pith. machine review for the scientific record. sign in

arxiv: 2604.25342 · v1 · submitted 2026-04-28 · 📊 stat.AP

Recognition: unknown

On the use of satellite information to estimate agricultural carbon footprint in a small area framework

Felicetta Carillo, Paolo Maranzano, Riccardo Borgoni, Riccardo Pajno, Timo Schmid

Pith reviewed 2026-05-07 14:12 UTC · model grok-4.3

classification 📊 stat.AP
keywords small area estimationsatellite dataagricultural carbon footprintammonia emissionsgeostatistical upscalingenvironmental statisticsPo Valley
0
0 comments X

The pith

Satellite ammonia emission data improves accuracy and stability of small-area agricultural carbon footprint estimates.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper builds a model-based framework for estimating agricultural carbon footprints at the scale of agrarian subregions in the Po Valley by combining survey and census data with satellite-derived information. Satellite-based ammonia emissions serve as auxiliary covariates that boost precision and spatial consistency while addressing mismatches between gridded satellite observations and administrative boundaries. A geostatistical upscaling step aligns the data sources, and a parametric bootstrap carries forward uncertainty from the covariate construction into the final estimates. This matters because reliable fine-scale indicators are needed for agricultural areas under climate and economic pressures where traditional data sources are limited or heterogeneous.

Core claim

The authors establish that incorporating satellite-derived ammonia emission data into a small-area estimation model substantially improves the accuracy and stability of carbon footprint estimates for agriculturally homogeneous municipalities while reducing dependence on large auxiliary datasets; the spatial misalignment between gridded satellite data and administrative units is handled through a geostatistical upscaling procedure whose uncertainty is propagated via parametric bootstrap.

What carries the argument

Geostatistical upscaling procedure that aligns gridded satellite ammonia emission data to agrarian subregion boundaries, combined with parametric bootstrap to propagate uncertainty from covariate construction.

If this is right

  • Subregional carbon footprint estimates become more precise and spatially coherent without requiring extensive heterogeneous auxiliary surveys.
  • Policy-relevant environmental indicators can be produced at finer scales in data-constrained agricultural zones.
  • The framework reduces overall reliance on traditional large-scale datasets for similar environmental statistics.
  • Uncertainty from auxiliary data construction is explicitly accounted for in the published estimates.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same satellite-augmented approach could extend to estimating other livestock-related emissions or environmental burdens in comparable intensive farming regions.
  • Official statistical agencies might adopt Earth-observation covariates to lower costs of maintaining fine-scale sustainability indicators.
  • If the proxy relationship holds, the method offers a scalable template for integrating remote-sensing data into model-based small-area statistics beyond agriculture.

Load-bearing premise

Satellite ammonia emission data acts as an accurate, unbiased proxy for agricultural carbon emissions and the upscaling procedure aligns gridded data to boundaries without material bias or unaccounted error.

What would settle it

Independent farm-level or ground-sensor measurements of carbon emissions in the Po Valley that show no accuracy gain from the satellite-enhanced model compared with the version using only survey and census data would falsify the improvement claim.

Figures

Figures reproduced from arXiv: 2604.25342 by Felicetta Carillo, Paolo Maranzano, Riccardo Borgoni, Riccardo Pajno, Timo Schmid.

