Cross scale spatial and temporal indicators for measuring the effects of landscape heterogeneity on pollination service

Spatial heterogeneity as well as landscape services’ provision are a function of spatio-temporal scales, therefore, pattern–process relationships must be assessed at the multiple scales. In this context, this research aims at: (1) analyzing at the regional scale how pollination service can be affected by landscape heterogeneity, using two landscape indicators useful to quantify the multiscale landscape composition and landscape configuration simultaneously; and (2) assessing the effect that the infection of Xylella fastidiosa has exerted on the pollination services. The multi-scale spatial assessment has been focused on two land-covers: forests and olive groves that can act as source of pollination services. The multi-temporal analysis, based on the annual NDVI, has been used to assess the functionality of olive groves before and at the beginning of the infection of Xylella fastidiosa, and currently. The results have shown that in 2012 the most representative cluster (C1) (73.6% for forests and 63% for olive groves) is in the lower left part of the multiscale metric space, meaning that both land-covers show a fragmented spatial configuration at small spatial scales and tend to be aggregated at large scales. The multi-temporal analysis has allowed to show the evident change in the landscape functioning in the provinces interested by the infection of Xylella fastidiosa (Lecce, Brindisi and Taranto) from 2013 to 2021, highlighting that the stability of the landscape has resulted completely changed for the loss of permanent land-covers (olive groves). In this study the spatio-temporal analyses have helped in giving a more complete indication in the assessment of landscape services where different factors can play a crucial role. The analysis of spatial patterns along a continuum of scale has been implemented by the analysis of multi-temporal dynamics to consider the effect of Xylella fastidiosa infection on pollination. The temporal behavior of NDVI has resulted completely changed in the provinces interested by this infection, meaning that recovering policies need to be undertaken to regenerate the landscape. When studying landscape services, some considerations must be considered in choosing the suitable spatial and temporal scale for its assessment. One of the recent drivers of change, represented by the infection of Xylella fastidiosa, and the planning of landscape functionality recovery interest higher spatial scales and will affect the scale at which landscape services, included pollination, are delivered