Previous Work at LTER Sites

Within the LTER network, there has been ongoing interest in the ENSO-related phenomena. A workshop, held in 1993, concentrated on the effect of El Niños and La Niñas at LTER sites (Greenland 1994a). Since the LTER network is spread across the North American and Antarctic continents (figure 1.1), it is natural that ENSO climatic signals should be stronger and more marked for some LTER sites than others. The line of LTER sites, from New Mexico through Colorado to the Pacific Northwest and into Alaska (JRN, SEV, NWT, SGS, AND, BNZ), follows the inverse influence of El Niño-related above-average precipitation in the Southwest to below-average precipitation in the Northwest. However, the intensity of the ecological response differs. The effect of low streamflows on the ecosystems in the Northwest is less marked than the large hydrologic and ecological impact documented by workers at the Seviletta (SEV) LTER site in New Mexico (Molles and Dahm 1990; Dahm and Molles 1992). Wetter than usual winters during El Niños have large effects on the aquatic and terrestrial ecosystems at the Sevilleta site. Plants, invertebrates, rodents, and rabbits all react to the increase in autumn and spring moisture associated with El Niño. Additionally, Dahm and Moore (1994) showed a series of dry La Niña episodes in the late 1940s and mid-1950s led to significant dieback of pinyon pine (Pinus edulis) and juniper (Juniperus monosperma) at the site. Ecosystem responses at SEV are treated in detail in chapter 15. At the NWT site, Woodhouse (1994) showed that wet springs occur the year after an El Niño and dry springs follow a La Niña. An ecological response is visible in tree ring chronologies, which indicated a more marked response for La Niñas than for El Niño. Tree growth response to SOI values varies over time at this site.

The first El Niño workshop (Greenland 1994a) demonstrated that the ENSO signal could be geographically subtle in its effects on ecosystems. For example, sophisticated analyses were made for the North Temperate Lakes (NTL) LTER sites in Wisconsin. Robertson et al. (1994) showed that although during El Niño events ice breakup dates were earlier and spring air temperatures were warmer than long-term averages across the whole state of Wisconsin, the effect was more extreme for southern lakes than northern lakes. Robertson and coworkers attribute the difference to the average breakup dates for the southern lakes being in late March, directly following a period when air temperatures are strongly related to El Niño events. In contrast, average breakup dates for northern lakes are in mid- to late April after a period when air temperatures are not significantly related to El Niño events. Later work by individual site investigators has identified an ENSO signal at other sites. An investigation by Schaefer (see chapter 8) suggests the Luquillo (LUQ) site in Puerto Rico has increased rainfall in May and decreased rainfall in October in El Niño years, whereas in La Niña years this pattern is reversed, although, in both cases, the signal is weak. The Andrews Forest (AND) site in the Pacific Northwest often suffers from below-average precipitation in El Niño years, whereas many severe floods and above-average precipitation years are associated with La Niñas (Greenland 1994b; see also chapter 19).

Until 1999 the effect of ENSO phenomena on LTER sites had not been investigated with a standardized methodology across all sites—a situation that made intersite comparison difficult. This was remedied by a systematic analysis on a common set of climatic data applicable to 17 of the LTER sites and designed to identify the relative strength of the El Niño and La Niña signal across the LTER network (Greenland 1999).

0 0

Post a comment