Figure 1
Figure 1. Figure 1: The Po Valley and agrarian sub-regions (i.e., at the farm level) concerning the Italian agricultural sector, with a particular focus on the economic and productive structure, as well as the environmental impact of farms, via a representative sample of the national system. Consequently, the FADN information will serve as the foundation for the survey statistics employed in SAE models. Other sources furnish … view at source ↗
Figure 2
Figure 2. Figure 2: Farm geolocalization by type of framing and economic size view at source ↗
Figure 3
Figure 3. Figure 3: Carbon footprint distribution conditioned to type of farming and economic size. Regarding the size, farms are classified based on whether their standard output was below 50, 000 euros (small farms) or above 50, 000 euros (large farms). We observed nearly equal numbers of small and large farms (1,517 versus 1,274). Small farms are more prevalent in Emilia-Romagna, while larger farms are concentrated in Veneto view at source ↗
Figure 4
Figure 4. Figure 4: Altitude and soil slope spatial distribution satellite-based analysis of agriculture and land use offers large opportunities to explore new methods to efficiently make use of this information. For Europe, the Copernicus Atmosphere Monitoring Service (CAMS) [CAMS, 2023], implemented by the European Center for Medium-Range Weather Forecasting (ECMWF) [Sabater, 2019], is one of the most recent global database… view at source ↗
Figure 5
Figure 5. Figure 5: Gridded map of the ammonia NH3 emissions in 2020 in the Po Valley from the CAMS database. Pixels are regular squares with 0.1 ◦ × 0.1 ◦ length. We use the subscript i below to indicate a given area, hence ni and Ni denote the sample size and the population size in area i, respectively. Let yik denote the variable of interest, i.e. the CF of farm k in area i. The Horvitz-Thompson (HT) estimator of the popul… view at source ↗
Figure 6
Figure 6. Figure 6: Top-left panel: (a) Estimated semi-variogram for the NH3 emissions on the 0.1 ◦ × 0.1 ◦ regular grid. Top-right panel: (b) Average 5-fold cross-validation RMSE by neighborhood size. Bottom-left panel: (c) CAMS data for NH3 emissions in the 222 considered ASRs with a 10km buffer. Bottom-right panel: (d) Estimated average NH3 emissions per ASR by block kriging. 5 Double parametric bootstrap procedures for un… view at source ↗
Figure 7
Figure 7. Figure 7: Left panel: (a) Horvitz–Thompson direct estimates of the agricultural carbon footprint per ASR (KTons CO2 eq.). Central panel: (b) empirical standard deviations of the Horvitz–Thompson direct estimates of the agricultural carbon footprint per ASR. Left panel: (c) histogram of the Horvitz–Thompson direct estimates of the agricultural carbon footprint per ASR. For these reasons, in the next section we introd… view at source ↗
Figure 8
Figure 8. Figure 8: It is also noteworthy that the largest values of the RMSE are observed to correspond to the view at source ↗
Figure 8
Figure 8. Figure 8: Spatial Fay–Herriot model with SAR random effects. Left panels: (a-d) ASR-level EBLUPs of agricultural CF and corresponding RMSEs from Model 1 (KTons CO2 eq.). Central panels: (b-e) ASR-level EBLUPs of agricultural CF and corresponding RMSEs from Model 2 (KTons CO2 eq.). Right panels: (c-f) Scatterplots comparing EBLUPs and RMSEs obtained from Model 1 and Model 2 (KTons CO2 eq.). 17/30 view at source ↗
Figure 9
Figure 9. Figure 9: Spatial Fay-Herriot model with SAR random effects. Left panel: (a) Estimated RMSE for the direct HT estimator. Central panel: (b) Estimated RMSE for the EBLUPs in Model 1. Right panel: (c) Estimated RMSE for the EBLUP in Model 2. Colors are on the same scale to provide are fair comparison. improvements we obtain moving from a direct estimation to a model-based estimation; second, relatively to the SFH mode… view at source ↗
Figure 10
Figure 10. Figure 10: Spatial Fay-Herriot model with SAR random effects. Left panels: (a-c) Q-Q normal plots for the model residuals. Right panels: (b-d) Q-Q normal plots for estimated SAR random effects. Region (NUTS-2) Direct est. Model 2 95% bootstrap percentile CI Emilia Romagna 3,179 2,542 (1,041; 7,997) Lombardy 6,754 4,730 (2,551; 24,079) Piedmont 3,654 3,726 (1,109; 9,601) Veneto 1,818 2,466 (1,965; 17,025) view at source ↗
Figure 11
Figure 11. Figure 11: Fay–Herriot model with independent random effects. Left panels: (a-d) ASR-level EBLUPs of agricultural CF and corresponding RMSEs from Model 1 (KTons CO2 eq.). Central panels:(b-e) ASR-level EBLUPs of agricultural CF and corresponding RMSEs from Model 2 (KTons CO2 eq.). Right panels: (c-f) Scatterplots comparing EBLUPs and RMSEs obtained from Model 1 and Model 2 (KTons CO2 eq.) view at source ↗
Figure 12
Figure 12. Figure 12: Fay-Herriot model with independent random effects. Comparison of the variability of different approaches without spatial autocorrelation view at source ↗
Figure 13
Figure 13. Figure 13: Fay-Herriot model with independent random effects. Left panels: (a-c) Q-Q normal plots for the model residuals. Right panels: (b-d) Q-Q normal plots for estimated SAR random effects 27/30 view at source ↗
Figure 14
Figure 14. Figure 14: Left panel: (a) Average number of simulated grid points per ASR. Right panel: (b) variogram envelope. where Yˆ (u) is the kriging prediction based on the real data, and Yˆ ∗ (u) is the kriging prediction based on the simulated values at the data locations. Yc(u) is thus a conditional simulation of the random field at location u, i.e., a random field consistent with both the observed data and the specified… view at source ↗
Figure 15
Figure 15. Figure 15: Left panel: (a) Average simulated ASR-level emission predictions plotted against the actual values. Right panel: (b) Envelope of the spatial distributions of average simulated ASR-level predictions view at source ↗
Figure 16
Figure 16. Figure 16: Left panel: (a) Spatial predictions of total CF across bootstrap replicates on the log scale; red points represent the ASR-level bootstrap averages plotted against the EBLUPs obtained from the actual data. Right panel: (b) Envelope of the spatial distributions of simulated CF totals per ASR on the log scale. 2.5th and 97.5th percentiles of the 1,000 bootstrap replicates. The plot shows that, on average, t… view at source ↗
Figure 17
Figure 17. Figure 17: MSE differences between the spatial predictions of total CF per ASR obtained from the model using the satellite covariate and: the HT direct estimates (in green), the model-based spatial predictions obtained using administrative covariates (in blue), and the spatial predictions from the bootstrap replicates (yellow envelope). All values are reported on the log scale. 30/30 view at source ↗
read the original abstract

