Coastal forests and shrublands can be vulnerable to climate-associated disturbances such as the strong winds and storm surges associated with hurricanes. Because of its extension into the Atlantic, eastern North Carolina is especially vulnerable to such storms, having been hit by several hurricanes during the MODIS period of record (i.e., since 2000). In 2014, Hurricane Arthur struck causing relatively minor damage to forests compared to two earlier events, namely 2003's Hurricane Isabel and 2011's Hurricane Irene.
During the first week of June 2014, hail caused significant damage to corn and soybean fields in eastern Nebraska and western Iowa. By June 9th, ForWarn's all-lands product was beginning to show the losses that these storms inflicted across multiple counties despite later than average crop growth in May, and when ForWarn's next product was produced for June 17th, the losses from hail were clearly visible. These maps were made by comparing current NDVI conditions to the all-year mean value for a 24-day period maximum covering the years 2000-2013.
ForWarn's true color product is produced every 8 days at 1km resolution to provide a quality control for ForWarn's disturbance maps. Clouds and snow cover that persists for the entire 24-day MODIS collection window can create the appearance of vegetation disturbance, when it is not. In addition, these true color images show variation in vegetation greenness seasonally and among years.
Year to year variation in climate has been extreme in California's Sierra Nevada. In particular, low winter snowpack raises concerns about municipal water supplies as reservoirs are fed by spring meltwater. However, low winter snow cover has implications for the region's forests as well. Less snow (and warmer temperatures) often mean an earlier onset of spring and more extended and dryer summer than would otherwise occur.
In early September of 2013, unusually heavy rain resulted in catastrophic flooding across Colorado’s Front Range and downstream along the South Platte River. Rainfall exceeded 20 inches northwest of Denver, which is close to the average rainfall for that area during an entire year. Flash floods led to massive evacuations, a severe loss of public and private property, and loss of human life. Oil and gas lines and tanks were damaged, causing the release of thousands of gallons of pollutants into the South Platte River.
ForWarn's standard calculation of the current state of vegetation is based on the maximum NDVI value observed for each 24 day period. This standardized window is needed to minimize the chance that clouds or other atmospheric effects will cause misleading results in the change product. For many periods and in some portions of the US, a much shorter window is sometimes possible. ForWarn's new Early Detect product uses the most recent clear view to capture change as soon as possible, even though clouds can sometimes cause issues.
Hurricane Sandy, one of the most damaging hurricanes to strike the United States, tested the monitoring ability of the ForWarn system. Detecting any disturbance is difficult during the spring and fall because vegetation is actively changing due to normal seasonal dynamics. Normal variation in weather can advance or retard the progress of these transitions by a week or more making it hard to know if observed change is from an early fall or an event. Hurricane Sandy struck the North Atlantic at the end of October which was in the middle of seasonal leaf decline.
The spring of 2012 was exceptional across much of the eastern US because it occurred weeks earlier than normal. Yet some forests also experienced strong spring hail storms that left leaves in tatters and set spring growth back. Forwarn successfully captured this localized hail event that struck the watershed that the City of Asheville uses as their primary water supply in early May. At the time, some was observed from lower elevation vantage points, but the remoteness of this area affected made it difficult to systematically monitor from the ground.
ForWarn's ability to capture short and long-term change is shown by this animated image pair that provides two different temporal contexts for the same date in May of 2012. The "all year" baseline shows change since 2000. Note the patchy forest loss from strip mines in eastern Kentucky and West Virginia. Twin tornadoes also show up, and these occurred in March 2012. As expected, they also appear on the second image that shows change since May 2011, but most of the strip mine activity does not, as it took place prior to 2011.