Place of publishing:
Subject and Keywords:
Greenhouse gases emission ; Emisja gazów cieplarnianych ; Statistical model ; Spatial inventory data ; Autoregressive model ; Spatial grid ; Model statystyczny ; Model autoregresyjny ; Siatka przestrzenna
In this paper we apply a linear regression with spatial random effect to model spatially distributed emission inventory data. The topic is related to the issue of disaggregation of national greenhouse gas emissions into fine spatial grid. Emission maps are typically produced from information on spatially explicit activities contributing to emissions, which are multiplied by emission factors. In our case study we have available N2O emission assessments for municipalities of southern Norway, as well as three kinds of covariate information for each region. Thus a regression model can be fitted to these data. We use conditionally autoregressive model to account for spatial correlation between municipalities. Estimation of parameters is based on the Bayes Theorem and the Gibbs sampler algorithm. The results suggest that one of initially considered covariates should be excluded from further analysis, and instead presence of another, spatially correlated factor(s) is suggested. Moreover, the model was capable to capture an outlier point source emission from nitric acid plant. The point of the contribution is that including spacial information helps in finding right explanatory variables. This way improved emission assessments can be provided based on future covariate data.
Language of abstract:
Projects co-financed by:
Operational Program Digital Poland, 2014-2020, Measure 2.3: Digital accessibility and usefulness of public sector information; funds from the European Regional Development Fund and national co-financing from the state budget.