The agricultural sector is undergoing rapid change due to climate pressures, demographic shifts, and uneven economic development, increasing the demand for reliable environmental indicators at fine spatial scales. However, limited data availability often constrains subregional analyses. This study develops a model-based framework for producing reliable small-area estimates for assessing the agricultural carbon footprint in the Po Valley (Northern Italy), a region characterized by intensive livestock farming and high environmental pressure. We integrate survey, census, and satellite-derived emission data into a unified framework and produce estimates at the level of Agrarian Subregions, defined as agriculturally homogeneous municipalities by the Italian National Institute of Statistics. Satellite-based ammonia emission data are incorporated as auxiliary covariates to improve precision and spatial coherence. A key methodological contribution is the treatment of spatial misalignment between gridded satellite data and administrative boundaries. This issue is addressed through a geostatistical upscaling procedure combined with a parametric bootstrap that propagates uncertainty from the covariate construction stage to the final small-area estimates. The results show that satellite-derived information substantially improves the accuracy and stability of carbon footprint estimates while reducing reliance on large, heterogeneous auxiliary datasets, illustrating the potential of Earth observation data in model-based environmental statistics.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. This paper develops a model-based small-area estimation approach for agricultural carbon footprint in the Po Valley region of Italy. It integrates traditional survey and census data with satellite-derived ammonia emission data as auxiliary covariates, using a geostatistical upscaling procedure to address spatial misalignment between gridded satellite data and administrative boundaries, along with a parametric bootstrap to propagate uncertainty. The authors claim that this integration leads to more accurate and stable estimates at the Agrarian Subregion level while decreasing reliance on large heterogeneous auxiliary datasets.

Significance. Should the quantitative improvements be demonstrated, the work would be significant for the field of environmental statistics and small-area estimation. It highlights the potential of satellite data to enhance precision in policy-relevant indicators for agriculture under climate pressures, offering a framework that could reduce the need for extensive ground-based data collection. The combination of geostatistical methods with small-area models and uncertainty propagation is a notable methodological contribution.

major comments (3)
  1. [Abstract] The assertion that satellite-derived information substantially improves accuracy and stability lacks any accompanying quantitative evidence, such as comparisons of mean squared error, bias, or cross-validation results between models with and without satellite covariates. This is central to validating the paper's main contribution.
  2. [Methods section on covariate construction] The treatment of ammonia (NH3) satellite data as a proxy for agricultural carbon footprint (encompassing CO2, CH4, N2O) requires justification, as the link is indirect through sources like livestock and soil processes. Without reported correlations or diagnostics at the subregion scale, the improvement claim risks being driven by the specific choice of covariate rather than a robust relationship.
  3. [Parametric bootstrap description] While the parametric bootstrap is intended to propagate uncertainty from the geostatistical upscaling of gridded data to administrative units, it is unclear from the description whether this procedure fully incorporates potential misspecification error in the NH3 proxy or only sampling variability in the upscaling step.
minor comments (2)
  1. The abstract would benefit from specifying the small-area model employed (e.g., whether it is a linear mixed model or Fay-Herriot type) to provide context for the integration.
  2. Ensure that all acronyms (e.g., GHG, NH3) are defined at first use in the main text.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed comments, which help strengthen the manuscript. We address each major comment below and describe the planned revisions.

read point-by-point responses

  1. Referee: [Abstract] The assertion that satellite-derived information substantially improves accuracy and stability lacks any accompanying quantitative evidence, such as comparisons of mean squared error, bias, or cross-validation results between models with and without satellite covariates. This is central to validating the paper's main contribution.

