I created DEM with Sentinel 1-A SLC. Then, I validate this elevation result with previous research and it is very different. In the previous research, it’s about 0.2 - 5 meters, but when i created this with SNAP, it’s about 20 - 40 meters.
Would you help me please tell me why it’s very different? The sentinel data used is S1A_IW_SLC__1SDV_20200929T110635_20200929T110704_034572_040654_1D01 and S1A_IW_SLC__1SDV_20201011T110635_20201011T110705_034747_040C74_225F
The main reason for the large differences is probably the high degree of decorrelation over the forest areas. It is nearly impossible to get proper data over vegetation with pairs of 12 (or even 6) days temporal baseline.
I’d say the interferogram as you showed it is not usable. You can try to find a better image pair, but over forest, nearly any SAR data will fail to give you reliable phase information.