The "Sequestering CARbon through Forests, AgriCulture, and land usE (SCARFACE)" project is a research initiative funded by the University of Milano-Bicocca (UniMiB) on October 2024 for two years.

The project emerged from the blending and cross-departmental research experiences of complementary and interdisciplinary research groups: a group of statisticians and data scientists (with relevant background on environmental and economic statistics) from the Department of Economics, Management and Statistics (DEMS) and a group of environmental and atmospheric chemistry of the Department of Earth and Environmental Sciences (DISAT) at UniMiB. Along with researchers from UniMiB, the project involves researchers from the Council for Agricultural Research and Analysis of Agricultural Economics (CREA) and other partner institutions.

The SCARFACE project is a follow-up to the AgrImOnIA (UniBg-UniMiB-UniTo-LUH) and AgroGeoStat (UniMiB & CREA) research projects, to which researchers from UniMiB's statistics department have contributed and inspired since 2022.

The scientific coordinator of SCARFACE is Dr. Paolo Maranzano.

Through the use of advanced geostatistical learning and data science techniques, SCARFACE tackles several research questions concerning the carbon sequestration capacity of agricultural activities in the Po Valley (Northern Italy), a region characterized by intensive livestock farming and agricultural land use and with very low air quality standards.

1. The first research question aims to provide a comprehensive overview of the current land use, soil pedological properties, and agronomic practices contributing to carbon sequestration in the Po Valley. This enables the identification of subareas that exhibit the optimal synergy between soil use and agricultural practices, which can then be utilized to plan effective agricultural policies that contribute to carbon sequestration;

2. The second objective is to identify critical sub-areas exhibiting the most significant weaknesses and requiring targeted interventions to enhance agricultural practices, with the aim of providing guidance to national and local institutions. 

The ground floor of the project involves extracting data on the structure of agricultural holdings in the area (including their techno-productive practices related to carbon sequestration), livestock farming, soil use and composition, weather and climate and airborne pollutants related to agricultural activities, in particular ammonia and particulate matters. The dataset will merge information collected from large sets of georeferenced data with different spatio-temporal frequencies with uneven spatial coverages and different spatial resolutions. In such a context, data acquisition, cleaning, harmonization, and validation steps are fundamental.

Using the novel spatio-temporal dataset on agricultural practice, we will apply statistical learning and data science techniques to study the spatio-temporal evolution of agricultural land use and carbon sequestration practices in the Po Valley. Specifically, on the one hand, we aim to construct a spatio-temporal annual mapping of agricultural soil use, and agronomic technical practices for the region; on the other hand, we aim to investigate spatio-temporal trends and structural characteristics of the agro-industry, with particular reference to the carbon sequestration practices connected to soil properties, land use, and the most relevant agronomic practices. Eventually, by implementing spatio-temporal clustering techniques, we aim at identifying critical areas in the region of interest that require enhanced agricultural carbon sequestration efforts through targeted micro (i.e., farm-level) and macro (i.e., public) policies. By complement, vituous areas with good agronomic practices can be detected as well.