    Authors: We agree that the abstract would be strengthened by including quantitative evidence. The results section already contains direct comparisons (including MSE reductions and cross-validation metrics) between the model with and without the satellite covariates. We will revise the abstract to report these specific improvements explicitly. revision: yes

  2. Referee: [Methods section on covariate construction] The treatment of ammonia (NH3) satellite data as a proxy for agricultural carbon footprint (encompassing CO2, CH4, N2O) requires justification, as the link is indirect through sources like livestock and soil processes. Without reported correlations or diagnostics at the subregion scale, the improvement claim risks being driven by the specific choice of covariate rather than a robust relationship.

    Authors: The choice of NH3 is justified by its established role as an indicator of intensive agricultural activity (livestock and fertilizer use) that drives both ammonia emissions and the greenhouse gas components of the carbon footprint in the study region. We will expand the methods section with supporting literature references and add reported correlations plus basic diagnostics between the NH3 covariate and the carbon footprint indicators at the agrarian subregion level. revision: yes

  3. Referee: [Parametric bootstrap description] While the parametric bootstrap is intended to propagate uncertainty from the geostatistical upscaling of gridded data to administrative units, it is unclear from the description whether this procedure fully incorporates potential misspecification error in the NH3 proxy or only sampling variability in the upscaling step.

    Authors: The parametric bootstrap as currently implemented propagates only the sampling variability associated with the geostatistical upscaling of the gridded NH3 data to administrative boundaries. It does not incorporate potential misspecification error in the NH3 proxy relationship itself. We will revise the methods description to clarify this scope and add a brief discussion of this as a methodological limitation. revision: yes

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The paper describes a standard model-based small-area estimation approach that combines survey/census data with satellite-derived ammonia emissions as auxiliary covariates, using geostatistical upscaling to handle spatial misalignment and a parametric bootstrap to propagate uncertainty from covariate construction. The reported accuracy gains are presented as empirical outcomes from comparing models with and without the satellite information, which constitutes an independent evaluation against external benchmarks rather than a reduction to fitted parameters or self-referential quantities by construction. No load-bearing self-citations, uniqueness theorems, or ansatzes are invoked in the provided text, and the central claims rest on the application of established geostatistical and SAE techniques to independent data sources without definitional equivalence between inputs and outputs.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The framework rests on standard assumptions of model-based small-area estimation and geostatistics. Key domain assumptions include the correlation between ammonia emissions and carbon footprint plus the validity of upscaling for misalignment. No new entities are introduced; free parameters are the usual model coefficients typical of SAE but unspecified here.

free parameters (1)
  • small-area model parameters
    Coefficients and variance components in the model-based estimator are fitted to the integrated data sources, as is standard in SAE but not detailed in the abstract.
axioms (2)
  • domain assumption Satellite ammonia emission data correlates sufficiently with agricultural carbon emissions to serve as an effective auxiliary covariate
    Invoked when incorporating satellite data to improve precision and spatial coherence.
  • domain assumption Geostatistical upscaling can align gridded satellite data to irregular administrative boundaries without substantial bias
    Central to the treatment of spatial misalignment.

pith-pipeline@v0.9.0 · 5515 in / 1470 out tokens · 73049 ms · 2026-05-07T14:12:41.171153+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

63 extracted references · 2 canonical work pages

  1. [1]

    Ali, A., Ahmad, M., Nawaz, M., and Sattar, F. (2025). Identifying and prioritizing spatial data required for effective agriculture policymaking: A comprehensive analysis using 20/30 analytical hierarchy process.Data Intelligence, 7(1):185–220. Auteri et al.,

    2025

  2. [2]

