Prognosis of forest fire danger | Meža ugunsbīstamības prognozes

English. Therefore aim of the study was to identify such and index and analyse its changes as a results of climate changes. Moisture of litter, duff and humus (or turf ) layers of soil was regularly measured from 2012 to 2014 (April–October) in altogether 69 forest stands with various age and dominant tree species (Scots pine, Norway spruce, silver birch, common aspen). Stands were representing different fire danger classes according to classification currently used in Latvia. In three sites meteorological stations were placed in such a way, that largest distance from sampled stand to the station would not exceed 2.5 km. Hourly data on temperature and precipitation were collected and used for calculations of following fire danger indexes: Nesterov, Modified Nesterov, Canadian Fire Weather (and its components (Fine fuel moisture code, Duff moisture code, Drought code). Calculation of climate change impact were based on ENSEMBLES models, downscaled for Latvia, A1B climate change scenario. Moisture of organic matter was strongly affected by group of forest types, representing a fire danger class. Litter layer had lowest moisture (on average 17.7%) and largest its fluctuations. Moisture of duff layer was always greater than that of litter layer (on average 30.4%) and smaller than that of humus layer (37.4%). Most of fire danger indexes correlated well with moisture of litter layer (r = –0.43....–0.50, except for one, where r = –0.25). Correlation with moisture of duff layer (depth 2–5 cm) was lower (r = –0.27...–0.39), and with moisture of humus layer (depth 6–10 cm) – lowest (r = –0.16...–0.41). Canadian Fire Weather index (FWI) and its drought code had strongest correlation with moisture of measured soil layers and therefore were recommended for characterisation of forest fire danger in Latvia.

Latvian. Pētījuma mērķis bija identificēt šādu indeksu un raksturot tā izmaiņas klimata izmaiņu ietekmē. Augsnes virsējā slāņa mitruma mērījumi triju gadu posmā (2012.–2014.) veikti trijās pētījuma vietās Latvijas rietumu un centrālajā daļā, kopumā 69 mežaudzēs (valdošās sugas: priede, egle, bērzs, apse), vienlaikus ievācot arī meteoroloģiskos datus. Novērtēta to korelācija ar četriem ugunsbīstamības indeksiem: Ņesterova, modificēto Ņesterova un Kanādas uguns laika apstākļu indeksu (un tā komponentēm). Lielākā daļa indeksu labi raksturoja nobiru mitrumu (r = –0,43....–0,50, izņemot vienu, kur r = –0,25), vājāk – daļēji sadalījušos nobiru slāņa (2–5 cm dziļumā) mitrumu (r = –0,27...–0,39), pavisam vāji – „trūdzemes” slāņa (6–10 cm dziļumā) mitrumu (r = –0,16...–0,41). Kanādas uguns laika apstākļu indeksam un tā sastāvā ietilpstošajam sausuma kodam bija augstākā korelācija ar visu vērtēto slāņu mitrumu, tādēļ to rekomendējams izmantot ugunsbīstamības raksturošanai Latvijā. Izmantojot Kanādas uguns laika apstākļu indeksu, aprēķināts, ka lielākajā daļā Latvijas teritorijas, laika periodā no 1980. līdz 2009. gadam, vidēji gadā bijušas 5 līdz 20 dienas ar sevišķi augstu ugunsbīstamību, taču nākotnē (2031.–2060. g.g.) šādu dienu skaits ievērojami pieaugs, sasniedzot 30...50. Tādēļ ir būtiski veikt investīcijas, lai ilgtermiņā uzlabotu meža uguns apsardzības efektivitāti.