Your statistical analysis is a bit strange. For example, the statement: "The impact was significant at the 95 percent level" - I'm sure you mean the 0.05 level. Also, the Mann-Whitney U test is for when you only have two categories. You appear to have more than two categories of elevation.
Thank you for usefull answer. I use automatic SPSS 22 and considering row data, software select the Mann-Whitney and I comared two aspect in the tests. You can see the graphs.