    Auteri, D., Attardo, C., Berzi, M., Dorati, C., Albinola, F., Baggio, L., Bucciarelli, G., Bussolari, I., and Dijkstra, L. (2024). The annual regional database of the european commission (ardeco): Methodological note. Report, JRC Working Papers on Territorial Modelling and Analysis. Baldoni et al.,

    2024

  3. [3]

    Baldoni, E., Coderoni, S., and Esposti, R. (2017). The productivity and environment nexus with farm-level data. the case of carbon footprint in lombardy fadn farms. Bio-based and Applied Economics, 6(2):119–137. Baldoni et al.,

    2017

  4. [4]

    Baldoni, E., Coderoni, S., and Esposti, R. (2018). The complex farm-level relationship between environmental performance and productivity: The case of carbon footprint of lombardy farms.Environmental Science & Policy, 89:73–82. Banca Dati RICA,

    2018

  5. [5]

    Battagliese et al.,

    Banca Dati RICA (2024).Sistema Documentale RICA. Battagliese et al.,

    2024

  6. [6]

    Battagliese, D., Pollice, A., Intini, M., and Bergantino, A. S. (2026). Downscaling direct estimators in small area models.Statistical Methods & Applications, pages 1–17. Belmonte et al.,

    2026

  7. [7]

    Belmonte, A., Riefolo, C., Buttafuoco, G., and Castrignan` o, A. (2025). An approach for spatial statistical modelling remote sensing data of land cover by fusing data of different types.Remote Sensing, 17(1):123. Bishop et al.,

    2025

  8. [8]

    M., Fienberg, S

    Bishop, Y. M., Fienberg, S. E., and Holland, P. W. (2007).Discrete multivariate analysis: Theory and practice. Springer Science & Business Media. Cameletti et al.,

    2007

  9. [9]

    Cameletti, M., G´ omez-Rubio, V., and Blangiardo, M. (2019). Bayesian modelling for spatially misaligned health and air pollution data through the inla-spde approach. Spatial Statistics, 31:100353. CAMS,

    2019

  10. [10]

    Carillo et al.,

    CAMS (2023).CAMS2-61: Global and European emission inventories - Documentation of CAMS emission inventory products. Carillo et al.,

    2023

  11. [11]

    C., and Salvatore, R

    Carillo, F., Maranzano, P., Marcis, L., Pagliarella, M. C., and Salvatore, R. (2024). The spatio-temporal fay-herriot model using the state-space method: an application to italian lombard agrarian sub-regions.Book of Short Papers - 2nd Italian Conference on Economic Statistics (ICES

    2024

  12. [12]

    Chi, G., Fang, H., Chatterjee, S., and Blumenstock, J. E. (2022). Microestimates of wealth for all low-and middle-income countries.Proceedings of the National Academy of Sciences, 119(3):e2113658119. Chiles and Delfiner,

    2022

  13. [13]

    and Delfiner, P

    Chiles, J.-P. and Delfiner, P. (2012).Geostatistics: Modeling Spatial Uncertainty. Wiley, Hoboken, NJ, 2nd edition. Coderoni et al.,

    2012

  14. [14]

    Coderoni, S., Bonati, G., Longhitano, D., Papaleo, A., and Vanino, S. (2013). Impronta carbonica aziende agricole italiane. Coderoni and Esposti,

    2013

  15. [15]

    and Esposti, R

    Coderoni, S. and Esposti, R. (2018). Cap payments and agricultural ghg emissions in italy. a farm-level assessment.Science of The Total Environment, 627:427–437. Coderoni et al.,

    2018

  16. [16]

    Coderoni, S., Esposti, R., and Baldoni, E. (2016). The productivity and environment nexus through farm-level data. the case of carbon footprint applied to lombardy fadn farms. InSeventh International Conference on Agricultural Statistics (ICAS VII), Roma, Italy, pages 26–28. Coderoni and Pagliacci,

    2016

  17. [17]

    and Pagliacci, F

    Coderoni, S. and Pagliacci, F. (2023). The impact of climate change on land productivity. a micro-level assessment for italian farms.Agricultural Systems, 205:103565. 21/30 Coderoni and Vanino,

    2023

  18. [18]

    and Vanino, S

    Coderoni, S. and Vanino, S. (2022). The farm-by-farm relationship among carbon productivity and economic performance of agriculture.Science of The Total Environment, 819:153103. Colombo et al.,

    2022

  19. [19]

