The overall aim of this study was to analyse the contribution of spatial data and GIS in strategic forest management at a regional planning level considering topical issues: windthrow, carbon sequestration and climate change. It also recommends how the methodologies might be transferred to other countries such as Bangladesh. A macro-based geographical information system (GIS) is proposed as a suitable tool for modelling the interaction between wind and forest areas. Using the forest area of the Snowdonia National Park in North Wales, UK (affected by endemic windthrow) as the study area, land use, soil and windiness (Detailed Aspect Method of Scoring of the Forestry Commission of Great Britain) data were incorporated within a GIS. Shapes of forest patches were estimated through FRAGSTATS statistical software and placed in a GIS to combine the patch shape contributions with windiness. Four scenarios were generated from four combinations of patch shape, soil class and leaf presence in winter with the DAMS score to analyse the performance of the GIS in the integration of spatial data and techniques. A geographical information system is used to combine and analyse spatial data. Forest site factors: soil, elevation and exposure to wind were considered with yield class prescribed by Pyatt (1977) and local expert advice (Stevens, 2001) with respect to soil classes. Adjusted yield classes (A YC) for the conifer plantations of the study area (Snowdonia National Park) were selected and validated to estimate the organic carbon in the organic carbon model. A GIS-Spreadsheet organic carbon model, to estimate the organic carbon stock of the woodland, was presented considering tree, litter and soil to a depth of one metre. The Willis-Price tree carbon model (Price, 2001) was used to estimate the carbon stock of the woodland. Published data were used to estimate the litter organic carbon and local expert advice and published data for the soil organic carbon to a depth of one metre. Sites which would fix an organic carbon stock as great as the conifer (Picea sitchensis) if the sites were replaced by the broadleaf species oak (Quercus spp.) were selected. The results of the study revealed that yield class 02, 04 and 06 of oak enabled the fixing of as much organic carbon than conifers of A YC up to 06 (scenario 1 ), I 2 (scenario 2: AYC 08 tol2), 16 (scenario 3: A YC 14 to 16) respectively. The total organic carbon stock (tree, litter and soil) of the conifer plantations was estimated and the consequences of the landuse changes (three scenarios to replace the selected part of the conifer plantation) in accordance to the organic carbon model were also estimated. A macro-language-based GIS model is proposed as an effective tool to show how climate change scenarios of UKCIP 1998 will be adopted as a decisive factor in forest management, such as replacing exotic conifers with native broadleaved woodland. Suitable sites for potential native broadleaved woodland (PNBW) for the future (2080) considering climate change scenarios were sketched and located by overlaying Mulligan's (1999) maps with the sites of the scenarios presented in the organic carbon model. Total organic carbon stocks which would be sequestered with these sites were also estimated. The themes of the study, i.e. using GIS in week-by-week activities of forest management regional planning, may be transferable with the generic version of the models, particularly with macro language. A hypothetical study was presented to show the possible applications of these stated models in Bangladesh All the models were generated with the macro language of the IDRISI GIS which can cope with any change by updating the spatial datasets and running the model within GIS for the week-by-week scenarios. This accommodative ability of models enables the manager to have an easy understanding and shows the scenario to the policy maker within short notice. In conclusion, it was suggested that spatial data sets and GIS might contribute significantly in forest management, furthermore accommodating week-by-week activities into the database to derive regional forest ecosystem management decisions.