The world is awash with vague and worthless generalizations and politically correct 'risk-vulnerability matrices'. But if you want worthwhile estimates of the future then there is no short cut. There is no alternative to actually calculating the specific results yourself. Firstly, one doesn't calculate the climate change impact by country. One calculates by grid block, as per the defined cells in atmospheric-ocean coupled global climatic model. You will need to assess the output of at least a dozen, preferably about 16, different models for maximum, minimum and mean monthly temperatures. Repeat this for your chosen carbon emissions scenarios. We are currently on track for the disastrous RCP 8.5 scenario, but if the world comes to its' collective sense then you could probably use RCPs 4.5 and 6.0. Forget about RCP 2.8, which is now just an exercise in wishful thinking. There are several climate-change sources, such as the World Bank climate change portal, on the internet from where the model outputs are published - albeit on models which are already out of date (they are being refined all the time). Having calculated the temperatures for each scenario, carry out your preferred statistical assessment of the wide range of model outputs for various time-slices for as far into the future as you feel confident about the model's capacity. This gives you the likely temperature changes for the given emissions scenario, and time horizon. Repeat all of this for as many grid cells as you deem representative of your country of interest.
I recommend that you do not use the published model outputs for rainfall, or indeed for any water-related parameter (humidity, soil moisture deficits, evapotranspiration, open-water evaporation, etc). GCMs were never designed for this, and are inaccurate to wildly misleading. For these parameters, use evidence-based trends from existing instrumental time-series of rainfall (corrected for systematic error and discontinuities), using rigorous statistical testing of confidence limits. Use the corrected rainfall, humidity, temperature, wind, and hours of bright sunshine to plug into the Penmann-Monteith equation for evapotranspiration (The solar radiation can be calculated by altitude and coordinates on the same internet site as the ET0 calculators). Use the ET0 data and FAO crop factors to estimate changes in crop water requirements for the time frame of your study. Use all of the above, and specialist subject area searches to assess the likelihood of such factors as land degradation, desertification, crop failure frequency, drought frequency, disease vector mobility, stress on stock, stress on humans, and other inland issues.
Sea level rise is something else which takes in coastal aquifer contamination, coastal flooding, ecological changes, fisheries breeding, coral bleaching and reef degradation, and synergies of temperature, dissolved oxygen, pollution and other human impacts, frequency of oceanic oscillations, and changes in coastal dynamics.
You can add sophistication to your climatic analysis by modifying your instrumental trend projections to a more reasonable non-linear form, such as an error-function transition to an assumed climatic plateau in the distant (post 2100) future. For the sea level rise there is no sign of any quasi- equilibrium being regained, at least before 2400 (if then), so a second-order sea level rise projection is about the best we can do at present, with the caveat that uncalibrated feedback processes are likely.
In short, what you are asking is at least a man-year of intensive work by someone who knows what he/she is doing - make that two man-years if you include the rising sea level. So it isn't surprising that you can't find the detailed studies that you seek. Some regions of some countries have gone through this exercise, generally with a team of specialists. In all such cases that I know of, the results are in official ministry reports, and are not easily found through public access. Further, even at the end of this lengthy process the results are still subject to many uncertainties and caveats, including those mentioned earlier by 'userLTK'.