    L., Pillon, S., and Lanzani, G

    Colombo, L., Marongiu, A., Malvestiti, G., Fossati, G., Angelino, E., Lazzarini, M., Gurrieri, G. L., Pillon, S., and Lanzani, G. G. (2023). Assessing the impacts and feasibility of emissions reduction scenarios in the po valley.Frontiers in Environmental Science,

    2023

  20. [20]

    and Coderoni, S

    Cortignani, R. and Coderoni, S. (2022). The impacts of environmental and climate targets on agriculture: Policy options in italy.Journal of Policy Modeling, 44(6):1095–1112. Dowd and Pardo-Ig´ uzquiza,

    2022

  21. [21]

    Dowd, P. A. and Pardo-Ig´ uzquiza, E. (2024). The many forms of co-kriging: a diversity of multivariate spatial estimators.Mathematical Geosciences, 56(2):387–413. Edochie et al.,

    2024

  22. [22]

    L., Ouedraogo, A., Sanoh, A., and Savadogo, A

    Edochie, I., Newhouse, D., Tzavidis, N., Schmid, T., Foster, E., Hernandez, A. L., Ouedraogo, A., Sanoh, A., and Savadogo, A. (2025). Small area estimation of poverty in four west african countries by integrating survey and geospatial data.Journal of Official Statistics, 41(1):96–124. EUROSTAT, 2024a. EUROSTAT (2024a).NUTS - Nomenclature of territorial un...

    2025

  23. [23]

    Fass` o et al., 2023.Fass` o, A., Rodeschini, J., Moro, A

    EUROSTAT (2025).Animal populations by NUTS 2 regions. Fass` o et al., 2023.Fass` o, A., Rodeschini, J., Moro, A. F., Shaboviq, Q., Maranzano, P., Cameletti, M., Finazzi, F., Golini, N., Ignaccolo, R., and Otto, P. (2023). Agrimonia: a dataset on livestock, meteorology and air quality in the lombardy region, italy.Scientific Data, 10(1):143. Fay and Herriot,

    2025

  24. [24]

    Fay, R. E. and Herriot, R. A. (1979). Estimates of income for small places: An application of james-stein procedures to census data.Journal of the American Statistical Association, 74(366a):269–277. Gardini et al.,

    1979

  25. [25]

    Gardini, A., De Nicol` o, S., and Fabrizi, E. (2025). A mixture-of-experts model to deal with the rural/urban dichotomy in small area estimation.Journal of the Royal Statistical Society Series C: Applied Statistics, 74(5):1255–1278. Georgakis et al.,

    2025

  26. [26]

    E., and Stamatellos, G

    Georgakis, A., Papageorgiou, V. E., and Stamatellos, G. (2023). Bivariate fay-herriot model for enhanced small area estimation of growing stock volume. In2023 International Conference on Applied Mathematics & Computer Science (ICAMCS), pages 162–168. Georgakis et al.,

    2023

  27. [27]

    E., and Stamatellos, G

    Georgakis, A., Papageorgiou, V. E., and Stamatellos, G. (2024). A new approach to small area estimation: improving forest management unit estimates with advanced preprocessing in a multivariate fay–herriot model.Forestry: An International Journal of Forest Research. Gotway and Young,

    2024

  28. [28]

    Gotway, C. A. and Young, L. J. (2002). Combining incompatible spatial data.Journal of the American Statistical Association, 97(458):632–648. Haining,

    2002

  29. [29]

    (1990).Spatial data analysis in the social and environmental sciences

    Haining, R. (1990).Spatial data analysis in the social and environmental sciences. Cambridge , MA: Cambridge University. Harmening et al.,

    1990

  30. [30]

    Harmening, S., Kreutzmann, A.-K., Schmidt, S., Salvati, N., and Schmid, T. (2023). A framework for producing small area estimates based on area-level models in r.The R Journal, 15:316–341. https://doi.org/10.32614/RJ-2023-039. 22/30 Hassouna et al.,

    work page doi:10.32614/rj-2023-039 2023

  31. [31]

    J., Beltran, I., Amon, B., Alfaro, M

    Hassouna, M., van der Weerden, T. J., Beltran, I., Amon, B., Alfaro, M. A., Anestis, V., Cinar, G., Dragoni, F., Hutchings, N. J., Leytem, A., Maeda, K., Maragou, A., Misselbrook, T., Noble, A., Rych la, A., Salazar, F., and Simon, P. (2023). Dataman: A global database of methane, nitrous oxide, and ammonia emission factors for livestock housing and outdo...

