ORIGINAL PAPER Assessing onset and length of greening period in six vegetation types in Oaxaca, Mexico, using NDVI-precipitation relationships L. Gómez-Mendoza & L. Galicia & M. L. Cuevas-Fernández & V. Magaña & G. Gómez & J. L. Palacio-Prieto Received: 16 February 2007 / Revised: 14 January 2008 / Accepted: 15 January 2008 / Published online: 26 February 2008 # ISB 2008 Abstract Variations in the normalized vegetation index (NDVI) for the state of Oaxaca, in southern Mexico, were analyzed in terms of precipitation anomalies for the period 1997–2003. Using 10-day averages in NDVI data, obtained from AVHRR satellite information, the response of six types of vegetation to intra-annual and inter-annual fluctuations in precipitation were examined. The onset and temporal evolu- tion of the greening period were studied in terms of precipitation variations through spectral analysis (coherence and phase). The results indicate that extremely dry periods, such as those observed in 1997 and 2001, resulted in low values of NDVI for much of Oaxaca, while good precipitation periods produced a rapid response (20–30 days of delay) from a stressed to a non-stressed condition in most vegetation types. One of these rapid changes occurred during the transition from dry to wet conditions during the summer of 1998. As in many parts of the tropics and subtropics, the NDVI reflects low frequency variations in precipitation on several spatial scales. Even after long dry periods (2001– 2002), the various regional vegetation types are capable of recovering when a good rainy season takes place, indicating that vegetation types such as the evergreen forests in the high parts of Oaxaca respond better to rainfall characteristics (timing, amount) than to temperature changes, as is the case in most mid-latitudes. This finding may be relevant to prepare climate change scenarios for forests, where increases in surface temperature and precipitation anomalies are expected. Keywords Climate variability . NDVI . Phenology . Tropical forest . Temperate forest Introduction Vegetation dynamics are related to climate (Justice et al. 1986; Jackson et al. 2001; Bonan 2002)—mainly temperature and precipitation variations (Churkina and Running 1998; Prasad et al. 2005). Phenology of vegetation is clearly coupled to annual and seasonal cycles of temperature and precipitation. Extreme climatic conditions on a regional scale may influence the length of the greening period and ecosystem capacity for CO 2 sequestration (Hill and Donald 2003). Consequently, a clear determination of the relationship between climate and vegetation health becomes an important issue when reforestation projects are considered to mitigate climate change. Remote sensing is a very useful tool with which to identify greening period and phenology on a large spatial scale (hundreds of kilometers) (Yu et al. 2003; Prasad et al. 2005). Several studies have found a close relationship between normalized vegetation index (NDVI) and vegeta- tion phenology (Kramer et al. 2000). This is due mainly to the fact that NDVI corresponds to a fraction of the photosynthetically active radiation absorbed by vegetation canopies (Tucker et al. 1985; Jobbágy et al. 2002). NDVI has also been considered for monitoring spatial gradients of structure and function of various types of vegetation on Int J Biometeorol (2008) 52:511–520 DOI 10.1007/s00484-008-0147-6 L. Gómez-Mendoza : L. Galicia (*) : M. L. Cuevas-Fernández : G. Gómez : J. L. Palacio-Prieto Instituto de Geografía, Universidad Nacional Autónoma de México, Circuito Exterior s/n, Ciudad Universitaria, C.P. 04510 Coyoacán, D.F., Mexico e-mail: lgalicia@igg.unam.mx L. Gómez-Mendoza : V. Magaña Centro de Ciencias de la Atmósfera, Universidad Nacional Autónoma de México, Circuito Exterior s/n, Ciudad Universitaria, C. P. 04510 Coyoacán, D.F., Mexico