National Research Fund (FNR) ATTRACT programme (A16/SR/11254288