    2023

  32. [32]

    Roma, Italy

    ISTAT (2006).Regione Agraria. Roma, Italy. ISTAT,

    2006

  33. [33]

    Kercuku et al.,

    ISTAT (2020).Tavole regionali censimento dell’agricoltura. Kercuku et al.,

    2020

  34. [34]

    Kercuku, A., Curci, F., Lanzani, A., and Zanfi, F. (2023). Italia di mezzo: The emerging marginality of intermediate territories between metropolises and inner areas.REGION, 10(1):89–112. Kreutzmann et al.,

    2023

  35. [35]

    Kreutzmann, A.-K., Pannier, S., Rojas-Perilla, N., Schmid, T., Templ, M., and Tzavidis, N. (2019). The R package emdi for estimating and mapping regionally disaggregated indicators.Journal of Statistical Software, 91(7):1–33. Maranzano et al.,

    2019

  36. [36]

    S., Otto, P., and Carillo, F

    Maranzano, P., McConville, K. S., Otto, P., and Carillo, F. (2023). A geostatistical investigation of the ammonia-livestock relationship in the po valley, italy.Book of the Short Papers SIS 2023, pages 200–205. Marongiu et al.,

    2023

  37. [37]

    Marongiu, A., Angelino, E., Moretti, M., Malvestiti, G., and Fossati, G. (2022). Atmospheric emission sources in the po-basin from the life-ip prepair project.Open Journal of Air Pollution, 11(3):70–83. Marongiu et al.,

    2022

  38. [38]

    G., Distefano, G

    Marongiu, A., Collalto, A. G., Distefano, G. G., and Angelino, E. (2024). Application of machine learning to estimate ammonia atmospheric emissions and concentrations.Air, 2(1):38–60. Morales et al.,

    2024

  39. [39]

    (2021).A Course on Small Area Estimation and Mixed Models

    Morales, D., Esteban, M., P´ erez, A., and Hobza, T. (2021).A Course on Small Area Estimation and Mixed Models. Springer. Newhouse et al.,

    2021

  40. [40]

    Newhouse, D., Ramakrishnan, A., Swartz, T., Merfeld, J., and Lahiri, P. (2025). Small area estimation of monetary poverty in mexico using satellite imagery and machine learning.Oxford Bulletin of Economics and Statistics. Otto et al.,

    2025

  41. [41]

    Otto, P., Fusta Moro, A., Rodeschini, J., Shaboviq, Q., Ignaccolo, R., Golini, N., Cameletti, M., Maranzano, P., Finazzi, F., and Fass` o, A. (2024). Spatiotemporal modelling of pm2.5 concentrations in lombardy (italy): a comparative study.Environmental and Ecological Statistics,

    2024

  42. [42]

    Pebesma, E. J. (2004). Multivariable geostatistics in s: the gstat package.Computers & Geosciences, 30(7):683–691. Permatasari and Ubaidillah,

    2004

  43. [43]

    and Ubaidillah, A

    Permatasari, N. and Ubaidillah, A. (2025). Small area estimation of poverty using remote sensing data.Statistical Journal of the IAOS, 41(1):180–190. Pratesi and Salvati,

    2025

  44. [44]

    and Salvati, N

    Pratesi, M. and Salvati, N. (2008). Small area estimation: the eblup estimator based on spatially correlated random area effects.Statistical methods and applications, 17:113–141. Raffaelli et al.,

    2008

  45. [45]

    Raffaelli, K., Deserti, M., Stortini, M., Amorati, R., Vasconi, M., and Giovannini, G. (2020). Improving air quality in the po valley, italy: Some results by the life-ip-prepair project.Atmosphere, 11(4):429. Rao and Molina,

    2020

  46. [46]

    Rao, J. N. K. and Molina, I. (2015).Small Area Estimation. Wiley, New York. 23/30 Rash et al.,

    2015

  47. [47]

    Rash, A. J. H., Khodakarami, L., Muhedin, D. A., Hamakareem, M. I., and Ali, H. F. H. (2024). Spatial modeling of geotechnical soil parameters: Integrating ground-based data, rs technique, spatial statistics and gwr model.Journal of Engineering Research, 12(1):75–85. Rawat et al.,

    2024

  48. [48]

    Rawat, P., Kumar, M., and Airon, A. (2025). Hierarchical bayes small area estimation for rice yield using remote sensing data.Model Assisted Statistics and Applications, 20(1):52–62. Rodeschini et al.,

    2025

  49. [49]

    Rodeschini, J., Fass` o, A., Finazzi, F., and Fusta Moro, A. (2024). Scenario analysis of livestock-related pm2.5 pollution based on a new heteroskedastic spatiotemporal model. Socio-Economic Planning Sciences, page 102053. Sabater,

    2024

  50. [50]

    (2019).ERA5-Land hourly data from 1950 to present

    Sabater, M. (2019).ERA5-Land hourly data from 1950 to present. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). (Accessed on 15-05-2025),. Shawky et al.,

    2019

  51. [51]

    A., Saad, A., and Elayouty, A

    Shawky, S. A., Saad, A., and Elayouty, A. (2025). A spatial and spatiotemporal statistical downscaling model for combining spatially misaligned maximum temperature data using r-inla.Environmetrics, 36(8):e70053. Singh et al.,

    2025

  52. [52]

    B., Shukla, G

    Singh, B. B., Shukla, G. K., and Kundu, D. (2005). Spatio-temporal models in small area estimation.Survey Methodology, 31(2):183. Slud and Maiti,

    2005

  53. [53]

    Slud, E. V. and Maiti, T. (2006). Mean-squared error estimation in transformed fay–herriot models.Journal of the Royal Statistical Society Series B: Statistical Methodology, 68(2):239–257. Tzavidis,

    2006

  54. [54]

    Tzavidis, N. (2025). Small area estimation in the era of machine learning and alternative data sources: Opportunities, challenges, and outlook.Journal of Official Statistics, 41(3):921–929. Villejo et al.,

    2025

  55. [55]

    J., Illian, J

    Villejo, S. J., Illian, J. B., and Swallow, B. (2023). Data fusion in a two-stage spatio-temporal model using the inla-spde approach.Spatial Statistics, 54:100744. Wang et al.,

    2023

  56. [56]

    Wang, H., Zhao, Z., Winiwarter, W., Bai, Z., Wang, X., Fan, X., Zhu, Z., Hu, C., and Ma, L. (2021). Strategies to reduce ammonia emissions from livestock and their cost-benefit analysis: A case study of sheyang county.Environmental Pollution, 290:118045. Wang et al.,

    2021

  57. [57]

    Wang, J., Wang, Y., Li, G., and Qi, Z. (2024). Integration of remote sensing and machine learning for precision agriculture: a comprehensive perspective on applications.Agronomy, 14(9):1975. Western et al.,

    2024

  58. [58]

    M., Sha, Z., Rigby, M., Ganesan, A

    Western, L. M., Sha, Z., Rigby, M., Ganesan, A. L., Manning, A. J., Stanley, K. M., O’Doherty, S. J., Young, D., and Rougier, J. (2020). Bayesian spatio-temporal inference of trace gas emissions using an integrated nested laplacian approximation and gaussian markov random fields.Geoscientific Model Development, 13(4):2095–2107. Whittle,

    2020

  59. [59]

    Whittle, P. (1954). ON STATIONARY PROCESSES IN THE PLANE.Biometrika, 41(3-4):434–449. Wieczorek,

    1954

  60. [60]

    Wieczorek, J. (2023). Design-based conformal prediction.arXiv preprint arXiv:2303.01422. Wyer et al.,

    work page arXiv 2023

  61. [61]

    E., Kelleghan, D

    Wyer, K. E., Kelleghan, D. B., Blanes-Vidal, V., Schauberger, G., and Curran, T. P. (2022). Ammonia emissions from agriculture and their contribution to fine particulate matter: A review of implications for human health.Journal of Environmental Management, 323:116285. Ybarra and Lohr,

    2022

  62. [62]

    Ybarra, L. M. and Lohr, S. L. (2008). Small area estimation when auxiliary information is measured with error.Biometrika, 95(4):919–931. 24/30 Zhao et al.,

    2008

  63. [63]

    Zhao, Q., Yu, L., Du, Z., Peng, D., Hao, P., Zhang, Y., and Gong, P. (2022). An overview of the applications of earth observation satellite data: impacts and future trends.Remote Sensing, 14(8):1863. 25/30 Appendix A: Comparison between Fay-Herriot model with independent random effects and with spatially-correlated random effects To assess the efficacy of...

    2022